2 .

1 INTRODU~ION
Metallis and NonmetalIic Maker& Physical and Me~hanieal Properties Stress a d Strain
2.2 STATIC &%OPERTIES

Tensile Test Compressron Tests Elardnea Tmting

lPll 2.1 INTRODUCTION

Manufacturing has been accurately defined as the activities that are performed in the conversion of "stuff" into "things." Successful products begin with appropriate materiStructure Properties als. You wouldn't build an airplane out of lead or an automobile out of concrete-you need to start with the right stuff. But "stuff" rarely comes in the right shape, size, and quantity for the desired use. Parts and components must be produced by subjecting engineering materials to one or more processes (often a series of operations) that alter lheir shape, their properties, or both. Much of a manufacturing education relates to an Performance understanding 01(1)the structure of materials, (2) the properties of materials, (3) the Processing processing of materials, and (4) the perfor~nanceof materials, and the interrelations FIGURE 2-1 between these four factors, as illustrated in Figure 2-1. The manufacturing relationships This chapter will begin to address the properties of engineering materials. among structure, properties, Chapters 3 and 4 will discuss the subject of "structure" and begin to provide the whys processing, and performance. behind various properties. Chapter 5 introduces the possibility of modifying structure to produce desired properties. Most engineering materials do not have a single set of properties but rather offer a range or spectrum of possibilities.Takingadvantage of this range, we might want to intentionally make a material weak and ductile for easy shaping (forming loads are low and tool life is extended) and then, once the shape has been produced, make thc material strong for enhanced performance in use. When selecting a material for a product or application, it is important to ensure that its properties will be adequate for the anticipated operating conditions. The various requirements of each part or component must first be estimated or determined. These requirements typically include mechanical characteristics (strength, rigidity, resistance to fracture, the ability to withstand vibrations or impacts) and physical characteristics (weight, electrical properties, appearance) as well as features relating to the service environment (ability to operate under extremes of temperature or to resist corrosion). Candidate materials must possess the desired properties within their range of possibilities. To help evaluate the properties of engineering materials, a variety of standard tests have been developed, and data from these tests have been tabulated and made readily available. Proper use of this data often requires sound engineering judgment. It is important to consider which of the evaluated properties are significant, under what conditions the test values were determined, and what restrictions or limitations should be placed on their use. Only by being familiar with the various test procedures, their

1><1

DeGarmo's

DeGarmo's

7 2 5
184

09
' / ? [ ) d l

e3

:>

J .5/9

'RING

Auburn University-Emeritus

1 - T. Black

Ronald A. Kohser
University o f MissourbRolla

John Wiley & Sons, Inc

Cover photos: The sketchbooks of Lconardo da Vinci (1452-1519) contain two drawings that are of parlicular interest to the authors of this tcxt Oneis a n u d e sketch of an underwater device, ur submarine, with the elongated sausage shape chwractesistic of many later day succcsscs.Thc other, reproduced on the cover of this edition. is amflying-machine," that bears an uncanny rescmblancr, 10 a modern-day helicopter. Unlikc many of Leonarda's creations, he apparently madc no attempt to lurther refine the-conccpts. since there was never a sub~equrnlskctch of oithsr.

We lrave no way of knowing, but he may have realized that the Was this man really such a ~enius'! construction materials of his day were totally inadeuuale [or either lask. One would not want to build a suhmal-ine or helicopter from wood. stone or leather. Today's submarines are constructedfrom curn,sion-resistant. hieh-strength metals that arc also selected for their ability to be fabricated by " welding. Aerospace materials must offcr high-strength and light-weight,along with fatigue- and Iraclurcresistancc.ne rotor arms of modrrn helicopters are now being made fro111fiber-reinforced composite materials.'lhc components of the engine and drive assembly havc some of the most demanding rcquiicmenls <,fm,~dern en@ncering.

the success of those matrriills and processes. Like Lronardo, however, today's dcsipners continue to push thc limits-lighter, stranger. more corrosion resistant,closcr to net-shape, more economical. New matcrials will certainly be developed, and new proccsscs will expand our capabilities. It is the goal of this tevlt<,prcsent the capabilities and limitations of current technology with a look toward future ldvilnces that hopefully will enable today's dreams to become tomorrow's reality.

EXFCUTIVE EDITOR EXECUTIVE PUBLISHER

MARKETING MANAGER SENIOR PRODUCTION EDITOR CREArlVE DIRECTOR DESIGNER MEDIA EDITORS PRODUCTION MANAGEMENT SERVICES COVER PHOTO CREDITS TOP PHOTO BOTTOM PHOTO

Jwcph Hayton Don Fnwley D a n ~ rSayre l Sandra Kim Phyllis Cerys Sandra Dumas Harry Nolan Hope Mlller Steve ChascyiLauren Saplrc Ingrao Aasoc~ates!Suzannelngrao

0Kran Collect~oniGetty Imagea

0 Michael MeifardiThe Image BanklGetty lrnagcs

Tnis book was set in l0il2Tirneslen by Prepare and printcd and bound by Courier/Wexiford The cover was printed by Courier/We.srford.

Thn hook is prfnted nn acid f r c c p ~ p e r

Copyrlghl Q ZOUX John W~ley 81 Sons. Inc All rlgits reserved No part of this publicati<mmay be reproduced,rtored in a rctricval system or transmitted in any form or by any means, electronic,mechmical, photocopying,rccording; scanning, or otherwise, except as permitted under Secdon 107 or 108 of the I976 United States Copyright Act, without either the prior writtrn permission of the Publisher, or authorization through paymenl of the appropriate pei-copy fee to the Copyright Clearance Ccntcr, Inc. 222 Rosewood Drive, Danvets, MA 01923, (978)750-84130, fax (978)750-4470. Requests to the Publisher for permission should be addrcsscd L o the Pelmissions Depanmenr,John Wiley & Sons, Inc., 111 River Street, Haboken, NJ 07030, (201)748-6011,fnx (201)748-6008, E-Mail: [email protected] order hooks or for customer service, please call 1-800-CALL WlLEY (225-3945).

ISBN 978-(1470 05512 0

Prlntcd In the Un~ted States of Amencd 1098765432

In the world of manufacturing, slgniticant changes and trends are having a profound Impact on our everyday tiveswhether we l i k e ~or t not, we all live m a technological society, a world of manufactured goods Every day we come in contact with hundreds of manufactured items, from the bedroom to the htchen, to the workplace, we use appliances, phones, cars, trains, and planes,TVs,VGRs,DVD's, furniture, cloCning, and so on. These goods are man~lfactured~n tactorles all over rhe world using manuiacturlng processes. What are the trends in the manufacturing world, and how do they impnct manufacturing processes?

I TRENDS IN MANUEACTURING
Fhst,manufacturing has become a global activity w ~ t h US. companles sending work to other countries ( C W , Taiwan, Meaco) to take advantage of low-cost labor, whlle many foreign companies are buildlng plants In the Wnitcd States, to be nearer their marketplace,The automobile manufacturers and then suppliers use just about every process described in this book and some that w e do not describe, often because they are closely held secrets. Second, many manufacturing companies are redesigning their factortes (thelr manufacturing systems) becoming lean producers. and learning how to make goods better (higher quality),cheaper, f a s t e ~ in aflexlble way (t.e., they are more responsive to the customers).Almost every plant that you can vislt these days is doing something to make itself leaner. Many of them have adopted some verslon of the Toyota Production System. More impottantly, these manufacturing factones are deslgned with the internal customer (the workforce) in mind, so things lrke ergonom~cs and safety are key deslgn requirements. So while this book IS all about materials and processes for making the products, the deslgn of the factory cannot be ignored when it comes to makmg the external customer happy \nth the product and the Internal customer satisfied with the employer. Third, the number and variety of products and the materials from which they are made continue to proliferate, while production quantities have become smaller. Existing processes must bc madifled to be more flexible, and new proccses must be developed. Fourth, consumers want better quality and reliability,so the methods, processes, and people responsible for the quality must be continually improved.7he trend toward zero defects and continuous improvement requites continuous improvement$ of the manufacturing system. Finally the new producldevelopment effort L o reduce the trme-to-marketfor new ar system mde perspective.s, products is conttnuing. Many companles are taEmg wholist~c includmg concurrent engmeering efforts to bring product deslgn and manufacturing clloser to the customer. There are two key aspects here. Erst, products are designed to be easier to manufacture and assemble (caneddesagnfor manufutture/msembly).Second, the manuf&cturlngsystem design is flexible (able to accept new products), so the company can be ~ompetitive in the global marketplace. Basically, manufacturing is a value-nddmg aciivlty, where the conversion of materxals rnto produds adds value to the original material. Thus, the objective o fa company engaged in manufacturing 1s to add value and do so in the most efficient manner, using the least amount of time,material, money,space, and labor.To min~mize waste and maximize efficiency, the processes and operations need to be properly selected and arranged to permit smooth and controlled flow of material through the factory and provide for product variety.Meeting these goals requlres a weU-designed and effictent manufacturing system.

I I

vi

Preface

PURPOSE OF THE BOOK

The purpose of this book is to give design andmanufacturlng engineers and techn~cians basic intormation on matenal$manufacturing processes and systems.The materials section focuses on properties and behavior.Thus,aspects of smelting and reflning (or other material producl~on processes) are presented only as they affect manufacturing and manufactured products. In terms of the processes used to manufacture Items (converting materials into products), this text seeks to promde a descr~ptive introduction to a wide variety of opt~ons, emphas~zing how each process works and its relative advantages and limitations. Our goal 1s to present this material In a way that can be understood by individuals seeing it for the very first time.Tlus 1s not a graduate text where the objective is to thoroughly understand and optimize manufacturmg processes. Mathematical models and analybcal equations are used only when they enhance the bas~c understanding of the material So, while the text is an introductory text, we do attempt to incorporate new and emerging technologies lihe a welding process that is being adapted to alter and Improve material properties and performance without creating a joint. The book also serves to introduce the language of mmufacrurtng Just as there is a big d~fference between a gun hand and a hand gun, there is a big lfference between an engtne lathe and a lathe engine Everyday English words (words like cltmb, Dluom, wllowance, chuck, coin, head, and ironmg) have entirely different meanings on the factory floor, a place where m~sunderstandings can be very costly. P ~ t y the engineer who has to go on the plant floor not knowing an engine lathe from a milling machine or what a press brake can do.This engineer quickly loses all credibility with the people who make the products (and pay the engmeers' salaries) However, the modern manufacturing engineer must be able to deal with real workplace problem solvmg techniques like Taguchi methods and six sigma and developing manufacturing cells to make product famhes.This requires redeqign of all the elements of the manufacturrng systems-the machlne tools and manufacturing processes, the workholding devices, the materlal handling equipment, and the retraining of the people who work in the system.

In 1957,E. Paul DeGarmo was a mechanical engineering professor at the University of Cahfornia, Berkley when he wrote the first edit~on of Materials and Processes m Manufacturing. The book quickly became the emulated standard for introductory texts in manufacturing. Second, third, and fourth editions followed in 1962,1969, and 1974. DeGarmo had begun teaching at Berkeley in 1937, after earning hls M S. in mechan~cal engineering from Callforma Institute of Technology. He worked as a factory control engineer at FirestoneTlre and Rubber Company while attending Caltech. DeGarmo was a founder of the Department of Industrial Englneering (now Industrial Engineering and Operat~ons Research) and served as its chair from 1956-1960. He was also assistant dean of the College of Engineering for three years while continuing hls teaching responsibd~aes. H e retired from active teach~ng in 1971 and he continued his research, wnting, and consulting for many years. In 1977, after the pnbllcat~on of the fourth edition of Matenu15 and Pruces~es in Manufacturing, he received a letter from Ron Kohser, then an assistant professor at Missouri-Rolla who had many suggestions regarding the materials chapters. DeGarmo asked Kohser to rewrite those chapters for the fifth ed~t~o which n , Ron did. After the fifth editlon DeGarmo decided he was really going to retire and after a national search, recru~ted J T. Black, then a Professor at Ohio State, to co-author the book. For the sixth edition, seventh edit~on, eighth andnmth editions (pubhshed in 1984,1988, and 1997,respectively, by Macmillan, PrenticeHall and 1999 and 2003 by John Wdey & Sons), Ron Koliser and J T. Black have shared theresponsiblhty for the text.The chapters on engmeenng materials, casting,forming,powder metallurgy, joimng and non-destructive testlng have been written or revised by Ron Kohser. J T. Black has assumed the responsibihty for the introduction and chapters on material removal, metrology, surface finishing, quality control and manufacturing systems design.

DeGarmo died in 2000, three weeks short of his 93'"lrthday. His wife Mary died in 1995;he is surv~ved by his sons, David and Richard, and many grandchildren.For this 10th edition, we honor our mentor E. Paul DeGarmo with a change in the title to include his name. We are forever indebted to Paul for selecting us to carry on the tradition of his book on its' fiftieth anniversary!

I5 0 ANNIVERSARY ~ ~ EDITION!
Any long-term user of this book will note a significant change in its title-from Materials and Processes i n Manufacturing by DeGarmo, Black, and Kohser to DeCarmo's Materials and Processes in Manufacturing by Black and Kohser.Pau1DeGarmo initiated this text in 1957 and nurtured it through a number of editions. Even after his retirement, through his death in 2000, Paul maintained an active interest and involvement. In recognition, the 9th edition, published in 2003, carried his name as a posthumous coauthor. For 50 years, this text has been known by many as simply "DeGarmo," andit is this identity that we wish to continue by moving his name to become a preface to the former title. In 1957 Dr. DeGarmo observed that engineering education had begun to place more emphasis on the underlying sciences at the expense of hands on experience. Most of his students were coming to college with little familiarity with materials, machine tools, and manufacturing methods that their predecessors had acquired through the old "shop"c1asses. If these engineers and technicians were to successfully convert their ideas into reality, they needed a foundation in materials and processes, with emphasis on their opportunities and their limitations. He sought to provide a text that could be used in either a one- or two-semester course designed to meet these objectives.The materials sections were written with an emphasis on use and application. Processes and machine tools were described in terms of what they could do, how they do it, and their relative advantages and limitations, including economic considerations. Recognizing that many students would be encountering the material for the first time, clear description was accompanied by numerous visual illustrations. Paul's efforts were well received, and the hook quickly became the standard text in many schools and curricula. As materials and processes evolved, advances were incorporated into subsequent editions. Computer usage, quality control, and automation were added to the text, along with other topics, so that it continued to provide state-ofthe-art instruction in both materials and processes. As competing books entered the market, one was forced to note that their subject material and organization tended to mimic the DeGarrno texi. Professors Black and Kohser are proud to continuePaul's legacy. It is fitting that _this 10th edition will be published in 2007, 50 years following the initial efforts of Professor DeGamo. It is further fitting that his name continue to appear on this 50th anniversary edition and any subsequent editions.

E. Paul DeGarmo wanted a book that explained to engineers how the things they designed are made. DeGarnzo's Materials and Processes in Munufacturing is still written providing a broad, basic introduction to the fundamentals of manufacturing. The book begins with a survey of engineering materials, the "stuff" that manufacturing begins with, and seeks to provide the basic information that can be used to match the properties of a material to the service requirements of a comp0nent.A variety of engineering materials are presented, along with their properties and means of modifying them. The materials section can be usedin curricula that lack preparatory courses in metallurgy, materials science, or strength of materials, or where the student has not yet been exposed to those topics. In addition, various chapters in this section can be used as supplements to a basic materials course, providing additional information on topics such as heat treatment, plastics, composites, and material selection. Following the materials chapters,measurement and nondestructivetesting are introduced with a manufacturingperspective. Then chapters on casting, forming,powder metallurgy, material removal, and joining are all developed as families of manufacturing processes

viii

Preface Each section beg~ns w~th a presentation of the fundamentals on which those processes are m followed by a discussion of the varlous process alternatives, which can be based T h ~ s selected to operate indiv~dually or be coinblned Into an integrated system. In the last two chapters there issome ~ndepth material on surface engineenng and quahty control. Engneers need to know how to determine process capab~lity and if they get involved in six sigma projects, to know what slgrna really measuresThere is also introductory material on surface mtegnty, slnce so many processes produce the finished surface and residual stresses in the components.

New chapter on measurement,inspection and testing New chapter on electronic processey New examples of basic calculations in machining chapters NC chapter reorganized with more examples Reclassification of metal deformation processes into bulk and sheet Expanded coverage of new and emerging technology,such as friction-stirwelding Expanded coverage of polymers; ceramic materials and composites, and the processes that are unique to those materials. Throughout the book,case studies have been designed to makestudents awareof the great importance of properly coordlnatmg deaign,mater~al select~on, and manufactur~ng to produce a satisfactory and reliable product. The text is intended for use by engineenng (mech,m~cal, manufacturing, and mdnstrial) and engrneering technology students, in both two- and four-year undergraduate degree programs. In addition, the book is also used by engineers and technologists m other disc~plines concerned w ~ t h design and manufactur~ng (such as aerospace and electronics). Factory personnel wrll find this book to be a valuable reference that concisely presents the varlous production alternahves and the advantage5 and limitat~ons of each.Addttiona1or more in-depth ~nformationon specific materials or processes can be found 1x1 the various references posted on the Internet along with chapters on rap~d prototypmng, automation and enterprise systems.

For instructors adopt~ng the text for use in their course, an rnvtructor snlutrons manual is available through the book wchs~te. www w~ley.com/college/degarmo. Also ava~lable on the website is a set of powerpoint lecture slrdes created by Philip Appel at Gonzaga University. Three add~tional chapters, as identified in the table of contents, are available on the front cover provides inforbook website. The registration card attached on the ins~de mat~on on how to access and download t h ~matenal s If the registration card 1s m~ssing, access can be purchased d~rectly on the website www.wiley.com/college/degarmo,by chcking on "student companion site" and then an the links to the chapter titles.

The authors wish to acknowledge the multitude of assistance, information, and organizations, lllustrat~ons that have been provrded by avanety of mdustries,profess~onal and trade associations. The text has become known for the large number of clear and helpful photos and illustrahons that have been grac~ously provided by a varlety of w~thout safety guards, so sources. In some cases, equipment is photographed or dep~cted as to show important details, and personnel are not wearing certam items of safety apparel that would be worn during normal operation Over the many editions, there have been hundreds of reviewers, faculty, and students who have made suggestions and corrections t o the text. We continue to be grate ful for the tlme and ~nterest that they have put into t h ~book. s In this e d ~ t ~ we o n benefited from the comments of the following reviewers J Don Book,P~ttsburg State Un~vers~ty;

Jan Brink, Midwestern State University; Rene A.Chapelle, University of Houston; Joe Chow, Florida International University; Kurt Colvin, California Polytechnic State University, Pomona; Snbi Dinda, Oakland University; Roman Dubrovsky, New Jersey Institute of Technology; Richard B. Griffin, Texas A&M University-Main; Rodney G. Handy. Purdue University; T. Kesavadas, State University of New York, Buffalo; John Lee, San Jose State University; H. Joel Lenoir, Western Kentucky University; Steven Y.Liang, Georgia Institute of Techno1ogy;Victor Okhuysen, California Polytechnic State University, Pomona; Lewis N. Payton, Auburn University; Zhijian Pei,Kansas State University; William Schoech, Valparaiso University; Mala M.Sharma, Bucknell University; Bharat S.Thakkar, Illinois Institute of Technology: and Alan Zoyhofski, Rochester Institute of Technology. The authors would also like to acknowledge the contributions of Dr. Elliot Stern for the dynamics of machining section in Chapter 20. Dr. Brian Paul for his work on the rapid prototyping and electronics chapters, and Dr. Barney Klamecki for his help with the 9th edition. As always, our wives have played a major role in preparing the manuscript. Carol Black and Barb Kohser have endured being "tcxtbook widows" during the time when the last four editions were written. Not only did they provide loving support, but Carol also provided hours of expert proofreading, typing, and editing as the manuscript was prepared. Finally special thanks to our acquisitions editor, Joseph P. Hayton, for putting up with two procrastinating professor$ who tried both his patience and his abilities as he coordinated all the various activities required to produce this text as scheduled. We also thank Suzanne Ingrao and Sandra Dumas for all their help in bringing the 10th edition to reality.

ABOUT THE AUTHORS

J T. Black received his Ph.D. from Mechanical and Industrial Engineering, University of Illinois, Urbana in 1969, an M.S. in Industrial Engineering from West Virginia University in 1963 and his B.S. in Industrial Engineering, Lehigh University in 1960. J T. is Prolessor Emeritus from Industrial and Systems Engineering in the Samuel Ginn College of Engineering at Auburn University. H c was the Chainnan and a Professor of Industrial and Systems Engineering at The University of Alabama-Huntsville He also taught at The Ohio State University, the University of Rhode Island, the University of Vermont, the University of Illinois and West Virginia University. JT. is a Fellow in the American Society of Mechanical Engineers, the Institute of Industrial Engineering and the Society of Manufacturing Engineers. J loves to write music (mostly down home country) and poetry. Co-authoring with Ron Kohser makes this book a success,just as picking his doubles partner in tennis has given him Lhe #1 doubles ranking for 65 year olds in the State of Alabama. Ron Kohser received his Ph.D. from the Lehigh University Institute for Metal Forming in 1975. Ron is currently in his 32nd year on the faculty of the University o f Missouri-Rolla, where he is a Professor of Metallurgical Engineering and Dean's Teaching Scholar. While maintaining a full commitment to classroom instruction, he has served as department chair and Associate Dean for Undergraduate Instruction. H e currently teaches courses in Metallurgy for Engineers, Introduction t o Manufacturing Processes, and Material Selection,Fahrication and Failure Analysis. In addition to the academic responsibilities, Ron and his wife Barb operate A Miner Indulgence,a bed-andbreakfast in Rolla, Missouri.

"

'TS
71

vii

Chapter 1 Introduction to DeGarmo's Materials and Processes in Manufacturing

Materials, Manufacturing, and the Standard of Living 1.2 Manufacturing and Production Systems Case Study Famous Manufacturing Engineers 1.1

Chapter 4 Equilibrium Phase Diagrams and the Iron-Carbon System

4.1 4.2 Introduction Phases

1
1

3 27

Chapter 2
2.1 2.2 2.3 2.4 2.5 2.6

Properties of Materials 28
28 30 42 47 50 50 52 53 55

Introduction Static Properties Dynamic Properties Temperature Effects (Both High and Low) Machinability, Formability, and Weldability Fracture Toughness and the Fracture Mechanics Approach 2.7 Physical Properties 2.8 Testing Standards and Concerns Case Study Separation of Mixed Materials

71 71 4.3 Equilibrium Phase Diagrams 71 4.4 Iron-Carbon Equilibrium Diagram 79 4.5 Steels and the Simplified Iron-Carbon Diagram 80 82 4.6 Cast Irons Case Study The Blacksmith Anvils 88

Chapter 5 Heat Treatment

Introduction Processing Heat Treatments Heat Treatments Used to Increase Strength Strengthe~ing Heat Reatments for Nonferrous Metals 5.5 Strengthening Heat Treatments for Steel 5.6 Surface Hardening of Steel 5.7 Furnaces 5.8 Heat Treatment and Energy Case Study A Carpenter's Claw Hammer 5.1 5.2 5.3 5.4

89
89 89 92 93 96 109 112 114 116

Chapter 3 Nature of Metals and Alloys

3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 3.13 3.14 3.15 structure-property-processingPerformance Relationships The Structure of Atoms Atomic Bonding Secondary Bonds Atom Arrangements in Materials Crystal Structures of Metals Development of a Grain Structure Elastic Deformation Plastic Deformation Dislocation Theory of Slippage Strain Hardening or Work Hardening Plastic Deformation in Polycrystalline Metais Grain Shape and Anisotropic Properties Fracture of Metals Cold Working, Recrystallization, and Hot Working

56
56
57 57 59 59 59 61 62
63 64

Chapter 6 Ferrous Metals and Alloys

6.1 Introduction to History-Dependent Materials 6.2 Ferrous Metals 6.3 Iron 6.4 Steel 6.5 Stainless Steels 6.6 Tool Steels 6.7 Alloy Cast Steels and Irons Case Study Interior Tub of a Top-Loading Washing Machine

118
118 118 119 120 132 134 136 138

'

64 65 66 66 66 68 68 68

Chapter 7 Nonferrous Metals and Alloys

7.1 7.2 7.3 Introduction Copper and Copper Alloys Aluminum and Aluminum Alloys

139
139 140 144 152 154 155

3.16 Grain Growth 3.17 Alloys and Alloy Types 3.18 Atomic Structure and Electrical Properties

7.4 Magnesium and Magnesium Alloys 7.5 Zinc-Based Alloys 7.6 Titanium and Titanium Alloys

xii
7.7 7.8

Contents 157 157 158 159 159 160 161 10.14 Ultrasonic Inspectlon 10 15 Radiography 10 16 Eddy-Current Test~ng 10.1 7 Acoustlc Em~ss~on Monrtorlng 10.18 Other Methods of Nondestructive Testing and Inspect~on 10 19 Dormant versus Cr~tical Flaws Case Study Measunng An Angle

Nlckel-Based Alloys 5uperalloys and Other Metal7 Designed for High-Temperature Serv~ce 7.9 Lead and hn, and Iheir Alloys 7.10 Some Lesser Known Metals and Alloys 7 11 Mrtalllc Glasses 7.12 Graph~te Case Study Nonsparking Wrench

Chapter 8 Nonmetallic Materials: Plastics, Elastomers, Ceramics, a n d Composites 162

8.1 lntroductlon 8.2 l'lastlcs 8 3 Elartomers 8.4 Ceramics 8.5 Composite Materials Care Study Two-Wheel Dolly Handles 162 163 173 175 183 194

Chapter 11 Fundamentals of Casting

11.1 Introduction to Materials Processing 11.2 Introduction to Casting 11.3 Casting Terlninology 11.4 The Solidification Process 11.5 Patterns 11.6 Design Considerations in Castings 11.7 The Casting Industry Case Study The Cast Oil-Field Fitting

Chapter 9
9.1 9.2

Material Selection

195
195 197 199 200 203 204 205 207 207 208 212

Lntroduct~on Mater~al Selection and Manufacturing Processes 9 3 The Design Process 9 4 Piocedu~es f o Mater~al ~ Selection 9.5 Addit~onal Factors to Conyrder 9 6 Consideration of the Manufacturing Process 9 7 Ultimate Oblect~ve 9 8 Materldls Substitution 9 9 Effect of Product Lrability on Materials Selectron 9 10 Ards to Material Select~on Case Study Materlal Select~on

Chapter 12 Expendable-Mold Casting Processes

Tntroduct~on Sand Casting Cores and Core Making Other Expendable-Mold Processes with Mult~ple-Use Patterns 12 5 Expendable-Mold Processes Using Smgle-Use Pattern?, and F~ntshing 12 6 Shakeout, Clean~ng, 12 7 Summary 12.1 12 2 12 3 12.4

Chapter 10 Measurement a n d Inspection a n d Testing

10 1 lntroductron 10.2 Standards of Measurement 10 3 Allowance and Tolerance 10 4 Inspection Methods for Measurement 10.5 Measullng Instruments 10 6 Vision Systems for Measurement 10.7 Coordinate MedSuIlng Machrnes 10 8 Angle-Measuring Instnunents 10 9 Gages for Attributes Medsurlng 10.10 Testlng 10.11 Vlsual Inspection 10 12 I.lqutd Penetrant Inspectron 10.13 Magnetic Partlcle Inspection

213
213 214 220 227 229 238 240 240 242 245 24 7 247 248

Chapter 13 Multiple-Use-Mold Casting Processes

13.1 13.2 13.3 13 4 13 5 13.6 13.7 13.8 13.9 Introductlon Permanent-Mold Castlng Dle Casting Squeeze Casting dnd Semaolid Casbng Centrifugal Casting Continuous Castlng Melting Pourlng Practice Ciean~ng, Fln~shing, and Heat Treatlng of Cast~ngs 13.10 Automation in Foundry Operations 13.11 Process Selection Case Study Baseplate for a Household Steam Iron

Contents

xi11

Chapter 14 Fabrication or Plastics, Ceramics, and Composites

14.1 Introduction 14.2 Fabrication of Plastics 14.3 Processing of Rubber and Elastomers 14.4 l'rocessing of Ceramics 14.5 Fabrication of Composite Materials Case Study Fabrication of Lavatory Wash Barms

17.6 Pipe Welding 17.7 Presses Case Study Fabrication of a One-Piece Brass Flashlight Case

Chapter 18
18.1 18.2 18.3 18.4

Powder Metallurgy

Chapter 15 Fundamentals of Metal Forming

Introduction Forming Processes: Independent Variables Dependent Variables Independent-Dependent Relationships Process Modeling General Parameters Friction and 1.ubrication under Metalworking Conditions 15.8 Temperature Concerns Case Study Repairs to a Damaged Propeller
'

15.1 15.2 15.3 15.4 15.5 15.6 15.7

Chapter 16 Bulk Forming Processes

6 Introduction 16.2 Classification of Deformation kocesses 16.3 16.4 16.5 16.6 Bulk Deformation Processes Rolling Forging Extrusion

'

Introduction The Basic Process I'owder Manufacture Rapidly Solidified Powder (Microcrystalline and Amorphous) 18.5 Powder Testing and Evaluation 18.6 Powder Mixing and Blending 18.7 Compacting 18.8 Sintering 18.9 Hot-Isostatic Pressing 18.10 Other Techniques to Produce High-Density l'/M Products 18.11 Metal Injection Molding (MIM) or Powder Injection Molding (I'IM) 18.12 Secondary Operations 18.13 Properties of PIM Products 18.14 Design of Powder Metallurgy Parts 18.15 Powder Metallurgy Products 18.16 Advantages and Disadvantages of Powder Metallurgy 18.17 Process Summary Case Study Impeller for an Automobile Water Pump

Chapter 19 Electronic Electrochemical Chemical and Thermal Machining Processes

19.1 Introduction 19.2 Chemical Machining Processes 19.3 Electrochemical Machining Processes 19.4 Electrical Discharge Machining Case Study Fire Extinguisher Pressure Gage

16.7 Wire, Rod, and Tube Drawing 6.8 Cold Forming, Cold Forging, and Impact Extrusion 6.9 Piercing -6.10 Other Squeezing Processes 16.11 Surface Improvement by Deformation Processing :ase Study Handle and Body of a Large Ratchet Wrench
'

Chapter 17 Sheet-Forming Processes

7.1 Introduction 7.2 Shearing Operations 7.3 Bending 17.4 Drawing and Stretching Processes 17.5 Alternative Methods of Producing 'Sheet-Type Products

Chapter 20 Fundamentals of Machining/Orthogonal Machining

20.1 20.2 20.3 20.4 20.5 20.6 20.7 20.8 Introduction Fundamentals Energy and Power in Machining Orthogonal Machining (Two Forces) Merchant's Model Mechanics of Machining (Statics) Shear Strain y and Shear Front Angle q Mechanics of Machining (Dynamics)

:1
'

X~V

Contents

20.9 Summary Case Study Orthogonal Plate Machining Experiment at Auburn University

Chapter 25 Workholding Devices for Machine Tools

25.1 Introduction 25.2 Conventional Fixture Design 25.3 Design Steps 25.4 Clamping Considerations 25.5 Chip Disposal 25.6 Unloading and Loading Time 25.7 Example of Jig Design 25.8 Types of Jigs 25.9 Conventional Fixtures 25.10 Modular Fixturing 25.11 Setup and Changeover 25.12 Clamps 25.13 Other Workholding Devices 25.14 Economic Justification of Jigs and Fixtures Case Study Fixture versus No Fixture in Milling

Chapter 21 Cutting Tools for Machining

21.1 Introduction 21.2 Cutting-Tool Materials 21.3 Tool Geometry 21.4 Tool Coating Processes 21.5 Tool Failure and Tool Life 21.6 Flank Wear 21.7 Economics of Machining 21.8 Cutting Fluids Case Study Comparing Tool Materials Based on Tool Life

Chapter 22 Turning a n d Boring Processes

22.1 Introduction 22.2 Fundamentals of Turning, Boring, and Facing Turning 22.3 Lathe Design and Terminology 22.4 Cutting Tools for Lathes 22.5 Workholding in Lathes Case Study Estimating the Machining Time for Turning

Chapter 26 Numerical Control (NC) a n d the A(4) Level of Automation

26.1 26.2 26.3 26.4

702
702 710 721 725 726 730

Chapter 23 Drilling and Related Hole-Making Processes

23.1 23.2 23.3 23.4 23.5 23.6 23.7 23.8 Introduction Fundamentals of the Drilling Process Types of Drills Tool Holders for Drills Workholding for Drilling Machine Tools for Drilling Cutting Fluids for Drilling Counterboring, Countersinking, and Spot Facing 23.9 Reaming Case Study Bolt-down Leg on a Casting

introduction Basic Principles of Numerical Control Machining Center Features and Trends Ultra-High-speed Machining Centers (UHSMCs) 26.5 Summary Case Study Process Planning for the M E

Chapter 27 Other Machining Processes

27.1 Introduction 27.2 lntroduction t o Shaping and Planing 27.3 Introduction to Broaching 27.4 Fundamentals of Broaching 27.5 Broaching Machines 27.6 lntroduction t o Sawing 27.7 Introduction t o Filing Case Study Cost Estimating-Planing vs. Milling

Chapter 24 Milling
24.1 Introduction 24.2 Fundamentals of Milling Processes 24.3 Milling Tools and Cutters 24.4 Machines for Milling Case Study HSS versus Tungsten Carbide Millir

Chapter 28 Abrasive Machining Processes

28.1 28.2 28.3 28.4 28.5 28.6 Introduction Abrasives Grinding Wheel Structure and Grade Grinding Wheel Identification Grinding Machines Honing

756
756 75 7 763 767 771 780

28.7 Superfinishing 28.8 Free Abrasives Case Study Overhead Crane Installation

Chapter 32 Resistance and Solid-State Welding Processes

32 1 32 2 32 3 32.4 Introduction Theory of Reststance Welding Resistance Welding Processes Advantages and Limitations of Resistance Welding 32.5 Sol~d-State Weldlng Processes Case Study Field Repair to a Power Transformer

871
871 871 874 879 879 888

Chapter 29 Thread and Gear Manufacturing

' 29.1 Introduction
29.2 Thread Making 29.3 Internal Thread Cutting-Tapping 29.4 Thread Milling 29.5 Thread Grinding 29.6 Thread Rolling 29.7 Gear Making 29.8 Gear Types 29.9 Gear Manufacturing 29.10 Machining of Gears 29.11 Gear Finishing 29.12 Gear Inspection Case Study Bevel Gear for a Riding Lawn Mower

Chapter 33 Other Welding Processes, Brazing and Soldering

33.1 Introduction 33.2 Other Welding and Cutting Processes 33.3 Surface Modification by Welding-Relate Processes 33.4 Brazing 33.5 Soldering

, :,

;
I, i

/\
.
,

Chapter 30 Fundamentals of Joining

30.1 Introduction t o Consolidation Processes 30.2 Classification of Welding and Thermal Cutting Processes 1.3 Some Common Concerns 30.4 Types of Fusion Welds and Types of Joints 30.5 Design Considerations 30.6 Heat Effects 30.7 Weldability or Joinability 30.8 Summary

Chapter 34 Adhesive Bonding, Mechanical Fastening, and Joining of Nonmetals

34.1 Adhes~ve Bondlng Fasten~ng 34 2 Mechan~cal 34.3 Joining of Plastics 34.4 Join~ng of Ceramics and Glass 34.5 Joining of Compos~tes Case Study Golf Club Heads wlth Insert

915
915 924 927 929 929 932

! .

i,
1,

Chapter 35

Surface Engineering

1
,

!;
,~

Chapter 31 Gas Flame and Arc Processes

31.1 31.2 31.3 31.4 31.5 31.6 31.7 31.8 .9 Oxyfuel-Gas Welding Oxygen Torch Cutting Flame Straightening Arc Welding Consumable-Electrode Arc Weldlng Nonconsumable-Electrode Arc Welding Welding Equipment Arc Cutting Metallurgical and Heat Effects in Thermal Cutting Ise Study Bicycle Frame Construction and Repair

1 :.
!'

35.1 Introduction 35.2 Mechanical Cleaning and Finishing Blast Cleaning 35.3 Chemical Cleaning 35.4 Coatings 35.5 Vaporized Metal Coatings 35.6 Clad Materials 35.7 Textured Surfaces 35.8 Coil-Coated Sheets 35.9 Edge Finishing and Burrs 35.10 Surface Integrity Case Study Dana Lynn's Fatigue Lesson

Chapter 36

Quality Engineering 969

969 970 981

36.1 Introduct~on 36 2 Determining Procesr Capability 36 3 Inspection to Control Quality

XV~

Contents

36.4 Process Capability Determination from Control Chart Data 985 36.5 Determining Causes for Problems in Quality 986 36.6 Summary 996 Case Study Boring QC Chart Blunders 1000

Chapter 39 Rapid Prototyping, Tooling And Fabrication (web-based chapter)

(w.wil~.com/college/DeGamo)

Chapter 37 Manufacturing Automation (web-based chapter)

(www.wiley.com/college/DeGanno)

Index Selected References for Additional Study (web-based)

Chapter 38 The Enterprise (web-based chapter)

(www.wiley.com/college/D&anno~

1 . 1 MATER~ALS, MANUFACTURWG, AND THE STANDARD OF LIVING

1.2 MANUFACTURING AND PRODUCT~ON SYSTEMS Pxoduction System-The Enterprise Manufacturing Systems Manufacturing Processes Job and Station Operation

Treatments Too!.s,TooIing, and Workholders Toohng for Measurement and Inspectton lntegratmg Inspection into the Process Products and Fabr~cahons Workpiece and its Configurat~on Roles of Engineers 1n Manufacturing

Changing World Competit~on Manufacturing System Des~gns Basic Manufactur~ng Processes Other Manufacturing Operat~ons Understand Your Process Technology Product Life Cycle and Life-Cycle Cost Manuractur~ng System Deslgn New Manufacturing Systems

1.1 MATERIALS, MANUFACTURING, AND THE STANDARD OF LIVING

Manufacturing 1s critical to a country's economic welfare and standard o C living because the standard of hving in any society 1s determmed, pnmar~ly, by the good! and sewLces that are dvailahle to its people. Manufacturmg companies contribute about 20% of the GNP, employ about 18% of the workforce, and account for 40% of the exports of the United States In most cases, matenals are utilized in the form of manufactured goods. Manufacturing and assembly repreaent the organized actlv~ties that convert raw materials into salable goods.The manuiacturcd goods are typically divlded into two classes: producer goods and consumer goods. Producer goods are those goods manufactured lor other companies to usc to manufacture either producer or consumer goods. Conwmer goods are those purchased directly by the consumer or the general public For example,someonc has to build the machine tool (a lathe) thar produccs (usingmachaing processes) the large rolls that are sold to the rolhng mill fdctory to be used to roll the sheets of steel that are then formed (using d m ) to become the body panels of yourcar. S~m~larly, many service industr~es depend heavily on the use of manufactured products, lust as the agricultural industry is heavily dependent on the use of ldrge farming machines for efficient production. Convert~ng materials from one form to another adds value to them The more efficiently mater~als can be produced and converted into the desi~ed products that funclion with the prescr~hed quahty, the greater w~ll he the compan~es' product~vity and the better w~ll be the standard of l~ving of the employees The h~story of man has been linked to h ~ ability s to work with matenals, beginn~ng with the Stone Age and ranging through the eras of copper and bronze, the Iron Age, and recently the age of steel.While ferrous materials still dominate the manufactunng world, we are entering the age of tallor-made plastics, composite materials, and exotic alloys. A good example of this progression is shown in F~gure 1-1.The goal of the manufacturar of any product or service is to cont~nually improve. For a glven product or service, this improvement process usually follows an S-shaped curve, as shown in Figure 1-1(a), often called a product hfe-cycle curve. After the in~tial inventronlcreation, growth in performance occurs, with relat~vely few resources reqmred. a penod of r a p ~ d However, each improvement becomes progress~vely more d~fficultFor a delta gain,

Product Develooment
S-Curve

Design optimization architectures

J
I

Develooment
I

Superposition &design enhancements

1945 1960 1975 1990

Tlme resource
(a1

2005

(b)

FIGURE 1-1 (a) A product development curve usually has an "S"-shape. (b) Example of the S-curve for the pneumatic radlal ttre (Courtesy of: Bart Thomax, M~chel~n)

more money and time and ingenuity are required. Finally, the product or service enters the matunty phase, during which additional performance gains become very costly. For example, ~nthe automob~le tlre industry,Flgure 1-lb shows the evolution of radial tire performance from its birth in 1946 to the present. Growth in performance is actually the superposition of many different improvementsin material,processes, and design. These innovations,known as susrozning technology, serve to continually bring more value to the consumer of existing products and services. In general, susta~ning manufacturing technology is the backbone of American ~ndustry and the ever-increasing productivity metnc. Although materials are no longer used only in their natural state, there is obvlously an absolute h t to the amounts of many materials available here on earth Therefore,as the variety of man-made materials continues to mcrease, resources must be used efficiently and recycled whenever possible. Of course, recycling only postpones the exhaustion date, Like materials, processes have also proliferated greatly in the last 40 years, with new processes being developed to handle the new materials more efficiently and with Less waste.A good example is the laser, invented around 1960,which now finds many uses in manufacturing, measurement, inspection, heat treating, welding, and more. New developments in manufacturing technology often account for rmprovements in productivity. Even when the technology is proprietary, the competition often gains access to it, usually quite quickly Start~ng with the product destgn, materials, labor, and equipment are interactive factors in manufacturing that must be combined properly (mtegrated) to achleve low cost, supenor quality,and on-time delivery Typically, as shown in F~gure 1-2,40% of the
Selling

price

Manufacturing cost ~ 4 0 % selling price

_____--cost, 15% marketing,

Equipmem Plant

Subassemblies

cost is the largest part of the selling prlce, usually around 40%. The largest part of the manufacturing cost IS materials, usually 50%.

FIGURE 1-2 Manufacturing

mbterials Manufactur~ng
-20%

selhng price of a product is manufacturing cosr. Slnce the selhng pnce IS determined by the customer,maintaining profit often depends on reducing manufacturing cost.The internal customers who really make the product, called d~rect labor, are usually the targets of automation, but typically they account for only about 10% of the manufacturing w s t even though they are the main element in increawng productivlty.In Chapter 39, a manufacturing strategy is presented that attacks the materials cost, indrrect costs, and general admiuistrat~on costs, in addition to labor costs The materials costs Include the cost of storing and handling the materials wthin the plant The strategy 1s called lean productron Referring again to the total expenses shown in F~gure 1-2 (sellmgpr~ce less profit), about 68% of dollars are spent on people, the breakdown being about 15%for engineers, 25% for marketing, sale& and general management people; 5 % for direct labor, and 10% for indlrect labor.The average labor cost in manufacturing in the Unlted States was around $15 per hour for hourly workers in 20(30.Reductions in direct labor w~ll have only marglnal effects on the total people costs. The optimal combination of factors for producing a small quantlty of a given product may be vcry inefficient for a larger quantity of the same product Consequently, a systems approach, taking all the factors into account,must be used. T h o requires a sound nnd broad understandrng on the par[ of the decision makers on the value of materzals,yrocedses, and equrpment to the company, accampunred by an understanding of the manufacturing systems Materials and processes in manufacturing systems are what thls book m all about

MANUFACTURING AND PRODUCTION SYSTEMS

Manufacturing is the econonuc term for making goods and servlces available to satisfy human wants. Manufacturing implies creating value by applying useful mental or physical 1abor.The manujactwringprocesses are collected together to form a manufacturzng elements system (MS).The manufacturing system 1s a complex arrangement of phys~cal characterized by measurable parameters (F~gure 1-3) Thc manufacturing system takes inputs and produces products for the external customer. The entire company is often referred to as the enterpr~se or,m this textbook, the production system.The production system includes the manulactunng system, as shown in Figure 1-4, and services it. In this book, aproductzon syrtern will refer to the total company and wdl mclude within it the manufacturing system (SPSs) The production systemincludes themanufactur~ng system plus all the otherhnctional areas of the plant for informat~on, des~gn, analysis, and control. These subsystems are connected by vanous means to each other to produce e~ther goods or services or both. Goods refer to material things. Servzces are nonmaterial things that we buy to sattransportaisfy our wants, needs, or deslres. Service productzon syytems (SPSs) ~nclude tlon, banking, finance, savings and loan. Insurance, ut~lities, health care, education, comrnun~cation, entertamment, sportmng events, and so forth They are useful labors that do not directly produce a product. Manufacturing has the responsibility for desigmng processes (sequences of operations and processes) and systems to create (make or
Inputs include
Raw materials

1 I

A Flow Shop Manufacturing System

I /
External

-WeerSubassembIy[

Components
FIGURE 1-3

The manufacturing system converts inputs to outputs using processes to add value to the goods for the external customer.

rnformarlon Sequence of Processes

*Goods

Products

CHAPTER 1 lntroductlon to DeGarmo's Mater~als and Processes In Manufacturing

Legend: information systems Instructions or orders Feedback

customer Deliver

Market

Distribution
I ,

department

Inspection (quality control) Finished

--. -Recommend
design changes

Finance department

,
Product

Purchasing department delivery

/

manufacturing Drawings, department specifications, and standards Production Process i I and inventory design How to make
(schedulesi

the product

f
i

L * 9

Material requisitions , Design &test& redesign new products

q ,

FIGURE 1-4 The functions and systems of the production system, whlch Includes (and services) the r~!;r!!.tl~cr,r~~iq ,y,rer!!. Trw f,tict u-tla.oepdrtn1e"ts are conn6cren 0).lormn aIw ~rilort!!~ of'rnidri'n

\).srems derqned to %er\'<e the rna~i..larl~.r'n~j \y%crn rtiat prodrces the qoods.

manufacture) the product as designed.The system must exhibit flexibility to meet customer demand (volumes and mixes of products) as well as changes in product design. As shown inTable 1-1,production terms have a definite rank of importance,somewhat like rank in the army. Confusing system with section is similar to mistaking a colonel for a corporal. In either case, knowledge of rank is,necessary.The terms tend to overlap because of the inconsistencies of popular usage. An obvious problem exists here in the terminology of manufacturing and production. The same term can refer to different things For example, drill can refer to the machine tool that does these kinds of operations: the operation itself,which can be done on many different kinds o f machines; or the cutting tool, which exists in many different forms. It is therefore important to use modifiers whenever possible: "Use the mdial drill press to drill a hole with a 1-in.-diameter spade drill."The emphasis of this book will be

p a m or products, mntEgretnon of all units of the

intetpnse ;r the business.

Hachine or mach~ne btacl or manuP@tunng proem?

Sob (sometimualled a smttm.a

collmuon ot tasks) Operatrow (sometrmes called

aprme~~)

matenak-handhng cqutpment, and operators A speclfic pmoc of equ~pment designed to accomplish spentlc process% often called a rnodrtne loo/, nlachlne tools linked tpgethcr to make a manufactur~ngsgstem. A collection 116 aperattons dnne on machmes or a collection of tasks perfoimed by one worker at one location on the assembly ltne Aspecrf~c actton or Lrcatmont,ofIen done on a machine, thc coilection qf which makes up the job at a worker

Spot weld~ng, m l h g mach~ne. lathe, dnli prcss,forge, drop hammor,d~ecaster. punch prerv gr~nder, etc

DnII, ream, bend,soider, turn,face,mlll extrude ~nspecl. load

Grinding whwl,dr~ll bit,end m1l11ng cutter.die, mold,clamp, three-jaw chuck, Gxture

To615 or tooling

directed toward the understanding of the processes, machines, and tools required for manufacturing and how they intera~t with the materials belng processed.Tn the last s e c tian of the book, an introduotion to systems aspects is presented.

PRODUCTION SYSTEM-THE

ENTERPRISE

The highest-ranking term in the hierarchy ts ~ F O ~ U G .systern.A ~ L O ~ production system includes people, money, equipment, materials and supplies, markets, management, and the manufacturing system. In fact, all aspects of commerce (manufacturing, sales, advertising, profit, and distribution) are involved. Table 1-2 provides a partial list of productian systems.

Aerosp'dce and a~rplanes

AppIlancRE Automouve (cars, trucks, vanb. wagons etc f Beverages Bulldmg suppljei. (hardware) Cement and asphalt Cerami~s Chemcds and allied indvstries Clothing (garments) Chnstrwctron Conshuct~on maten&ls(bnck,block, @ends) Dl ugs, soaps, cosmel~cs Electrical and inrcroeleetroniu Energy (power, gas, electric) Engineering Kqmpment and mnch~nery fagrlwitura1,umsfNctrm and electrical products. eleotronrcs,housahold products, industrial m d m o tools, office equipment, computers, power generators)

Food, (canned,dalry. meats, etc ) Footwear

firmrum
Glass
Hospital wpplrers

Leather and fur goods Machines Marine mglnrarmg Metals [steel, dlummum, e t c ) Natural resources (oil, cod, forest, pulp and paper) Puhlwhmgaod prrnflng (bonk~,CDs,newspapers) Restaurants Retail (food, department stores, etc) Shlp hu~ldnng Texules Tire and rubber Tobacco Transportatlen veh~cles (ratlroad, airlina truck, bus) Vehlcles (bikea, cyoles,ATVs, snuwmob~lesi

CHAPTER 1 lntroduct~on to DeCarmo's Materials and Processes In Manufacturing

Advertising and marketing Communication (telephone, computer networks) Education Entertainment (radio,TV, movie~plays) Equipment and furnnure 1enta1 F~nancial (banks, mveslmenl cnmpnmcx loan ~ornpan~es) Health care Insurance Transportation and car rental Travel (hotel, motel, cruse ilnes)

require an SPS for proper product sales. This is particularly true in industries, such as the food (restaurant) industry, in which customer service is as important as quality and on-time delivery.Table 1-3provides a short List of service industries.

MANUFACTURING SYSTEMS A collection of operations and processes used to obtain a desired product(s) or componentjs) is called a manufacturing system. The manufacturing system is therefore the design or arrangement of the manufacturingprocesses. Control of a system applies to overall control of the whole, not merely of the individual processes or equipment.The entire manufacturing system must be controlled m order ta schedule and control production, inventory levels, product quality, output rates, and so forth.

MANUFACTURING PROCESSES A manufatturingprocm converts unfinished materials to finished products, often using machines or machine tools For example, injection molding, die casting, progressive stamping, nulling, arc welding, pamtmg, assembling, testing, pasteurizing, homogenizIng, and annealing are commonly calledprocesses or manufactunngprocesses.The term process often implies a sequence of steps, processes, or operations for production of goods and services, as shown in Figure 1-5,which shows the processes to manufacture an Olympic-type medal. A machrne tool is an assembly of related mechanisms on a frame or bed that together produce a desired result. Generally, motors, controls, and auxiliary devices are included. Cutting tools and workhold~ng devices are considered separately. A machine tool may do a single process (e.g., cutoff saw) or multiple processes, or it may manufacture an entire component. Machine sizes vary from a tabletop drill press to a 1000-ton forging press. JOBAND STATION In the classical manufacturing system, a job iq the total of the work or duties a worker perf0rms.A station IS a location or area where a production worker performs tasks or his job. A job is a group of related operations and tasks performed at one station or series of stations in cells. For example, the job at a final assembly statlon may consist of four tasks.

1. Attach carburetor. 2. Connect gas line. 3. Connect vacuum line. 4. Connect accelerator rod.
The job of a turret lathe (a semiautomat~c machsne) operator may include the following operations and tasks: load, start, index and stop, unload, mspect.The operator's job may also include setting up the machine (i e., getting ready for manufacturing). Other machme operahons Include drilling, reaming, facing, turning, chamfering, and knurling. The operator can run more than one machine or service at more than one station. The termslob andsturion have been carr~ed over to unmanned machines. A lob is a group of related operations generally performed at one station, and a station is a positlon or location in a machsne (or process) where specific operations are performed.A simple machine may have only one station. Complex machines can be composed of many stations. The job at a station often includes many simultaneous operations, such as "drill all five holes" by multiple spindle drills In the planning of a lob, a process plan is often developed (by the engineer) to describe how a component 16 made using a

SZ~TION 1 . 2

Manufacturing and Production Systems

How an olympic medal is made using the CAD/CAM process

'll An oversized 30 plaster model 1 s made

from the artlst's conceptual drawlngs

.*

-.*

2 ) The model 1 s scanned with a laser to

produce a dlgltal computer called a Computer-aided desrgn (CAD).

Ton

,Die set

Bottom to01

3) The computer has software to produce a program to drive numetical c6mrol machlne to cut a dte set.

41 Blanks are cut from brenze medal sheet

abrasrve water jet under 2-axis CNC control

5) The blanks are

heated and placed between the top die and bottom
VWSSUF~ IS

applred by a press at very slow rates. deforms ~rito the press medal Th~g

water inlet

Addltlonal finishing steps in the proaess Include chem~cal etching; gold o l sllver platrng; psckaglng

FIGURE 1-5 The manufacturing process for making Olymprc medals has many steps or operations, beginning wlth design and lnclud~ng dm making.

and Processes In Manufacturing CHAPTER 1 lntroductron to DeGarmo's Mater~als

1
Raw material bar stock cylinder with flat ends

S~mpllfled Sequence o f

Operatrons

Cut bar stock to length,

centerdr~ll ends

Multiule cvlinders

Turn (on a lathe), rough turn and f~nish turn

Three external
cyl~nders and four flats

Turn Ion a lathe)the smaller

external cyllndr,eal surfaces.

FlatThree cylinders and six flats

Mill the flat on the r ~ g h end. t M ~ l the l slot on the left end.

~nternal holes

Dr~li four holes on left end Tau [~nternal threads) holes.

FIGURE 1-6 The component called a ptnlon shaft 1 s manufactured by a "sequence of operatlons'' to produce vanous geometrtc surfaces The enqlneer figures out the sequence and selects the tool~ng to perform the steps.

requrnce of operatzan. So, for example, the part shown in Figure 1-6 is produced by a set of machining operations This Information can be embedded in a CNC program,as shown In F~gure 1-5.

OPERATION
An operatLon is a dtstmct action performed to produce a desired result or effect.Typical manual machine operations are loading and unloading. Operations can be divided into suboperational elements. For example,loadmgis made up of ptcking up a part,placIng part in jig, clostng jig. However, suboperational elements will not be discussed here. Operattons categor~red by functlon are:

1. Materials handling and tran~porting: change in posltion of the product

2. Proccsszng:change in volumc and quality, ~ncludtng assembly and dsassembly; can Include packaging 3. Packaging:special processmng; may be temporary or permanent for shipping 4. lnspecang and test~ng: companson to the standard or check of process behavior 5. Stonng: time lapses without further operatlons

SECTION 1.2 Manufactur~ng and Production Systems

These basic operations may occur more than once ln some processes,or they may sometimes be omitted. Remember, ~t is the mufacturingprocesres that chenge the vulue and quality of the materials. Defective processes produce poor quality or scrap. Other operations may be necessary but do not, in general, add value, whereas operations performed by mach~nes that do material processing usually do add value.

TREATMENTS Treatments operate continuously on the workpiece They usually alter or modify the product-in-process without tool contact Heat treating, curing, galvanizmg, plating, finrshing, (chemical) cleaning, and painting are examples of treatments. Treatments usually add value to the part. These processes aredifficult to includeincells because they often have long cycle times, are hazardous to the workers' health, or are unpleasant to be around because of h~gh heat or chemicals. They are often done in large tanks or furnaces or rooms. The cycle time for these processes may dictate the cycle times for the entlre system These operations also tend to be materlal specific. Many manufactured products are given decoratlve and protective surface treatments that conlrol the finished appearance. A customermay not buy a new vehicle because it has a vlslble defect m the chrome bumper, although this defect will not alter the operation of the car. TOOLS, TOOLING, AND WORKHOLDERS The lowest mechanism in the production term rank is the tool. Tools are used to hold, cut, shape, or form the unfimshed product. Common hand tools include the saw, hammer, screwdnver, chisel, punch, sandpaper, drill, clamp, file, torch, and grindstone Basically,mach~nes are mechanized versions of such hand tools and are called cutling tools Some examples of tooh for cutting are drill bit&reamers, single-point turning tools, milling cutters, saw blades, broaches, and grindrng wheels. Noncuttlng tools for forming Include extrusion dies, punches, and molds. Tools also include workholders, jlgs, and fmtures.Thehe tools and cuthng tools are generally referred to as the tooling, which usually must he considered (purchased) separate from machine tools. Cutting tools wear and fail and must be periodically replaced before parts are ruined The workholding devices must be able to locate and secure the workpieces during processing in a repcatable, mistake-proof way. TOOLING FOR MEASUREMENT AND INSPECTION Measuring tools and instruments are also import.dnt for manufacturing. Common examples of mcasuring tools are rulers, calipers,micrometers, and gages Precision devices that use laser optics or vision systems coupled with sophlsticared electronics are becoming commonplace Vluon systems and coordinate measuring machines are becommg critical elements for ach~eving superior quality. INTEGRATING INSPECTION INTO THE PROCESS The Integration of the inqpection process into the manuCactunng process or the manufacturing system is a critical step toward building products of supenor quality. An example will help. Compare an electric typewriter w ~ t h a computer that does word processing.The electric typewriter is flex~ble. It types whatever words are wanted in whatever order. It can type in Pica,Ehte, or Orator, but the font (disk or ball that has the appropriate type size on it) has to be changed according to the size and face of type wanted. The computer can do all 01 this but can also, through its software, do italics, darken the words, vary the spacing to justify the right margin, plus many other functlons It checks immediately for Incorrect spelling and other defects like repeated words. The software system provides a signal to the hardware to flash the word so that the operator w~ll know something is wrong and can make an immediate correction. If the system were des~gned to prevent the typ~st from typing repeated words, then thrs would be a poka-yoke, a defect prevention. Defect prevention is better than immed~ate defect detection and correction. Ultimately, the system should be able to iorecast the probability

10

CHAPTFR 1 Introduction to DeCarmo's Mater~als and Processes In Manufacturing

of a defect, correcting the problem at the source. This means that the typist would have to be removed from the process loop, perhaps by having the system type out what it is told (convert oral to written directly). Poka-yoke devices and source inspection techniques are keys to designing manufacturing systems that produce superior-quality products at low cost.

PRODUCTS AND FABRICATIONS In manufacturing, material things (goods') are made to satisfy human wants. Products result from manufacture. Manufacture also includes conversion processes such as refining, smelting, and mining. Products can be manufactured by fabricating or by processing. Fabricating is the manufacture of a product from pieces such as parts, components, or assemblies. Individual products or parts can also be fabricated. Separable discrete items such as tires, nails, spoons, screws, refrigerators, or hinges are fabricated. Processing is also nsed to refer to the manufacture of a product by continuous means, or by a continuous series of operations, for a specific purpose. Continuous items such as steel strip, beverages, breakfast foods, tubing, chemicals, and petroleum are "processed." Many processed products are marketed as discrete item&such as bottles of beer, bolts of cloth, spools of wire, and sacks of flour. Separable discrete products, both piece parts and assemblies, are fabricated in a plant,fuctory, or mill, for instance, a textile or rolling mill. Products that flow (liquids, gases, grains, or powders) are processed in aplant or refinery.The continuous-process industries such as petroleum and chemical plants are sometimes called processing industries or flow industries. To a lesser extent, the terms fabricating industries and manufacturing industries are nsed when referring to fabricators or manufacturers of large products composed of many parts,such as a car, a plane, or a tractor. Manufacturing often includes continuousprocess treatments such as electroplating,heating, demagnetizing, and extrusion forming. Construction or building is making goods by means other than manufacturing or processing in factories. Construction is a form of project manufacturing of useful goods like houses, highways, and bnildings.The public may not consider construction as manufacturing because the work is not usually done in a plant or factory, but it can be.There is a company in Delaware that can build a cusLom house of any design in its factory, truck it to the building site, and assemble it on a foundation in two or three weeks. Agriculrure,j%heries,and commercialfishing produce real goods from useful labor. Lumbering is similar to both agriculture and mining in some respects and miningshould be considered processing. Processes that convert the raw materials from agriculture, fishing, lumbering, and mining into other usable and consumable products are also forms of manufacturing. WORKPIECE AND ITS CONFIGURATION In the manufacturing of goods, the primary objective is to produce a component having a desired geometry, size, and finish. Every component has a shape that is bounded by various types of surfaces of certain sizes that are spaced and arranged relative to each other. Consequently, a component is manufactured by producing the surfaces that bound the shape. Surfaces may be:
1 . Plane or flat 2. Cylindrical (external or internal) 3. Conical (external or internal) 4. Irregular (curved or warped) Figure 1-6 illustrates how a shape can be analyzed and broken up into these basic bounding surfaces. Parts are manufactured by using a set or sequence of processes that will either (1) remove portions of a rough block of material (bar stock, casting,forging) so as to produce and leave the desired bounding surface, or (2) cause material to form into a stable configuration that has the required bounding surfaces (casting, forging). Conse-

SECTION 1.2 Manufactunng and Produalon Systems

11

quently, in designing an object, the designer speciftes the shape, size, and arrangement of the boundingsurface.The part des~gnmust be analyzed to determine what materials will provide the desired properties, including mating to other components, and what processes can best be employed to obtaln the end product at the most reasonable cost. This IS often the job of the manufactmng engineer.

ROLES OF ENGINEERS I N MANUFACTURING Many engineers have as their function the designing of products The products are brought into reallty through the processing or fabncation of matenals. In this capacity designers are a key factor in the material selection and manufacturingprocedure. A desrgn engcneer, better than any other person, should know what the design is to accomplish, what assumptions can be made about service loads and requirements, what service envuonment the product must withstand, and what appearance the final product is to have.To meet these requrements, the materzal(s) to beused must be selected and specified In most cases, to utllize the material and to enable the product to have the desired form, the designer knows that certain manufacturingprocases will have to be employed In many instances, the selection of a specific matenal may dictate what processing must be used. On the other hand, when certain processes must be used, the design may have to be modified in order for the process to be utllized effectively and economically. Certain dimensional sizes can dictate the processing, and some processes require certain sizes for the parts going into them. In converting the design into reality, many dec~sions mu% be made. In most mstances, they can b e made most effectively at the design stage. It is thus apparent that design engineers are a wtal factor in the manufacturing process, and it is indeed a blessing to the company if they can des~gn for manufactunng, that is, design the product so that ~tcan be manufactured and/or myembled economicaUy (i.e., at low unit cost). Design for manufacturing uses the knowledge of manufacturing processes, and so the design and manufacturing engineers should work together to integrate deslgn and manufacturing achvities. Manufacturing engzneers select and coordinate spec~fic processes and eqmpment to be used,or supervise and manage their use. Some design special tooling is used so that standard machines can be utllized in producing specific products.These engineers must have a broad knowledge of manufacturingprocesses and material behavior so that desired operations can be done effectively and efficiently without overloading or damagmg machines and without adversely affecting the materials being processed. Although it is not obvious, the most hostile environment the material may ever encounter in itslifetune is the processing environment. Industrral or manufacturing engineers are responsible for manufacturing systems deslgn (or layout) of factorlesThey must take Into account the interrelationships of the design and the properties of the materials that the mach~nes are going to process as well as the interreact~on of the materials and processesne choice of machines and eqnipment used in manufacturing and thelr arrangement in the factory are also design tasks. Materzals engzneers devote their major efforts to developing new and better materials. They, too, must be concerned with how these matenals can be processed and with the effects that the processing will have on thepropert~es of the materials Although their roles may be quite different, it is apparent that a large proportion of engineers must concern themselves with the interrelationships of materials and manufacturing processes. As an example of the close interrelationship of design,materials select~on, and the selection and use of manufacturing processes,consider the common desk stapler. Suppose that this item is sold at the retail store for $20. The wholesale outlet sold the stapler for $16 and the manufacturer probably received about $10 Tor it. Staplers typically consist of 10 to 12 parts and some rivets and pins.Thus the manufacturer had to produce and assemble the 10 parts for about $1per part. Only by giving a great deal of attention to design, selection of materials, select~on of processes, select~on of equipment used for manufacturing (tooling), and utilization of personnel could such a result be achieved. The stapler is a relatively simple product, yet the problems involved in its mannfacture are typical of those that manufacturing indnstnes must deal with.The elements

I
-

12

C H A P ~1R lntroductlon to DeGarmo's Mater~als and Processes I n Manufactur~ng

of design, materials, and processes are all closely related, each having its effect on the performance of the device and the other elements For example, suppose the designer calls for the component that holds the staples to be a metal part. Will it be a machined part rather than a formed part? Ent~rely different processes and materials need to he specifled depending on the cho~ce. Or, if a palt 1s to be changed from metal to plast~c, then a whole new set of fundamentally different matelials and processes would need to come into play. Such changes would also have a s~gnlficant impact on cost.

CHANGING WORLD COMPETITION In recent year4major changes in the world of goods manufacturing have taken place. Three of these are:
1. Worldwide compet~tion for global products and their manufacture 2. Hlgh-tech manufacturing or advanced technology 3. New manufacturing systems deslgns, strategic?, and management Worldwide (global) competition 1s a fact of manufacturing life, and it w11I get stronger in the future. The goods you buy today may have been made anywhere in the world. The second aspect, advanced manufacturing technology, usually refers to new mach~ne tools or processes with computer-a~dedmanufacturing Produc~ng machine tools 1s a small industry with enormous leverage Improved processes lead to better components and mole durable goods. However, the new technology is often purchased from companies that have developed the technology, ro this dpproach 1s Important but may not provide a unique compet~tive advantage if your competitors can also buy the technology, provided that they have the cap~tal. Some companies develop their own unique process technology and try to keep it proprietary as long as they can. A good example of unique process technology is the numerical control machine tool, shown in F~gure 1-5 and d~scussed in Chapter 27 Computer-controlled machines are now common to the factory floor. The t h ~ r d change and perhaps the real key to succesv in manuCacturmg is to build a manufacturing system that can dehver, on time to the customer, super-quahty goods at the lowest posslble cost in a flexible way. This change reflects an effort to Improve markedly the methodology by which goods are produced rather than simply upgrading the mannfactur~ng process technology. Manufacturing system des~gn is discuased extens~vely in the l a ~section t of the book, and we recommend that students examine this material closely after they have gamed a working knowledge of materials and processes. Thc next sectlon provides a bnef ~ntroduct~on to manufacturing system designs.

MANUFACTURING SYSTEM DESIGNS Flve manufacturing system designs can be ident~fied: the job shop, the flow shop, the Imked-cell shop, the project shop, and the contlnuous process. The latter system prlmar~ly deals withl~quidsandlor gases (such as an oil refinery) rather than solids or discrete parts. The most common of these layouts is thelob shop, character~zed by large varleties of components, general-purpose mach~nes, and a funct~onal layout (Fignre 1-7). Thls means that mach~nes are collected by function (all lathes together, all broaches together, all mill~ng machines together) and the parts are routed around the shop in small lots to the various machines.The inset shows the mult~ple paths through the shop and a detall on one of the seven broaching machine lools The materlal is moved from machme to machine in cartc or containers and is called the lo1 or butch. Flow shops are character~zed by larger volumes of the same part or assembly,spec~al-purpose machmes and equipment, less variety, and more mechanization. Flow shop layouts are typically either contlnuous or interrupted and can be for manufacturing or assembly, as shown in Figure 1-8.If contznuouA, a production line is built that basically runs one large-volume complex item in great quantlty and nothing else. The common l~ght bulb is made this way A transfer lme producing an engine block is another typical

SECTION 1.2

Manufacturing and Production Systems

13

FIGURE 1-7 This rack bar machining area is functionally des~gned so it operates l~ke a job shop, with lathes, broaches, and grinders llned up.

.& s y
Steel fragments

% .%

$.

1 EleCfrontc pads ]

FIGURE 1-8 The moving assembly line for cars is an example of the flow shop.

14

CHAPTER 1

lntroduct~on to Decarmo's Mater~als and Processes In Manufadurlng

BEFORE: Layout with conveyor-Subassembly with two conveyors

- -~~

~~

Workers

Measurable Parameters
Before

AFTER: U-Shape layout-Conveyors removed

In-process rnventory Personnel Daily output per person cycle tlme

After: 1056 unrtscshi* 8 units

8 operators 132 unlE$

Standing, walking

workers

FIGURE 1-9 The tradit~onal subassembly lines can be redesigned into U-shaped cells as part of the conversion af mass production to laan pduction.

example if intwuptt.$, the line manufactures large lots but is periodically "chansed over" to run a simil~r but different component. The linked-cell manufactuing system (L-CMS) is composed otmanufacturing and subassembly cells (Figure 1-9) connected to final assembly [linked) usinga unique form of inventory and mfonation wniro1 called kanbnn. The L-CMS is used in lean pioductim systems where manufacturing processes and subas6emblies are restructured mnte U-shaped cells so they can operate on a one-piece-flow basis, like final assembly. The prolect shop is charatterked by the immobility of the item being manufacturd. Ln the consb.u&on indusfry, bridges and roads are good examples. In the manufactufe o f goods, large dirplanes, ships, large machine tools, and locomotives are manufactured in projecf shqs. It is necessary that the workers, machines,and materials come to the dte. l b e number of end items is not very large, and therefore the lor sizes a f the unnponenfs going into the end item are not large. Thus the job shop usually supplies parts and subassemblies to the pr~~&.Ct shop in small lots. Continuous processa are used to manufacture liquid%011%gases, and powders. These manufacturing systems are usually large plants producing goods for either producers or mass-producing canned or bottled goods for consumers.Tbe manufacturing engineer in these f a c t o h IS often a chemkal engineer. Natwalllt, there are many hybrid forms of these manufacturing systems, but the job shop is the most common system. Bemuse of its design,the job shop has been shown to be the least cost-efficient of all the systems. Component parts in a typical job shop spend only 3% ofthelr time in machines and the rest of the bme waiting orbeingmovedfrom one functional mea to the next Qnce the part is on the machine, it is actually being processed (i.e., having value added to it by the changmg of its shape) only about 30% to 40% a f the time. The rest of the time parts are being loaded, unloaded, inspected, and so on. The advent of nunaericgl cmtrol machines increased thepereentage of $$me that themachine is making because tool movements are programmed and the machines can automatically change tools or load or unload parts.

SECTION 1.2 Manufacturing and Production Systems

15

However,there are a number of trends that are forcing manufacturingmanagement to consider means by which the job shop system itself can be redesigned to improve its overall efficiency.These trends have forced manufacturing companies to convert their batch-oriented job shops into linked-cell manufacturing systems, with the manufacturing and subassembly cells structured around specific products. Another way to identify families of products with a similar set of manufacturing is called group - processes iechnology. Group technology ( G T )can be used to restructure the factory floor. GT is a concept whereby similar parts are grouped together into part families. Parts of similar size and shape can often be processed through a similar set 0.fprocesses.A part family based on manufacturing would have the same set or sequences of manufacturing processes.The set of processes is called a cell. Thus, with GT, job shops can be restructured into cells, each cell specializing in a particular family of parts.The parts are handled less,machine setup time is shorter, in-process inventory is lower, and the time needed for parts to get through the manufacturing system (called the throughput time) is greatly reduced.
A

BASIC MANUFACTURING PROCESSES It is the manufacturing processes that create or add value to a product. The manufacturing processes can be classified as:
Casting, foundry, or molding processes Forming or metalworking processes Machining (material removal) processes Joinlng and assembly Surface treatments (finishing) Rapid prototyping Heat treating Other These classifications are not mutually exclusive. For example, some finishing processes involve a small amount of metal removal or metal forming. A laser can be used either for joining or for metal removal or heat treating. Occasionally, we have a process such as shearing, which is really metal cutting but is viewed as a (sheet) metalforming process. Assembly may involve processes other than joining. 'The categories of process types are far from perfect. Casting and molding processes are widely used to produce parts that often require other follow-on processes, such as machining. Casting uses molten metal to fill a cavity. The metal retains the desired shape of the mold cavity after solidification.An important advantage of casting and molding is that, in a single step, materials can be converted from a crude form into a desired shape. In most cases, a secondary advantage is that excess or scrap material can easily be recycled. Figure 1-10 illustrates schematically some of the basic steps in the lost-wax casting process, one of many processes used in the foundry industry. Casting processes are commonly classified into two types: permanent mold (a mold can be used repeatedly) or nonpermanent mold (a new mold must be prepared for each casting made). Molding processes for plastics and composites are included in the chapters on forming processes. Forming and shearing operations typically utilize material (metal or plastics) that has been previously cast or molded. In many cases the materials pass through a series of forming or shearing operations, so the form of the material for a specific operation may be the result of all the prior operations.The basic purpose of forming and shearing is to modify the shape and size andlor physical properties of the material. Metalfornzing and shearing operations are done both "hot" and "cold," a reference to the temperature of the material at the time it is being processed with respect to the

The Lost-Wax Casting Process

The most common castlng method through the ages wasfllt are perdue or lost-wax process Although evpenslve end tlme consuming, tho lost way method allows the artst to accurately reproduce the del~cata nuances of the original model Modes depicting thecarting of Rodin's Sarrow118891.

The anist creates a sculpted rnodei, generally made of plaster, clay, marble, Stone or wood

5 After cio~ing the mald around the clay model, hot wax is poured into the gap between the madel and the mold. This stage is crucial in producing a perfect reproductionof the initial sculpture.The rssult is a ciay model covered with wax, which is then hand-finishedto fidelity, incorporating the artist's signature, cast numberand afoundry seal.

Fireproof clay is carefully put n t o the Impression maklng asharpiy def~ned duplicate of the artist's ortg>nalmodel

The surface of thrs second cia" model is ~llghtly scr$ped away When this second model I S returned to the mald there is a gap between the model and the mald Ths gap is where the war w ~ lbe l poured The f~nal bionzewlll be of the same thickness as the gap that s created by the scraping

7
,\ ns$,.< r . < . ~ ,.t$< 3 ;>c., c,, c .,
igr. i
Y ' 1

10.11 lllc,le: FPPSU B nn tn- n , A !c s'; .,I :

i

:I" 031;. sea. .51'0 t r 4

'

r n.r tr T.
i sr,

6:

p prs n

. .

S~,O,~V :r I l i 0.0 1r.n l,lr,., :..:I

.",<,-J' 0-, o
,

"

.I
(<

. c:
<.!>

c
-I
'

I Ill n

I l r a C a l l C B C .PZ

< ?O,.r::

*. *)-..
t.'
,

A - -%?I, gr..?., .a!*, 1 ,?'or,' < :~:f!.~!:.ocrk > . ~ C . L & :)I :~~,:~>~ ,ts. , u , :tj,.. u.!: 8. 8, Ic c . i in C , i l c :c:!sa TPB r ~ s . . 7 % : j ?r ?rtmrni mo o r s - ? n : e, an,: nea,i: c ? . s r p l b d :.a*:o r e t >.I^ 1 o., 0 .: 0 . , r e r r o .: e i . ng 9 < n i x ,c t . , c C " I": I r.. r . $ b . , 11 : ,, 7 , ,,I,! b?<: ,e ,,.:,,,,,, c.,,, ,,<,lJ ACLI.I<: J , I I : I I . L I I I L I .> I . . , . I ' ; Cj: > B d & . C C O S *

Tne n . + s r l l e r ! mn n .,!rrn r vI!~.U IL n y.l l?,llp.,al.r6, u.&r I,:[? F C<:o~!ldr 2 LA : ? : . J 1 : ,LC :-c q. 7 cror'ze 31 cn,
:c:; c! L T O ore0
' .a
XIIF
3

3 a l e ' 01 C d T 0 11 ? G1018C1 V F 11P.i 100' nq A * cn ?>..%I : ~ r o . n p l ? r ,

(11,

OF'J'B Dl3^2e

;.. .',IS

U;(l

"S.

y, ' .. .

6 1 0
11

t t . ~ si?ce
. ,? ,

'

...
5:

eil i., .n,. n ? .An7." ~

".J

I.,,,'

,,.".':""

Wner me r-3; r g s I n snec -01 0.r" rl cx oes r ,>p :a I(. s:Idc.. c.1 !r,,,r<,.,i ., uc,: 111 .1 CI
: Y c ;

:,

,I:

: . n l i,y
S,V>G.>I

rn$c

ilJ\II

i,.lC trc (1 38d d1)0 ' 1 F T i ' 11

:I.ICLI,II

-1 5 2 ) F I
Or

-.

1i l a . ~ r~ tv

iwfr:'

L.JII>I 1. .

_I,

: i b FO

, P ~ ~ ? . Tne ; t~onlc
d r r l :e .F,-o: 3 1 1 q.11~; ?,i dl r.kl~' IZP'O ,:.I cr

. p:..re

. ! a , 1 1 At,, ,en?.s 0':1el r e p r ; l f ~ i , moan e': ne a? ' l e

n .I' 8, 2 0 . . 1 sci. '(0 :-c: ; .:. nr Ih.. o . ~ c ' ~ ~ L

1'CPP

s qn.;nou~

:,r~c~.!*..?
"1

-f!n- ndx

? i :I

w c n d e a ' e 2 rc 1?7~.er~ C I .

.-n i e ns: .: i c . 1I...

I!,,! "~<:'IZC

FIGURE 1-10 A manufacturing process represents a sequence of steps. Here are the steps in the lost-wax castlng process.

S ~ c n 1.2 o ~ Manufacturing and Production Systems

Metalforming Process for Automobile Fender
Sheet metal bend~ng iforming S~ngle draw punch and dle

17

Blank

I'

Cast billets of metal are passed through successive rollers to produce sheets of steel rolled stock.

The flat sheet metal is "formed" Into a fender, uslng sets of dles mounted on stands of large presses.

Metal shear~ng Sheet metal

Die

The fender 1 s cur out of the sheet metal In The last stage uslng shearlng processes.

Sheet metal shearing processes are l~ke sclssors cuttlng paper

Next, the sheet metal parts are welded into the body ofthe car.

FIGURE 1-11 The forming process used to make a fender for a car

temperature at which this material can recrystallize (i.e., grow new grain structure). Figure 1-11 shows the process by which the fender of a car is made using a series ofmetalforming processes. Machining or metal removalprocesses refer to the removal of certain selected areas from a part in order to obtain a desired shape or finish. Chips are formed by interaction of a cutting tool with the material beingmachined.Figure 1-12shows a chip being formed by a single-point cutting tool in a machine tool called a 1athe.The manufacturing engineer may be called upon to specify the cutting parameters such as cutting speed, feed, or depth of cut (DOC).The engineer may also have to select the cutting tools for the job. Cutting tools used to perform the basic turning on the lathe are shown in Figure 1-12. The cutting tools are mounted in machine tools, which provide the required movements of the tool with respect to the work (or vice versa) to accomplish the process desired. In recent years many new machining processes have been developed. The seven basic machining processes are shaping, drilling, turning, milling, sawing, broaching, and abrasive machining. With the exception of shaping, each of the basic processes has a chapter dedicated to it. Historically, eight basic types of machine tools

The Machining Process (tum~ng on a lathe)

Cutting

The workpiece is mounted in a machine tool (lathe) with a cutting tool.

The workpiece is rotated while the tool i s fed at some feed rate (inches per revolution]. The desired cutting speed V determines the rpm of the workpiece. This process is called turning.

The cutting tool interacts with the workpiece to form a chip by a shearing process. The tool shown here is an indexable carbide insert tool with a chipbreaking groove.

FIGURE 1-12 Single-po~nt

metalcutting process (turning) produces a chip while creating a new surface on the workpiece.

have been developed to accomplish the basic processes.These machine tools are called shapers (and planers), drill presses, lathes, boring machines, milling machines, saws, broaches, and grinders. Most of these machine tools are capable of performing more than one of the basic machining processes. Shortly after numerical control was invented, machining centers were developed that could combine many of the basic processes, plus other related processes, into a single machine tool with a single workpiece setup. Included with the machining processes are processes wherein metal is removed by chemical, electrical, electrochemical, or thermal sources. Generally speaking, these nontraditional processes have evolved to fill a specific need when conventional processes were too expensive or too slow when machining very hard materials. One of the first uses of a laser was to machine holes in ultra-high-strength metals. Lasers are being used today to drill tiny holes in turbine blades for jet engines. Because of its ability to produce eomponents with great precision and accuracy, metalcutting, using machine tools, is recognized as having great value-adding capability. In recent years a new family of processes has emerged called rapidprototyping. These additive-type processes produce first, or prototype, components directly from the software using specialized machines driven by computer-aided design packages. The prototypes can be field tested and modifications to the design quickly implemented. Early versions of these machines produced only nonmetalic components, but modem

SEcTroN 1.2 Manufacturing and Production Systems

19

machines can make metal parts. In contrast, the machining processes are recognized as having great value-adding capability, that is, the ability to produce components with great precision and accuracy. Perhaps the largest collection of processes, in terms of both diversity and quantity, are the joiningprocesses, which include the following: 1. Mechanical fastening 2. Soldering and brazing 3. Welding 4. Press, shrink, or snap fittings 5. Adhesive bonding 6. Assembly processes Many of these processes are often found in the assembly area of the plant. Figure 1-13 provides one example where all but welding are used in the sequence of operations to produce a computer. At the lowest level, microelectronic fabrication methods produce entire integrated circuits (ICs) of solid-state (no moving parts) components, with wiring and connections, on a single piece of semiconductor material, usually single-crystalline silicon. Arrays of ICs are produced on thin, round disks of semiconductor material called wafers. Once the semiconductor on the wafer has been fabricated, the finished wafer is cut up into individual ICs, or chips. Next, at Level 2,these chips are individually housed with connectors or leads making up "dies" that are placed into "packages" using adhesives.The packages provide protection from the elements and a connection between the die and another subassembly called the printed circuit board (PCB). At level 3, IC packages, along with other discrete components (e.g., resistors, capacitors, etc.), are soldered onto PCBs and then assembled with even larger circuits on PCBs.This is sometimes referred to as electronic assembly. Electronic packages at this level are called cards or printed wirina assemblies (PWAs). Next. series of cards are combined on a backuanel PCB. also known as a motherboard or simply a board. This level of packaging is sometimes referred to as card-on-board packaging. Ultimately, card-on-board assemblies are put into Silicon wafer,

Microelectronic

. . .

manufacturing

Level 1 Die

. -

or Integrated circuit (ICI

__------

on chip Level I
PCB

:assemblv

Level 5
FIGURE 1-13

Level 4

How an electronic product is made.

20

and Processes In Manufactur~ng CHAPTFR 1 lntroductlon to DeCarmo's Mater~als housings using mechanical fasteners and snap fittmg and finally integrated with power supphes and other electronic peripherals through the use of cables to produce final commercial products. Finishcngprocerces are yet another class of processes typically employed for cleaning, removlng burrs left by machining, or providmg protective andlor decorative surfaces on workpieces. Surface treatments include chemical and mechanical cleaning, deburring painting, platmg, buffing, galvanmng, and anodiz~ng. Heai treatment is the heating and cooling of a metal for the specific purpose of altering its metallurgical and mechanical properties. Because chang~ng and controlling these propert~es is so important in the processing and performance of metals,heat treatment is a very important manulacturing process. Each type of metal react3 differently to heat treatment Consequently, a designer bhould know not only how a selected metal metal w ~ lreact, l can be altered by heat ircatment but, equally important, how a ~elecfed favorably or unfavorably, to any heat~ng or cooltng that may be cncidentul to the manufacturing processes.

OTHER MANUFACTURING OPERATIONS In ddd~tlon to the processes already described, there are some other fundamentdl manufactur~ng operations that we must consider. Inspectton determines whether the desired objectives stated by the des~gner in the specifications have been achieved.This activ~ty provides feedback to design and manufacturing with regard to the process behavior. Ebsential to this mspection function are measurement activities. In teshng, a product is tried by actual function or operation or by subjection to external effects. Although a test is a form of inspection, it is often no1 viewed that way. In manufacturing,parts and matenals are inspected for conformance to the d~mens~onal and physical specifications,while testlug may simulate the environmental or usage demands to be made on a product after it is placed in aervice. Complex processes may require inany tests and inspection?. Test~ng includes life-cycle tests, destructive tests, nondestructive testmg to check for processing defects, wnd-tunnel tests, road tests, and overload tests. Trunsportation of goods in the factory is often referred to as matertnl handling or conveyance of the goods and refers to the transporting of unfinished goods (workin-process) in the plant and supplies to and from, between, and durmg manufactunng aperatlons. Loading, positioning, and unloading are also material-handling operations Transportation, by truck or train, is malerial handling between factorles Proper manufacturing system design and mechanization can reduce material handling in countless ways. Automatic material handling is a critical part of continuous automatic manufachandling Material hanturing.The word automairon is derived from automatic mater~al dhng, a fundamental operation done by people and by conveyors and loaders, often includes pos~tioning the workpiece wlthin the machine by indexing, shuttle bars, shdes, and clamps In recent years, wlre-guided automated guided vehicles (AGVs) and automatic storage and retrieval systems (ASIRSs) have been developed in an attempt to replace forklift trucks on the factory floor. Another form of material handling, the meehan~zed removal of waste (chips,trimming, and cutoffs), can be more difficult than handling the product. Chip removal must be done before a tangle of scrap chips damages tooling or creates defective workpieces Most texts on manufacturing processes do not mentlon packagmg, yet the packagmg is often the first thing the customer sees Also, packaging often maintains the product's quaIity between completion and use. (Packagmg 1s also used in eleclronics manutzdcturing to refer to placing microelectronic chips in containers for mountlug on circuit boards.) Packaging can also prepare the product for dehvery to the user. It varies from filltng ampules with ant~biot~cs to steel-strapping a l u m u m ingots into pallet~zed loads A product may require several packag~ng operations. For example,Hershey Kisses are (1)indiv~dually wrapped m foil, (2) plaeed in bags, (3) put Into boxes, and (4) placed in shipping cartons.

Weighing, filling,sealing, and labeling are packaging operations that are highly automated in many industries. When possible, the cartons or wrappings are formed from material on rolls in the packaging machine. Packaging is a specialty combining elements of product design (styling), material handling, and quality control. Some packages cost more than their contents, for example, cosmetids and razor blades. During storage,nothing happens intentionally to the product or part except the passage of time. Part or product deterioration on the shelf is called sheIflife, meaning that items can rust, age, rot, spoil, embrittle, corrode, creep, and otherwise change in state or structure, while supposedly nothing is happening to them. Storage is detrimental, wasting the company's time and money.The best strategy is to keep the product moving with as little storage as possible. Storage duringprocessing must be eliminated, not automated or computerized. Companies should avoid investing heavily in large automated systems that do not alter the bottom line. Have the outputs improved with respect to the inputs, or has storage simply increased the costs (indirectly) without improving either the quality or the throughput time? By not storing a product, the company avoids having to (1) remember where the product is stored, (2) retrieve it, (3) worry about its deteriorating, or (4) pay storage costs. Storage is the biggest waste of all and should be eliminated at every opportunity.

UNDERSTAND YOUR PROCESS TECHNOLOGY Understanding the process technology of the company is very important for everyone in the company. Manufacturing technology affects th8 design of the product and the manufacturing system,the way in which the manufacturing system can be controlled, the types of people employed, and the materials that can be processed.Table 1-4 outlines the factors that characterize a process technology. Take a process you are familiar with and think about these factors. One valid criticism of American companies is that their managers seem to have an aversion L o understanding their companies' , Mechanics (statics and dynamics of the pmccss) manufacturing technologies. Failure to understand How does the procmss work? the company business (i.e., its fundamental process What are the process mechanics (statics, dynamics,friction)l What physically happens, and what makes it happen? (Undrrstand the physia.) technology) can lead to the failure of the company. Economics or costs The way to overcome technological aversion is What are the tooling costs, the cngilleerinp costs? :I to run the process and study the technology. Only Which costs are short term. which long tcrm? ! someone who has run a drill press can understand ; What are the setup costs? the sensitive relationship between feed rate and drill Timespans torque and thrust. All processes have these "know., How long docs it lake to set up the prncess initially? What is the throughput time? how" features. Those who run the processes must How can these times beshortened? be part of the decision making for the factory.The Haw long does it lake lo run a par1 once it isset up (cycle time)? ; Whatprocess parameters affect the cycle time? CEO who takes a vacation working on the plant ., Constraints floor and learning the processes will be well on the What are the proccss limits? way to being the head of a successful company.

What cannot he done? What constrains this process (sizes, speeds,forces, volurnrs,power,cost)? What is very hard to do within an acceplahle tirndcost f m e ? , Uncertainties and proeessrelinbility : What can go wrong? How can this machine fail? What do people worry about with this process? Is this a ieliable,stable proccss? Skills What operator skills are critical? What is not done automatically? ! HW long docs it take to learn to do this process? Flexibility Can this process be adapted easily for new parts of a new design or material7 How does the process react to changes in part design and demand? What changcs are easy ta do? Process capability ,' What are thc accuracy and precision of the process? What tolerances does the proce~s meet? (What is the process capability?) How repeatable are those tolerances?

'(

PRODUCT LIFE CYCLE AND LIFE-CYCLE COST Manufacturing systems are dynamic and change with time. There is a general, traditional relationship between a product:^ life cycle and the kind of manufacturing system used to make the product. Figure 1-14 simplifies the life cycle into these steps:
1. Start~lp. New product or new company, low volume, small company. 2. Rupidgrowth. Products become standardized and volume increases rapidly. Company's ability to meet demand stresses its capacity. 3. Maturation. Standard designs emerge. Process development is very important.

Time

,
&-gn

-..: ,-..---

I
i

MrrnUfruswIng

Job show

Startup - ....7--

Rapid growth

Prgduckion bb shop
.

i..: i
.

with some . * , w

Commodity or decline - - - -. .---A Product~on job shup . Mote ilow 1 wth some few lines mass-produce and assembly lines
Maturation
.

I -

.-7--.

Productvariety:

Greatvariety; Increasing standardization; Emergence of a dominant High standardiproduct less variety standard design zation innovation "Commodity" great characteristics Fallout and consolidation Few large companies "Survivors" become commodities Price with consistent quality
Medium
High

Industry structure: Many small competitors Form of competition:

Product Product quality, cost, characteristics and availability Medium to high Medium

Price and quality with reliability

High

Process innovation: Low Automation: Low

Medium to high

FIGURE 1-14 Product life-cycle costs change with the classic manufacturing system designs.

4. Commodity. Long-life, standard-of-the-industry type of product or 5. Decline. Product is slowly replaced by improved products.
The maturation of a product in the marketplace generally leads to fewer competitors, with competition based more on price and on-time delivery than on unique product features. As the competitive focus shifts during the different stages of the product life cycle, the requirements placed on manufacturing--cost, quality,flexibiiity,and delivery dependability-also change.The stage of the product life cycle affects the product design stability, the length of the product development cycle, the frequency of engineering change orders, and the commonality of components, all of which have implications for manufacturing process technology. During the design phase of the product, much of the cost of manufacturing and assembly is d e ~ e r m i n e d ~ ~ s e mof b lthe y product is inherently integrative as it focuses on pairs and groups of parts. It is crucial to achieve this integration during the design phase because about 70% of the life-cycle cost of a product is determined when it is designed. Design choices determine materials, fabrication methods, assembly methods, and, to a lesser degree, material-handling options, inspection techniques, and other aspects of the production system. Manufacturing engineers and internal customers can influence only a small part of the overall cost if they are presented with a finished design that does not reflect their

SECTION 1.2 Manufacturing and Production Systems

23

concerns.Therefore all aspects of production should be included if product designs are to result in real functional integration. Life-cycle costs include the costs of all the materials, manufacture, use, repair, and disposal of a product. Early design decisions determine about 60% of the cost, and all activities up to the start of full-scale development determine about 75%. Later decisions can make only minor changes to the ultimate total unless the design of the manufacturing system is changed. In short, the concept of product life-c)'cle provides a framework for thinking about the product's evolution through time and the kind of market segments that are likely to develop at various times. Analysis of life-cycle costs shows that the design of the manufacturing system determines the cost per unit, which generally decreases over time with process improvements and increased volumes. For additional discussion on reliability and maintainability of manufacturing equipment, see SAE publication M-110.2. The linked-cell manufacturing system design discussed in Chapter 39 enables companies to decrease cost per unit significantly while maintaining flexibility and making smooth transitions from low-volume to high-volume manufacturing. Low-cost mannfacturing does not just happen.There is a close, interdependent relationship among the design of a product, the selection of materials, the selection of processes and equipment, the design of the processes, and tooling selection and design. Each of these steps must be carefully considered, planned, and coordinated before manufacturing starts.This lead time, particularly for complicated products,may take months, or even years, and the expenditure of large amounts of money may be involved.Typically, the lead time for a completely new model of an automobile or a modern aircraft may be two to five years. Some of the steps involved in getting the product from the original idea stage to daily manufacturing are discussed in Chapter 9 in more detail.The steps are closely related to each other. For example, the design of the tooling is dependent on the design of the parts to be produced. It is often possible to simplify the tooling if certain changes are made in the design of the parts or the design of the manufacturing systems. Similarly, the material selection will affect the design of the tooling or the processes selected. Can the design be altered so that it can be produced with tooling already on hand and thus avoid the purchase of new equipment? Close coordination of all the various phases of design and manufacture is essential if economy is to result. With the advent of computers and computer-controlled machines, the integrarion of the design function and the manufacturing function through the computer is a reality. p i s is usually called CADICAM (computed-aided designlcomputer-aided manufactnring).The key is a common database from which detailed drawings can be made for the designer and the manufacturer and from which programs can be generated to make all the tooling. In addition, extensive computer-aided testing and inspection (CATI) of the manufactured parts is taking place.There is no doubt that this trend will continue at ever-acceleratingrates as computers become cheaper and smarter, but at this time, the computers necessary to accomplish complete computer-integrated manufacturing (CIM) are expensive and the software very complex. Implementing CIM requires a lot of manpower as well.

MANUFACTURING SYSTEM DESIGN

When designing a manufacturing system, two customers must be taken into consideration: the external customer who buys the product and the Internal customer who makes the product. The external customer a hkely to be global and demand greater variety with superlor quality and reliabllity.The internal customer is often empowered to make crit~cal decis~ons about how to make the products The Toyota Motor Company 1s makm n g vehmcles in 25 countries.Their truck plant m Indiana has the capacity to make 150,000 vehicles per year (creating 2300 new jobs), using the Toyota Production System (TPS) An appreciation of the complexity of the manufacturing system design problem is shown in Figure 1-15, where the choices in system design are reflected against the number of different products, or parts being made in the system, often called variety. Clearly, there are many choices regarding which method (or system) to use to make the goods

CHAWLK 1 Introduction to DeGarmo's Materials and Processes in Manufacturing

This part variety-production rate matrix shows examples of particular manufacturing system designs. This matrix was developed by Black based on real factory data. Notice there is a large amount of overlap in the middle of the matrix, so the manufacturing engineer has many choices regarding which method or system to use to make the goods. This book will show the connection between the process and the manufacturing system used to produce the products, turning raw materials into finished goods.

Variety of parts per system

FIGURE 1-15 Differentmanufacturing system designs produce goods at differentproduction rates.

A manufacturer never really knows how large or diverse a market will be. If a diverse and specialized market emerges, a company with a focused flow-line system may be too inflexible to meet the varying demand. If a large hut homogeneous market develops, a manufacturer with a flexible system may find production costs too high and the flexibility unexploitahle. Another general relationship between manufacturing system designs and production volumes is shown in Figure 1-16.

NEW MANUFACTURING SYSTEMS

The manufacturing process technology described herein is availablc worldwide. Many countries have about the same level of process development when it comes to manufacturing technology. Much of the technology existing in the world today was developed in the United States, Germany, France, and Japan. More recently Taiwan, Korea, and China have been making great inroads into American markets, particularly in the automotive and electronics industries. What many people have lailed to recognize is that many companies have developed and promoted a totally different kind of manufacturing production system.This new system, called theToyota Production System, will take its place with the American Armory System and the Ford System for mass production. This new system, developed by theToyota Motor Company, is known worldwide as the lean manufacturing system. Many American companies have successfully adopted some version of ihe Toyota system. The experience of dozens of these companies is amalgamated into 10 steps, which, if followed, can make any company a factory with a future. For the lean production to work, 100%-good units flow rhythmically to subsequent processes without interruption. In order to accomplish this, an integrated quality control (IQC) program has to be developed. The responsibility for quality has been given to manufacturing. All the employees are inspectors and are empowered to make it right the first time. There is a companywide attitude toward coilstant quality improvement. Make quality easy to see, stop the line when something goes wrong, and inspect things 100% if necessary to prevent defects from occurring. The results of this system are astonishing in terms of quality, low cost, and on-time delivery of goods to the customer. The most important factor in economical and successful manufacturing is the manner in which the resources-labor, materials, and capital-are organized and managed so as to provide effective coordination, responsibility, and control. Part of the success of

Review Questions

25

Disconnected line Interrupted flow shop Batch processing Job shop ad hoc flow
Jobshop

I efficient

I
Commod~ty Few types H~gh volume High volume

more flexible
Multiple

The figure shows in a general way the relationship between manufacturing systems and production volumes. The upper left represents systems with low flexibilitybut high efficiency compared to the lower right, where volumes are low a n d so is efficiency. Where a particular company lies in this matrix is determined by many forces, not all of which are controllable.The job of manufacturing and industrial engineers is to design a n d implement a system which can achieve low unit cost, superior quality, with on-time delivery in a flexible way. Very low volume
r

pr~ducts Low volume

FIGURE 1-16 This f~gure shows in a general way the relationsh~pbetween manufacturlng systems and production volumes.

lean production can be attributed t o a different management approach. T h ~ approach s 1s charactenzed by a holistic attitude toward people. T h e real secret of successful manufacturing hes in designing a manufacturing system in which everyone who works in the system understands how the system works, how goods are controlled, with the declsion maklng placed at the correct 1evel.The engineers also must possess a broad fundamental knowledge of design, metallurgy, processing, economcs, accountmg, and human relations. In the manufactunng game,low-cost mass production 1s the result of teamwork w ~ t h i n an integrated manuactunng/production system. This is the key t o p r o d u c ~ n g superior q u a l ~ t y a t less cost with on-tlme delivery.

Bl Key Words
Ssembllng estmg consumer goods continuous process desrgn engineer fabncatmg fihngpg flow shop fomlng group technology heat treatment inspection lob job shop
joining

lean product~on lean shop lost-wax castlng I~nked-cell manufactunng system

machlue tool mach~mg manufacturlng cost manufacturrng englneer manufacturlng process manufacturlng system moldlug numerical control
operation

product hfe cycle product~on system project shop shearlng statlon toohng tools treatments weld~ng

producer goods

III Review Questions

1. What role does manufacturing play relative to the standard of living of a country? 2. Aren't all goods really consumer goods, depending on how you define the customer? Discuss. 3. Give examples of a job shop, flow shop, and project shop. 4. How does a system differ from a process? From a machine tool? From a job? From an operation? 5. Is a cutting tool the same thing as a machine tool? 6. What are the major classifications of basic manufacturing processes?
7. Could casting be used to produce a complex-shaped part to be made from a hard-to-machine metal? How else could the part be made? 8. In the lost-wax casting process, what happens to the roam? 9. In making a gold medal, what do we mean by a"re1ief image'' cut into the die? 10. How is a railroad station like a station on an assembly line? 11. Since no work is being done on a part when it is in storage, it does not cost you anything.True or false? Explain. 12. What forming processes are used to make a paper clip?

26

CHAPTER 1

Introduction to DeGarmo's Materials and Processes in Manufactur~ng

25. It has been said that low-cost products are more likely to be

13. We can analogize your university to a manufacturing system

14.

15. 16. 17. 18.

19 20. 21. 22.

23. 24.

that produces graduates Assuming that it takes four years to get a college degree and that each course really adds value to the student's knowledge base, what percentage of the four years is "value adding" (percentage of time in class plus two hours of preparation for each hour in class)? It is acknowledged that chip-type machining is basically an inefficient process.Yet it is probably used more than any other to produce desired shapes. Why? Compare Figure 1-1 and Figure l-l4.What are the stages of the product life cycle for an audiocassette tape? In a modern safety razor with three or four blades that sells for $1, what do you think the cost of the blades might be? List three purposes of packaging operations. Assembly is defined as "the putting together of all the different parts to make a complete machine." Think of (and describe) an assembly process. Is making a club sandwich an assembly process? What about carving a turkey? Is this an assembly process? What are the physical elements m a manufacturmgsystem? In the prodnct~on system, who usually figures out how to make the product? In Figure 1-7, what do the lines connecting the processes represent? Characterize the process of squeezing toothpaste from a tube (extrusion of toothpaste) using Table 1-4 as a guideline. See the index for help on extrusion. What are the major process steps in the assembly of an automobile? What difficulties would result if production planning and scheduling were omitted from the procedure outlined in Chapter 9 for making a product in a job shop?

26.

27.

28.
29.

30.

31. 32. 33 34. 35.

36.

more carefully designed than high-priced items. D o you think this is true? Why or why not? Proprietary processes are closely held or guarded company secrets. The chemical makeup of a lubricant for an extrusion process is a good example. Give another example of aproprietary process. If the rolls for the cold-rolling mill that produces the sheet metal used in your car cost $300,000 to $400,000,how is it that your car can still cost less than $20,000? Make a list of service production systems, giving an example of each. What is the fundamental difference between an SPS and an MPS? In the process of buying a calf, raising it to a cow, and disassembling it into "cuts" of meat for sale, where is the "value added? What kind of process is powder metallurgy:casting or forming? In view of Figure 1-2, who really determines the selling - price per unlt? What costs make up manufactur~ng cost (sometimes called factory cost)? What are major phases of a product life cycle? How many different manufacturing systems might be used to make a component with annual projected sales of 16,000 parts per year with 10 to 12 different models (varieties)? In general, as the annual volume for a product increases, the unit cost decreases. Explain.

Problems
1. TheToyota truck plant in Indiana produces 150,000trucks per year. The plant runs one eight-hour shift, 300 days per year, and makes 5W trucks per day. About 1300 people work on the final assembly line. Each car has about 20 labor hours per car in it. a. Assuming the truck sells for $16,000 and workers earn $30 per hour in wages and beuefits,what percentage of the cost of the truck is in direct labor? b. What is the production rate of the final assembly line? 2. Suppose you wanted to redesign a stapler to have fewer components. (You should be able to find a stapler at a local discount store.) How much did it cost? How mauy parts does it
have? Make up a "new parts" list and indicate which parts would have to be redesigned and which parts would be elimiuated.Estimate the manufacturing cost of the stapler assuming that manufacturing costs are 40% of the sellingprice.What are the disadvantages of your new stapler design versus the old stapler? 3. A company is considering making automobile bumpers from aluminum instead of from steel. List some of the factors it would have to consider in arriving at its decision. 4. Many companies are critically examining the relationship of product design to manufacturing and assembly. Why do they call this concurrent engineering?

case study

27

anufacturing engineering is that englneerlng function charged wlth the responsibility of Interpreting product design in terms of manufacturing requlrements and process capability. Specif~cally. the manufacturlng englneer

* Determne how the product isto be made in terms of

specific manufaclurlngprocesses. Des~gn workholdlng and work transporting toollng or Select the tools (includingthe tool materials) that will machlne or form the work materials. Sele~t, deslgn, and specify devices and instruments which that which hasbeen manufactured to determlne its quallty. Design and evaluate the performance of the manufacPerform all them f u ~ i o n s (and many more) relafedto the actual making of the product at the most reasonable cost per unit w~thout sacrifice of the functional requlrements or the users' servlce llfe. There's no great glory in being a great manufacturing engineer (MfE) Ifyou want to be a manufacturingenglneer, you had better be ready to got your hands dirty. Of course, there are excepttons.There have been some very famous manufacturing engineers. John Wllkinson of Bersham England built a bonng mill in 1775 to bore the cast Iron cylinders for James Watt's steam englne. How good was thls machine? Eli Whitney was said to have Invented the cotton gin, a machine to separate seeds from cotton. His machine was patented but was so $~mple, anyone could make one. He was cedlted with "lnterchangeabillty" - but we know Thomas Jefferson observed lnterchangeability in France In 1785 and probably the French gunsmlth LeBlanc 1 s the real inventor here. Jefferson tried to bnng the Idea to Americaand Whltney certainly did He took 10 muskets to congress, disassembled them, and scattered the pieces. Interchangeable parts permitted them to be reassembled. He was given a contract for 2,000 guns to be made in 2 years. But whatrs the rest Joe Brown started a business in Rhode Island In 1833 mawng lathes and small tools as well as timep~eces (watchmaker). Luaan Sharp jolned the company in 1848 and developed a pocket sheet metal gage in 1877 a 1 Inch micrometer and in 1862 the universal mllllng

voyage. He saved his money and had models of a gun built in Hartford by Anson Chase for which he got a patent. He set up a factory In New Jersey - but he could not sell hls guns to the army - too complicated. Hesoidto theTexas Rangers and the Florida Frontiersmen but he had to close the piant. In 1846 the Mexican war broke out. General Zachary Taylor and Captarn Sam Walters wanted to buy guns. Colt had none but accepted orders for 1000 guns and constructed a m&el (Walker Colt) and arranged to have them made at Whitney's (now 40 years old) plant in Whitneyvllle. Here he learned about mass production methods. In 1848 he rented a Plant Hartford and the Cob legend spread. In 1853 he had bullt one of the warlds largest arms plant In Connecticut which had 1400 machme tools. Colt helped start the careers of 0 E K Root mechanic and super~ntendent - paylng him a salary of $25,000 in the 1800's. Abolished hand work -jigs and fixtures. 0 Francls Pratt and Amos Whitney' -famous machine tool bu~iders o Wlll~am Gleason -gear manufacturer o E P.Bullard Invented the Muit-An-Matlc Multiple sp~ndle machine which cut the time to make a over 1 mlnute. flywheel fmm 18 mmutes to sl~ghtly Sold this to Ford. 0 Christopher Sponer. a E. J. Klngsbury Invented a drilllng machine to drill hales through toy wheel hubs that had a sprlng loaded cam which enables the head to sense the condltlon of the casting and modify feed rate automatically Now here are some more names from the past of famous and not so famous manufacturing. mechanical, and Industrialengineers. Relate them to the development of manufactur~ng processes or manufacturing system designs. UiWhltney Henry Ford Charles Sorenson Sam Colt John Parsons Eijr Toyoda Eiisha Root John Hall Thomas Blanchard Fred Taylor Talichi Ohno Ambrose Swasey

. . . .
.

SEcno~ 2.1

Introduction

29

capabilities, and their limitations can one determine if the resulting data are applicable to a particular problem.

METALLIC A N D NONMETALLIC MATERIALS While engineering materials are often grouped as metals, ceramics, polymers, and composites, a simpler distinction might be to separate them into metallic and nonmetallic. The common metallic materials include iron, copper, aluminum, magnesium, nickel, titanium, lead, tin, and zinc as well as the alloys of these metals, such as steel, brass, and bronze. They possess the metallic properties of luster, high thermal conductivity, and high electrical conductivity; they are relatively ductile; and some have good magnetic properties. Some common nonmetals are wood, brick, concrete, glass, rubber, and plastics.Their properties vary widely, but they generally tend to be weaker, less ductile, and less dense than the metals, and to have poor electrical and thermal conductivities. Although metals have traditionally been the more important of the two groups,the nonmetallic materials have become increasingly important in modern manufacturing. Advanced ceramics, composite materials, and engineered plastics have emerged in a number of applications. In many cases, metals and nonmetals are viewed as competing materials, with selection being based on how well each is capable of providing the required properties. Where both perform adequately, total cost often becomes the deciding factor, where tolal cost includes both the cost of the material and the cost of fabricating the desired component. Factors such as product lifetime, environmental impact, energy requirements, and recyclability are also considered. PHYSICAL A N D MECHANICAL PROPERTIES A common means of distinguishing one material from another is through their physical properties. These include such features as density (weight); mclting point; optical properties (transparency, opaqueness, or color); the thermal properties of specificheat, coefficient of thermal expansion, and thermal conductivity: electrical conductivity; and magnetic properties. In some cases, physical properties are of prime importance when srlecting a material, and scveral will be discussed in more detail near the end of this chapter. More often, however, material selection is dominated by the properties that describe how a material responds to applied loads or forces. These mechanicalproperties are usually determined by subjecting prepared specimens to standard test conditions. When using test results, however, it is important lo remember that they apply only to the specific conditions that were employed. The actual service conditions of engineered products rarely duplicate the conditions of laboratory testing, so considerable caution should be exercised when applying test results. STRESS AND STRAIN When a force or load is applied to a material, it deforms or distorts (becomes strained),and internal reactive lorces (stresses) are transmitted through the solid. For example,if a weight, W, is suspended from a bar of uniform cross section and length L,as inFigure 2-2, the bar will elongate by an amount AL. For a given weight, the magnitude of the elongation, A L, depends on the original length of the bar. The amount of elongation per unit length, expressed as e = A l J L , is called the unit stmin.Although the ratio is that of a length to another length and is therefore dimensionless, strain is usually expressed in terms of millimeters per meter, inches per inch, or simply as a percentage. Application of the force also produces reactive stresses, which serve to transmit the load through the bar and on to its supports. Stress is defined as the force or load being transmitted divided by the cross-sectional area transmitting the load. Thus, in Figure 2-2, the stress is S = W I A ,whereA is the cross-sectional area of the supporting bar. Stress is normally expressed in megapascals (in SI units, where a pascal is 1 newton per square meter) or pounds per square inch (in the English system). In Figure 2-2, the weight tends to stretch or lengthen the bar, so the strain is known as a tensile strain and the stress as a tensile stress. Other types of loadings produce other types of stresses and strains (Figure 2-3). Compressive forces tend to shorten the material and produce compressive stresses and strains. Shear .stresse$ and strains result when two forces acting on a body are offset with respect to one another.

FIGURE 2-2 Tenslon loading

nd the resultant elongation.

30

CHAPTER 2 Properties of Ailaterials

T
L

f
AL

Tension

Compression

Shear

FIGURE 2-3 Examples of tens~on, cornpresslon, and shear load~ng, and their response.

When the forces that are applied to a matenal are constant, or nearly so, they are said to be static. S i static loadings are observed in many applications,it is important to characterize the behavior of materials under these conditions. For design engineers, the strength of a material may be of primary concern, along with the amount of elastic stretch~ng or deflection that may be experienced while under load. Manufacturing engineers, looking to shape products, may be more concerned with the abihty to mechanically deform the material without fracture. As a result, a number of standardized tests have been developed to evaluate the staticproperhes of engineering materials.Test results can be used to determine if a given material or batch of material has the necessary properties to meet specified requirements. Other tests provide the materials characterization base used for material selection. In all cases, it is important to determine that the service conditions are indeed similar to those of testing. Even when the service conditions differ, the results of standard tests may be helpful in qualitatively rating and comparing various materials.

TENSILE TEST The most common of the static tests is the uniaxial tensile test.The test begins with the preparation of a standard specimen with prescribed geometry, like the round and flat specimens described in Figure 2-4. The standard specimens ensure meaningful and re-

,;;f

FIGURE 2 4 Two common

types of standard tensile test specimens: (a) round; (b) flat. Dimensions are in inches, with millimeters in parentheses.

SECTION 2.2 Static Properties

31

FIGURE 2-5 (a) Hydraulic universal (tenslon and compresston) testing machlne; (b) schematic of the load frame showlna how u~ward motton of the darkened voke can Droduce tension or corn~ression with resuect to the statlon6 (whlie) crosspiece. (Courtesy of Sate;Systerns, jnc, Crow Oty, PA.)

producible results, and are designed to produce uniform uniaxial tension in the central portion of the specimen while ensuring reduced stresses in the enlarged ends or shoulders that are gripped.
Strength Properties. The standard specimen is then loaded in tension in a testing machine like the one shown in Figure 2-5. A force or load, W, is applied and measured by the testing machine, while the elongation or stretch ( A L) of a specified length (gage l e n g t h ) is simultaneously monitored. A plot of the coordinated load-elongation data produces a curve similar to that of Figure 2-6. Since the loads will differ for differentsized specimens and the amount of elongation will vary with different gage lengths, it is important to remove these geometric or size effects if we are to produce data that are characteristic of a given material, not a particular specimen. If the load is divided by the

MPa

lo3 Psi
70

60

Upper yleld
polnt

Breaking strength

Lower vield Yield-point elongation

FIGURE 2-6 Engineering
0

.stress-strain diagram for a lowcarbon steel.

I 5

I 10

15

I 20

0
25

Strain-% (inchedinch x l o ' )

II
1,

IU.

A ?

I
I

' :
;
/ / / /
!

8 6
2

/ /

Strain - % IinJin.)
FIGURE 2-7 Stress-strain niagram for a material not n d v l n q a well-defned yi?,a po~nt,-show~ng the offset method for determ~n~ng y~eld strength. S, 1s the 0.1% offset s the 0.2% yield strength; S, 1 offset y~eld strength.

ortginal cross-sectional area, A,, and the elongation is div~ded by the original gage length, L,, the size effects are eliminated and the resulting plot becomes known as an engineermg mess-engtneerlng strarn curve (see Figure 2-6).Thls is simply a load-elongation plot with the scales of both axes modified to remove the effects of specimen size. In Figure 2-6 it can be noted that the initial response is linear Up to a certain point, the stress and strain are directly proportional to one another.The stress at wh~ch this proportionality ceases is known as theproportfional limit. Below this value, the material obeys Hookek law,which states that the strainis directly proportional to the stress. The proportionality constant, or ratio of stress to strain, 1s known as Young's modulus or the modulus of ehtictty This 1s aninherent property of a given material' andis of considerable engineenng importance. As a measure of saffness, it indicates the ability of a material to resist deflection or stretching when loaded and is commonly designated by the symbol E. Up to a certaln stress, if the load is removed, the speclmen will return to its original length The response is elastic or recoverable, like the stretching and relaxat~on of a rubber band. The uppermost stress for which this behavior is observed is known as the elmtrc limit. For most materials the elastic limit and propohonal limit are almost identical, with the elastic limit being slightly higher. Neither quantity should be assigned great engmeering sigmficance, however, because the determmed values are often dependent on the sensitivity and prec~sion of the test equipment. The amount of energy that a matenal can absorb while in the e l a s k range is called the resrl~ence.The area under a load-elongation curve is the product of a force and a distance, and is therefore a measure of the energy a b w b e d by the specimen. If the area is dctermned up to the elasticlimit, the absorbed energy will be elastic (or potential) energy and is regamed when the speclmen is unloaded. If we perform the same calculabon on an engineering stress-engineering strain diagram, the area beneath the elastic region corresponds to an energy per unit volume and is known as the modulus ofresthence. Elongation beyond the elasticlimit becomes unrecoverable and is known asplastlc deformation. When the load is removed, only the elastic stretching will be recovered, and the specimen will retain a permanent change m shape. For most components, the onset of plastic flow represents failure, since the part dimensions will now be outside of allowable tolerances. In manufactnnng processes where plastic deformation m used to produce a deslred shape, the applied stresses must be sufficiently above the elasticlimit to induce the required amount of plastic flow. Permanent deformation, therefore, may be either desirable or undesirable, and it is Important to determine the conhtions where elastic behavior transitions to plastic flow. Whenever the elastic limit is exceeded, increases in stram no longer require proportionate increases in stress. For some materials, a stress value may be reached where adhtional strain occurs w~thout any further increase in stress.This stress is known as the yreld point, or yield-poinf sfsess. For low-carbon steels, with curves like that in Figure 2-6. two distinct points are sign~ficant. The highest stress preceding extensive strain a known as the upper yield poiar, and the lower, relatively constant, "run-out" value is known as the lower yieldpoint. The lower value is the one that usually appears in tabulated data. Mast materials, however, do not have a well-defined yield pomt and exhibit stress-strain curves more like that shown in Figure 2-7. For these materials, the elasticto-plastic transition is not d~stinct, and detection of plastic deformation would be dependent upon machine sensitivity.To solve this dilemma, we elect to define a useful and easily detetmlnedproperty known as the offset yieldstrength. Offset yield strength does nor descnbe the onset of plastic deformation but instead defies the stress required to produce a given, but tolerable, amount of permanent strak. By setting this strain, or. "offset," to 0.2% (a common value), we can determine the stress required to plastically

SECTION 2.2

Static Properties

33

deform a 1-inch length to a final length of 1.002 inches (a 0.2% strain). If the applied stresses are then kept below the 0.2% offset yield strength of the material, the user can be guaranteed that any resulting plastic deformation will be less than 0.2% of the original dimension. While 0.2% is a common offset for many mechanical products, applications that cannot tolerate that amount of deformation may specify offset values of 0.1% or even 0.02%. Offset yield strength is determined by drawing a line parallel to the elastic line, but displaced by the offset strain, and reporting the stress where the constructed line intersects the actual stressstrain curve. Figure 2-7 shows the determination of both 0.1% offset and 0.2% offset values, S , and S2, respectively.The intersection values are reproducible and independent of equipment sensitivity. Offset yield values are meaningless unless they are reported in conjunction with the amount of offset strain used in their determination.The 0.2% value is most common and is generally assumed unless another number is specified. As shown in Figure 2-6, the load (or engineering stress) required to produce additional plastic deformation continues to increase. The load that a material or specimen can bear (load-bearing ability) can be computed by multiplying the material strength times its cross-sectional area. During tensile deformation,the specimen is getting longer. The cross-sectional area is decreasing, but the load-bearing ability of the specimen continues to increase! For this to occur, the material must be getting stronger. The mechanism for this phenomenon will be discussed in Chapter 3, where we will learn that the strength of a metal continues to increase with increased deformation. During the plastic deformation portion of a tensile test, the weakest location of the specimen undergoes deformation and becomes stronger. Since it is no longer the weakest location, another location assumes that status and deforms. As a consequence, the specimen strengthens uniformly and maintains its original cylindrical or rectangular geometry. As plastic deformation progresses, however, the additional increments of strength decrease in magnitude,and a point is reached where the decrease in area cancels or dominates the increase in strength. When this occurs, the load-bearing ability peaks, and the force required to continue straining the specimen begins to decrease, as seen in Figure 2-6.1Ple stress at which the load-bearing ability peaks is known as the ultimate strength, tensile strength, or ultimate tensile strength of the material. The weakest location in the test specimen at that time continues to be the weakest location by virtue of the decrease in area, and further deformation becomes localized. This localized reduction in cross-sectional area, known as necking, is shown in Figure 2-8. If the straining is continued, necking becomes intensified and the tensile specimen will ultimately fracture.The stress at which fracture occurs is known as the breaking strength or fruclure strength. For ductile materials, necking precedes fracture, and the breaking strength is less than the ultimate tensile strength. For a brittle material, fracture usually terminates the stress-strain curve before necking, and possibly before the onset of plastic flow.
Ductility and Brittleness. When evaluating the suitability of a material for certain manufacturing processes or its appropriateness for a given application, the amount of plasticity that precedes fracture, or the ductility, can often be a significant property. For metal deformation processes, the greater the ductility, the more a material can be deformed without fracture. Ductility also plays a key role in toughness, a property that will be described shortly.

..FIGURE 2-8 A standard 0.505-

.'

...

in.-diameter tensile specimen showing a necked region developed prior to failure.

34

CHAPTER 2 Properties of Mater~als

~ ' .l. .l . l. .l . . l .l l ; ~
8"

b b b b b b b a

-------o q q 0 9 q o
(a)

FIGURE 2 3 Final elongation in various segments of a tensile test soecirnen: (a) oriainal geon;etry; (b) ;h&e ifter fracture.

: w a e : - h $ h . :

? -

One of the simplest ways to evaluate ductility is to determine the percent elongution of a tensile test specimen at the time of fracture. A s shown in Figure 2-9, ductile materials do not elongate uniformly when loaded beyond necking. If the percent change of the entire 8-in. gage length is computed, the elongation is 31%. However, if only the center 2-111. segment is considered, the elongation of that portion is 60%. A vahd comparison of material behavior, therefore, requires similar specimens wlth the same standard gage length. In many cases,material "failure"is defined as the onset of localized deformation or necklng. Consider a sheet of metal being formed Into an automobile body panel If we are to ensure uniform strength and corrosion resistance in the final panel, the operation must be performed in such a way as to ma~ntain uniform sheet thickness For this application, a more meaningful measure of mater~al ductility would be the uniform elongation or the percent elongation prior to the onset of necking.This value can be determined by constructing a line parallel to the elastic portion of the diagram, passing through the point of h~ghest force or stress.The intercept where the line crosses the strain axis denotes the available uniform elongation Since the add~tional deformation that occurs after necking is not considered, uniform elongation is always less than the total elongation at fracture (the generally reported elongation value). Another measure of ductility is the percent reduct~on m urea that occurs in the can be computed as necked region of the specrmen.Th~s

R.A. =

A0

- Af x 100%

where A, is the ong~nal cross-sectional area and A,is the smallest area in the necked region. Percent reduction in area, therefore, canrange from 0% (for a brittle glass specimen that breaks with no change in area) to 100% (for extremely plastic soft bubble gum that pinches down to a point before fracture). When materials fail mth little or no ductihty, they are said to be bnttle. Brittleness. however, is simply the lack of ductility and should not be confused with a lack of strength. Strong materials can be brittle, and brittle materials can he strong.
Toughness. Toughness,or modulus of toughness,is the aork per unit volume required to fracture a material.The tensile test can provide one measure of this property,slnce toughness corresponds to the total area under the stress-strain curve from test initiation to fracture, and thereby encompasses both strength and ductility. Caution should be exercised when using toughness data, however, because the work or energy needed to fracture can vary markedly mth different conditions of testing. Variations in the temperalure or the speed of loading can significantly alter both the stress-strain curve and the toughness

In most cases, toughness is associated wlth unpack or shock load~ngs, and the values obtained from high-speed (dynamic) impact tests often fail to correlate with those obtained from the relahvely slow-speed (static) tensile test.
True Stress-True Strain Curves. The stress-stram curve in Figure 2-6 is a plot of engmeering stress, S, versus engrneenng stram, e, where S is computed as the applied load divlded by the original cross-sectional area and c 1s the elongation, AL, divided by the orignal gage length, L,. As the test progresses, the cross section of the test specimen changes continually,frrst in a uniform manner and then nonuniformly after necking begins. The actual stress should be computed based on the Instantaneous cross-sectional area,not the original.Smce the area is decreasing, the actual or true stress w i l l be greater 2-6. True stress, cr, can be computed by takthan the engineering stress plotted in F~gure ing s~multaneous readings of the load and the mmlmum specimen diameter.The actual area can then be computed, and true stress can be determined as u = W/A

The determination of true strain 1s a bit more complex In place of the change in length divided by the onglnal length that was used to compute engineer~ng strain, true strain is def~ned as the summation of the incremental strains that occur throughout the test. For a specmen that has been stretched from length Lo to length L, the true, nalural, or loganthmzc main would be:

The last equality makes use of the relationship for cyltndrical specimens

that applles only up to the onset of necking. Figure 2-10 depicts the type of curve that results when the data from a uniaxial tensile test areconverted to the form of true stress versus true strain. Since the true stress 1s a measure of the material strength at any point during the test, it will contmue to rise even after necking. Data beyond the onset of necklng should be used with extreme caution, since the geometry of the neck transforms the stress state from uniaxlal tension (stretchmg in one direct~on w~th compensating contractions in the other two) to trlaxial tension, in which the material is stretched or restrained In all three directions. Because of the tnaxtal tension, voids or cracks (Figure 2-11) tend to form in the necked region and serve as a precursor to final fracture. Measurements of the external diameter no longer reflect the true load-bearing area, and the data are further distorted.

Uniform

,deformation,

Nonun~form deforrnat~on

Maxrrnum

load Yield polnt

FIGURE 2-11 Sectloo of a
True strain

2-10 True stress-true strain curve ngineering metal.

tens~le test speclmen stopped just prtorto failure, showing a crack already started in the necked region (Courtesy of E. R. Parker)

36

CHAITBR 2 Properties of Mater~als

Strain

FIGURE 2-12

Stress-strain diagram obtained by unloading and reloading a specimen.

1
3 +

Y!

1,I
/

Large n

Strain Hardening and the Strain-Hardening Exponent. Figure 2-12 is a true stress-true strain diagram, which has been modified to show how a ductile metal (such as steel) will behave when subjected to slow loading and unloading. Loading and unloading within the elastic region will result in simply cycling up and down the linear portion of the curve between points 0 and A.However, if thc initial loading is carried through point B (in the plastic region), unloading will follow the path BeC, which is appraximately parallel to the line OA, and the specimen will exhibit a permanent elongation of the amount OC. Upon reloading from point C, elastic behavior is again observed as the stress follows the line CfD,a slightly different path from that of unloading. Point D is now the yield point or yield stress for the material in its partially deformed state.A comparison of points A and D reveals that plastic deformation has made the material stronger. If the test were again interrupted at point E, we would rind a new, even higher-yield stress. Thus, within the region of plastic deformation, each of the points along the true stress-true strain curve represents the yield stress for the corresponding value of strain. When metals are plastically deformed, they become harder and stronger, a phenomenon known as strain hardening. If a stress is capable of producing plastic deformation, an even greater stress will he required to continue the tlow. In Chapter 3 we will discuss the atomic-scale features that are responsible for this phenomenon. Various materials strain-harden at different rates; that is, for a given amount of deformation different materials will exhibit different increases in strength. Onc method of describine this behavior is to mathematically fit the plastic region of the true mess-true s t r a n curve to the equatlon
u =

K E ~

and determine the best-fit value of n., the strain-hardening exponent.2 As shown in Figure 2-13, a material with a high value of n will have a significant increase in material strength with a small amount of deformati0n.A material with a small n value will show little change in strength with plastic deformation.

Damping Capacity. In Figure 2-12 the unloading and reloading of the specimen follow slightly different paths. The area between the two curves is proportional to the amount of energy that is converted from mechanical form to heat and is therefore absorbed by True strain the material. When this area is large, the material is said to exhibit good damping capacFIGURE 2-13 True stress-true ity and is able to absorb mechanical vibrations or damp them out quickly.This is an imstrain curves for metals with portant property in applicationssuch as crankshafts and machinery bases. Gray cast iron hardening. is used in many applications because of its high damping capacity. Materials with low large a n d small Metals with larger n values damping capacity, such as brass and steel, readily transmit both sound and vibrations. experiencelargeramounts of for a givenstrain' strengthening Rate Considerations. The rate or speed at which a tensile test is conducted can have

a significant effect on the various properties. Strain rate sensitivity varies widely for engineering materials. Plastics and polymers are very sensitive to testing speed. Steels are also sensitive, but aluminum is rather insensitive. Those materials that are sensitive to speed variations exhibit higher strengths and lower ductility when speed is increased. It is important to recognize that standard testing selects a standard speed, which may or may not correlate with the conditions of product application.

COMPRESSION TESTS When a material is subjected to compressive loadings, the relationships between stress and strain are similar to those for a tension test. Up to a certain value of stress, the material behaves elastically. Beyond this value, plastic flow occurs. In general, however, a compression test is more difficult to conduct than a standard tensile test.Test specimens must have larger cross,sectional areas to resist bending or buck1ing.A~ deformation proceeds, the material strengthens by strain hardening and the cross section of the specimen increases, combining to produce a substantial increase in required load. Friction between

regton would h e n

the testing machine surfaces and the ends of the test specimen will alter the results if not properly considered. The type of service for which the material is intended, however, should be the primary factor in determining whether the testing should be performed in tension or compression.

HARDNESS TESTING The wear resistance and strength of a material can also be evaluated by assessing its "hardness." Hardness. is actually a hard-to-define property of engineering materials, and a number of different tests have been developed using various phenomena. T h e most common of the hardness tests are based on resistance to permanent deformation in the form of penetration or indentation. Other tests evaluate resistance to scratching, wear resistance, resistance to cutting or drilling, or elastic rebound (energy absorption under impact loading). Since these phenomena are not the same, the results of the various tests often do not correlate with one another. Caution should be exercised to ensure that the selected test clearly evaluates the phenomena of interest.
Brinell Hardness Test. The Brinellhardness test was one of the earliest accepted methods of measuring hardness. A tungsten carbide or hardened steel ball 10 mm in diameter is pressed into the flat surface of a material by a standard load of 500 or 3000 kg, and the load is maintained for 10 to 15 seconds to permit the full amount of plastic deformation to occur.The load and hall are then removed, and the diameter of the resulting spherical indentation (usually in the range of 2 to 5 mm) is measured using a special grid or traveling microscope. The BrineN hardness number ( B H N ) is equal to the load divided by the surface area of the spherical indentation when the units are expressed as kilograms per square millimeter. In actual practice, the Brinell hardness number is determined from tables that correlate the Brinell number with the diameter of the indentation produced under the various loads. Figure 2-14 shows a typical Brinell tester, along with a schematic of the testing procedure. Portable testers are available for use on pieces that are too large to be brought to a benchtop machine. The Brinell test measures hardness over a relatively large area and is somewhat indifferent to small-scale variations in the material structure. It is relatively simple and

Force (kgf)

FIGURE 2-14 (a) Brinell hardness tester; (b) Brinell test sequence showing loading and measurement of the indentation under magnification with a scale calibrated in millimeters. [(a) Courtesy of Wilson Instruments Division, lnstron Carp., Norwood, MA]

38

CHAPTER 2 Properties of Materials

easy to conduct, and is used extensively on Irons and steels. On the negative side, however, the Brinell test has the following limitations:

1. It cannot be used on very hard or very soft materials. 2. The results may not be valid for thm specimens It is best if the thickness of material is at least 10 times the depth of the indentation. Some srandards spec~fy the minimum hardnesses for which the tests on thm specimens will be considered valid. 3. The test is not valid for case-hardened surfaces. 4. The test must be conducted far enough from the edge of the materlal so that no edge bulging occurs. 5. The substantial indentation may be objectionable on finished parts. 6. The edge or rim of the indentation may not be clearly defined or may be dithcult to see.
The Rockwe11 Test. The widely used Rockwell hardness test 1s simllar to the Brinell test, with the hardness value again being determmed through an indentation produced under a static load. F~gure 2-15a shows the key features of the Rockwell test. A small indenter, e~ther a small-diameter steel ball or a diamond-t~pped cone called a brale, is first seated firmly against the material by the application of a 10-kg "minor" load. This causes a slight elastic penetration into the surface and removes the effects of any surface irregularities. The indicator on the screen of the tester, like the one shown in Figure 2-15b, is then set to zero, and a "malor" load of 60,100, or 150 kg is applied to the indenter to produce a deeper penetration (i.e., plastic deformation). When the Indicating pointer has come to rest, the major load is removed. With the minor load still applied, the tester now indicates the Rockwell hardness number on either a dial gage or digital &splay.This number is really an indication of the depth of the plastic or permanent penetration that was produced by the major load, with each unit representing a penekatlon depth of 2 fim. Different combinations of major loads and Indenters are designated by letters and are used for materials with various levels of strength.Table 2-1 prov~des a partial listing of the Rockwell scales and typical materials for which they are used Because of the diEerent scales, a Rockwell hardness number must be accompanied by the letter corresponding to the particular combination of load and indenter used in its determination. The notation Rc60 (or Rockwell C 60), for example, indicates that a 120" diamond-tipped brale indenter was used in combination with a major load of 150 kg, and

FIGURE 2-15 fa) ODeratinCl

principle of tne Rockwe1 hardness tester: (0) typical Rockwell hardness tester w~th diqital reado~t. [(a) Counery of &&on Instruments Division, lnstron Carp., Norwood, M& (6) csunesy of MTI Corporatron, Aurora, IL.]

Scale Symbol
A B C D E

Panetrator
Brale &1n. b d l Brale Brale m ban &-I". baU h-m ball i - m ball

Load (kg)
60

Typ~cal Materials
Cemented carbides, thln steel, shallow case-hardened steel Copper day$ soft steels, aluminum alloys,malleable Iron Steel, hard cast irons, tltmmm, deep case-hardened steel Thm steel, medrum case-hardened steel Cast rron,alumnum, magnesium Annealed coppers, thm soft ~ h e emetals t Hard copper alloys,malleable irons Alummum,zmc, lead

F G H

100 1% 100 100 60

150
M)

a reading of 60 was obtained.The B and Cscales are used more extensively than the others, with B being common for copper and aluminum and C for steels3 Rockwell tests should not be conductedon thrn materials (typically less than 1.5mm o r 1/16 in.), on rough surfaces, or on materials that are not homogeneous, such as gray cast iron. Because of the small slze of the indentation, variations in roughness, composition, or structure can greatly influence test results.For thin materials, or where a very shallow ~ndentation is desired (as in the evaluation of surface-hardenmg treatments such as mtriding or carburizing), the Rockwell super&~al hurdness test is preferred. Operating on the same Rockwell principle, this test employs smaller major andmnor loads (15 or 45 kg and 3 kg, respectively) and uses a more sensitive depth-measuring dev~ce. Incomparison with the BrineU test, the Rockwell test offers the attractive advantage of direct readings in a single step. Because it requires little (if any) surface preparation and can be conducted quite rapidly (up to 300 tests per hour or 5 per minute), it is often used for quahty control purposes, such as determining if an incommg product meets specification, assuring that a heat treatment was performe? properly, or simply monitoring the properties of products at various stages of manufacture. It has the additlonal advantage of producing a small mdentation that can be easily concealed on the finished product or easily removed in a later operation.
Wckem Hardness Test. The V i i k r s hardness test 1s also similar to the Bnnell test but uses a 136' square-based diamond pyrarmd as the indenter and loads between 1and 120kg. Ltke the Brinell value, the Vickers hardness number is also defined as load divided by the surface area of the indentation expressed in units of kilograms per square millimeter.Tne advantages of the Vickers approach include increased accuracy in determining the diagonal of a square impression as opposed to the diameter of a circle and the assurance that even light loads will produce some plastic deformation. The use of diamond as the indenter material enables the test to evaluate any mater~al and effectively places the hardness of all materials on a single scale. Like the other indentation or penetration methods, the Vickers test has a number of attractive features: (1)it is simple to conduct, (2) little time is involved, (3) little surface preparation is required, (4) the marks are quite small and are easily hidden or removed, (5)the test can be done on location, (6) it is relatively inexpensive,and (7) it provides results that can be used to evaluate material strength or assess product quality.

FIGURE 2-16 Microhardnesr tester (Courtesy of LECO Corporat~on. St. joseph, MI)

Mivohardnerr Tests. Various rnrcrohardness tests have been developed for apphcations where it is necessary to determine the hardnes5 of a very precise area of material or where the material or modified surface layer is exceptionally thin.These tests might be more appropriately termed mrcroindentation hardness rests, since it is the size of the indentation that n extremely smal1,not the measured value of hardness. Special machines, such as the one shown in Figure 2-16,
'The Rockwell C number 1s computed as 1M) - (depth of penetration in em12 pm), while the Rockwell B number is 130 - (depth of penetration m &m/2 wm)

40

CHAmen 2

Properties of Materials

Knoop

(a)

(b)

FIGURE 2-17

(a) Comparison of the dlarnond-tipped indenters used in the Vickers and Knoop hardness tests

(b) Series of Knoop hardness indentations progressing left-to-r~ght across a surface-hardened steel specimen (hardened surface to unhardened core). (Courtesy Buehler Ltd., Lake Bluff, IL)

have been constructed for this type of testing.The locat~on for the test is selected under high magnification. A small diamond penetrator is then loaded with a predetermined load ranging from 25 to 3600 g. In the Knoop test, an elongated diamond-shaped mdenter (long diagonal seven times the short diagonal) IS used and the length of the mdentation 1s measured with the aid of a microscope Figure 2-17 compares the indenters for thevickers and Knoop tests, and shows a series of Knoop indentations progressing left-to-r~ghtacross a surface-hardened steel specimen,from the hardened surface to the unhardened core.The hardness value,known as the Knoup hardness number,is again obtained by dividing the load in kilograms by the projected area of the indentat~on, cxpressed in square mil1imeters.A light-load V~ckers test can also be used to determine microhardness.
Other Hardness Determinations. When tesbng soft, elastic materials,such as rubbers and nonrigid plastics, a durometer can be used T h i s instrument, shown in F~gure 2-18, measures the resistance of a material to elastic penetration by a spring-loaded conical steel indenter No permanent deformation occurs. A slmllar test, used to evaluate the strength of molding sands used in the foundry industry, wll be described in Chapter 14. In the scleroscope test, hardness is measured by the rebound of a small diamondtipped "hammer" that is dropped trom a fixed height onto the surface of the material to be tested.This test evaluate5 the resll~ence of a mater~al, and the surface on which the test is conducted must have a fairly high polish to yield good results. Because the test is based on resilience, scleroscope hardness numbers should only be used to compare similar materials A comparison between steel and rubber, for example, would not be valid. Another definition of hardness 1s the ability of a material to reslst being scratched. A crude but useful test that employs this principle is the file test, where one determines if a material can be cut by a simple metalworking file.The test can be either a pass-fa1 test using a slngle file or a semiquantitative evaluation usmg a series of files that have been pretreated to various levels of known hardness. Refattonships among the Various Hardness Tests. Since the vanous hardness tests often evaluate different material phenomena, there are no simple relationships between the different types of hardness numbers. Approximate relationships have been developed, however, by testing the same material on a variety of dev~ces.Table 2-2 presents a correlation of hardness values for plain carbon and low-aUoy steels It may be noted that for Rockwell C numbers above 20, the Brinell values are approximately 10 times the Rockwell number. Also,for Br~nell values below 320, the Vickers and Brinell values agree quite closely. S~nce the relationships among the vanous tests will differ with material, mechanical processmg, and heat treatment, correlations such as Table 2-2 should be used w~th caution.

FIGURE 2-18 Durometer hardness tester. (Coueesy of Newage Testing Instruments, Southompton, PA )

SECTION

2 . 2

Static Properties

41

Rockwell Number
Bnnell Number
7578 722"
6%"

Tensile Seength

V~ckers

Number
940

C
68

Sderoscope Number
97
92

ksi
368
352

MPa
2537
2427

860

66
64
62

800
745

88
84

337
324

2124
2234

140

148

78
76

22
21

71
68

490
4fi9

135
131

142

137

78

20
20

66

455

126 121 112

a

132
121

72 70 66

64
62
58

441
427

114

Tungsrm, carbtdr hall, other6 standard ball

H2

Properties ~ of Mater~als

FIGURE 2-19 Relationship of

hardness and tensile strength for

a group of standard alloy steels. (Courtesy of ASM Internabonal, Materfals Park, OH)
0 200 400 600 800

Br~nell hardness
Relationship of Hardness to Tensile Strength. Table 2-2 and Figure 2-19 show a definite relationship between tensile strength and hardness. For plain carbon and lowalloy steels, the tensile strength (in pounds per square inch) can be estimated by multiplying the Brine11 hardness number by 500. In this way, an inexpensive and quick hardness test can be used to provide a close approximation of the tensile strength of the steel. For other materials, however, the relationship is different and may even exhibit too much variation to be dependable.The multiplying factor for age-hardened aluminum is about 600, while for soft brass it is around 800.

In many engineering applications, products or components are subjected to various types of dynamic loading. These may include (1) sudden impacts or loads that vary rapidly in magnitude, (2) repeated cycles of loading and unloading, or (3) frequent changes in the mode of loading, such as from tension to compression. To handle these conditions, we must be able to characterize the mechanical properties of engineering mateiials under dynamic loadings. Most dynamic tests subject standard specimens to a weU-controlled set of test conditions. The conditions of actual application, however, rarely duplicate the controlled conditions of a standardized test. While identical tests on different materials can indeed provide a comparison of material behavior, the assumption that similar results can be expected for similar conditions may not always be true. Since dynamic conditions can vary greatly, the quantitative results of standardized tests should be used with extreme caution, and one should always be aware of the test limitations.

IMPACT TEST
Several tests have been developed to evaluate the toughness or fracture resistance of a material when it is subjected to a rapidly applied load, or impact. Of the tests that have become common, two basic types have emerged: (1)bending impacts, which include the standard Charpy and Izod tests, and (2) tension impacts. The bending impact tests utilize specimens that are supported as beams. In the Charpy test,shown schematically in Figure 2-20, the standard specimen is a square bar containing a V-, keyhole-, or U-shaped notch. The test specimen is positioned horizontally, supported on the ends, and an impact is applied to the center, behimd the notch, to complete a three-point bending.The Izod test specimen, while somewhat similar in size

SECTION 2.3

Dynamic Properties

43

Blow

'1
FIGURE 2 - 2 0

\ 1.58 (1116")Saw cut

2

Blow

(0.079)Drill

Charpy test

(a) Standard Charpy impact specimens. Illustrated are keyhole and U notches; dimensions are in millimeters with inches in parentheses. (b) Standard V-notch specimen showing the three-point bending type of impact loading.

I
(a) FIGURE 2 - 2 1

lzod test
lb)

(a) lzod impact specimen; (b) cantilever mode of loading in the lzod test. and appearance, is supported vertically as a cantilever beam and is impacted on the unsupported end, striking from the side of the notch (Figure 2-21). Impact testers,like the one shown in Figure 2-22, supply a predetermined impact energy in the form of a swinging pendulum. After breaking or deforming the specimen, the pendulum continues its upward swing with an energy equal to its original minus that absorbed by the impacted specimen. The loss of energy is measured by the angle that the pendulum attains during its upward swing. The test specimens for bending impacts must be prepared with geometric precision to ensure consistent and reproducible results. Notch profile is extremely critical, for the test measures the energy required to both initiate and propagate a fracture.'Ihe effect of notch profile is shown dramatically in Figure 2-23. Here two specimens have been made from the same piece of steel with the same reduced cross-sectional area. The one with the keyhole notch fractures and absorbs only 43 ft-lb of energy, whereas the unnotched specimen resists fracture and absorbs 65 ft-lb during the impact.

FIGURE 2 - 2 2 Impact testing machine. (Courtesy of Tinius Olsen lnc., Horsham, PA.)

FIGURE 2 - 2 3 hotcheo and ~nnotched;mpact specimens before and after test:nq. Both sper'mens had the same cross-secttonal area, out tne notched specimen fractures while the other doesn't.

44

CHAFTER 2 Properties of Mater~als

FIGURE 2-24 Tensile impact

test.

Caution should also be placed on the use of lmpact data for design purposes. The test results apply only to standard specimens containing a standard notch Moreover, the tests evaluate material behav~or under very specific conditions. Changes in the form of the notch, minor variations in the overall specimen geometry, or faster or slower rates of loading (speed of the pendulum) can all produce s~gnificant changes in the results. Under condit~ons of sharp notches, w d e specunens, and rap~d loading, many ductlie materials lose their energy-absorbing capab~l~ty and fail 1n a bnttle nlanner. [For example, the standard impact test should not be used to evaluatematerials for bullet-proof armor, since the velont~es of loading are extremely drfferent.] The results of standard tests, however, can be quite valuable in assessing a material's sens~tivity to notches and the mult~axial stresses that exlst around a notch. Materials whose properties vary wlth notch geometry are termed notch-sensikve. Good surface finish and the absence of scratches, gouges, and defects in workmanshlp w ~ lbe l key to satisfactory performance. Materials that are notch-insensctcve can often be used mth ascast or rough-machined surfaces w~th no risk of premature failure. Impact testing can also be performed at a variety of temperatures.As will be seen later m this chapter, the evaluabou of how fracture resistance changes mth temperature can he crucial to success when selecting engineer~ng materials for low-lemperalure service. The tensrle impacz test, illustrated schematically in Figure 2-24, ellminates the use of anotched specimen and thereby avoids many of the objections inherent in the Charpy and Izod tests. Turned specimens are subjected to uniaxial impact loadmgs applied through drop weights, modified pendulums, or variable-speed flywheels.

FATIGUE AND THE ENDURANCE LIMIT Materials can also fail by fracture if they are subjected to repeated applications of stress, even though the peak stresses have magnitudes less than the ultimate tensile strength and usually less than the yield strength.This phenomenon, known as fatzgue, can result from either the cyclic repetitlon of a particular loading cycle or entirely random vanations in stress. Almost 90% of all metallic fractures are ~nsome degree attributed to fat~gue. For experimental simplicity, a periodic, sinusoidal loading is often utilized, and conditions of equal-magnitude tension-compression reversals prov~de further simplification.These conditions can be achleved by placing a cyhndrical specimen in a rotating dr~ve and hangmg a we~ght so as to produce elastic bendmg along the axls, as shown in Figure 2-25.As a result of the elasf~c bending, material at the bottom of the specmen 1s stretched, or loaded m tenslon, while material on the top surface is compressed. As the specimen turns, the surface of the speeimen experiences a sinnsoidal application o f tension and compression with each rotation. By conducting multiple tests, subjecting identical specimens to different levels of maxlmum loading, and recording the number of cycles necessary to ach~eve fracture, curves such as that in Figure 2-26 can be produced These curves are known as stress versus number ofcycles, or S-N, curves. If the material being evaluated m Flgnre 2-26 were subjected to a standard tensile test, it would requue a stress in excess of480 MPa (70,000 psi) to mduce failure. Under cyclic loadmg with a peak stress of only 380 MPa

Specimen

Motor

Revolution counter

FIGURE 2-25 Schernat~c

diagram of a Moore rotatingbeam fat~gue machine (Adapted from Hoyden eta/., "The Structure and Properties of Materials", Vol3, p. 15, Wfley,
1965)

Sm.-rroN 2.3 Dynamic Properties

MPa 500

45

Psi x lo3

170

FIGURE 2-26 Typical 5-N curve for steel showlng an endurance h i t Speclflc numbers will vary wlth the type of steel and treatment

Cycles (55,000 psi), the specimen will fail after about 100,000 cycles. If the peak stress were further reduced to 350 MPa (51,000 psi), the fatigue lifetime would be extended by an order of magnitude to approximately 1,000,000 cycles. With a further reduction to any value below 340 MPa (49,000 psi), the specimen would not fail by fatigue, regardless of the number of stress application cycles. The stress below which the material will not fail regardless of the number of load cycles is known as the endurance limit or endurancestrength, and may be an important criterion in many designs. Above this value, any point on the curve is the futigue strength; the maximum stress that can besustained for a specified number of loading cycles. A different number of loading cycles is generally required to determine the endurance limit for diffecent materials.For steels,10million cycles areusually sufficient.For several of the nonierrous metalq500 million cycles may be required. For aluminum, the curve continues to drop such that, if aluminum has an endurance limit, it is at such a low value that a cheaper and much weaker material could bc used. In essence, if aluminum is used under realistic stresses and cyclic loading, it will fail by fatigue after a finite lifetime. The fatigue resistance of an actual product is sensitive to a number of additional factors. One of the most imporrant of these is the presence of stress raisers (or stress concentrators), such as sharp corners, small surface cracks, machining marks, or surface gouges. Data for the S-N curves are obtained from polished-surface, "flaw-free" specimens, and the reported lifetime is the cumulative number of cycles required to initiate a fatigue crack and then grow or propagate it to lailure. If a part already contains a surface crack or flaw, the number of cycles required for crack initiation can be reduced significantly. In addition, the stress concentrator magnifies the stress experienced at the tip of the crack, accelerating the rate of subsequent crack growth. Great care should be taken to eliminate stress raisers and surface flaws on parts that will be subjected to cyclic loadings. Proper design and good manufacturing practices are often more imporlant than material selection and heat treatment. Operating temperature can also affect the fatigue performance of a material. Figure 2-27 shows S-N curves for Inconel 625 (a high-tempcrature Ni-Cr-Fe alloy) determined over a range of temperatures. As temperature is increased, the fatigue strength drops significantly. Since most test data are generated at room temperature, caution should be exercised when the product application involves elevated service temperatures.
MPa Ksi
8"

600

FIGURE 2-27 Fatigue strength of lnconel alloy 625 at various temperatures (Courtesy of Huntrngfon Alloy Products D~vlvon,The Internatronal Nrckel Company, Inc, Toronto, Canada)

1o3

104

lo5

lo@

lo7

lo8

Cycles to fallure

46

CHAWER 2

Propert~es of Materials

Fatigue lifetime can also be affected by changes in the environment. When metals are subjected to corrosion during cyclic loadings, the condition is known as corrosion fatigue, and both specimen lifetime and the endurance limit can be significantly reduced. Moreover, the nature of the environmental attack need not be severe. For some materials, tests conducted in air have been shown to have shorter lifetimes than those run in a vacuum, and further lifetime reductions have been observed with increasing levels of humidity.The test results can also be dependent on the frequency of the loading cycles. For slower frequencies, the environment has a longer time to act between loadings. At high frequencies, the environmental effects may be somewhat masked. The application of test data to actual products, therefore, requires considerable caution. Residual stresses can also alter fatigue behavior. If the specimen surface is in a state of compression,such as that produced from shot peening, carburizing, or burnishing, it is more difficult to initiate a fatigue crack, and lifetime is extended. Conversely, processes that produce residual tension on the surface, such as welding or machining,can significantly reduce the fatigue lifetime of a product. If the magnitude of the load varies during service, the fatigue response can be extremely complex. For example, consider the wing of a commercial airplane. As the wing vibrates during flight, the wing-fuselage joint is subjected to a large number of low-stress 1oadings.The large number of these load applications ,. . . . may be tar less damaging, however, than afew high-stress loadings, urmce timit to Tensile like those that occur when the plane contacts the runway during qr Vsrious Materials landing. From a different perspective, however, the heavy loads may be sufficient to stretch and blunt a sharp fatigue crack, reRatio Mater~al quiring many additional small-load cycles to "reinitiate" it. EvalAlurnlnurn 0.38 uating how materials respond to complex patterns of loading is an 0.29 Beryllium copper (heat-treated) area of great importance to design engineers. 0.33 Copper, hard Since reliable fatigue data may take a considerable time to generate, we may prefer to estimate fatigue behavior from propMagnesturn erties that can be determined more quickly. Table 2-3 shows the Steel approximate ratio of the endurance limit to the ultimate tensile AISI 1035 strength for several engineering metals. For many steels the enScrew stock durance limit can be approximated by 0.5 times the ultimate tenAISI 4140 normallzed sile strength as determined by a standard tensile test. For the Wrought lron nonferrous metals, however, the ratio is significantly lower.

&

.' '

FIGURE 2-28 Progressive fracture of an axle within a ballbearing ring, starting at two points (arrows).

FATIGUE FAILURES Components that fail as a result of repeated or cyclic loadings are commonly called fatigue failures.These fractures form a major part of a larger group known as progressive fractures. Consider the fracture surface shown in Figure 2-28.The two arrows identify the points of fracture initiation, which often correspond to discontinuities in the form of surface cracks, sharp corners, machining marks, or even "metallurgical notches," such as an abrupt change in metal structure. With each repeated application of load, the stress at the tip of the crack exceeds the strength of the material, and the crack grows a very small amount. Crack growth continues with each successive application of load until the remaining cross section is no longer sufficient to withstand the peak stresses. Sudden overload fracture then occurs through the remainder of the material. The overall fracture surface tends to exhibit two distinct regions: a smooth, relatively flat region where the crack was propagating by cyclic fatigue, and a coarse, ragged region, corresponding to the ductile overload tearing. The smooth areas of the fracture often contain a series of parallel ridges radiating outward from the origin of the crack. These ridges may not be visible under normal examination, however.They may be extremely fine;they may have been obliterated by a rubbing action during the compressive stage of repeated loading; or they may be very few in number if the failure occurred after only a few cycles of loading ("low-cycle fatigue"). Electron microscopy may be required to reveal the ridges, or fatigue striations,

SECTION 2.4 Temperature Effects (Both H~gh and Low)

47

FIGURE 2-29 Fatigue fracture of AlSl type 304 stainless steel vtewed in a scannlng electron microscope at 81 OX. Wellstriations are vls~ble def~ned (From "Inferpretatron of SEM
Fractographs," Metals Handbook, Vol. 9, 8th ed, ASM Internabonal, Materials Park, OH, 1970, p. 7 0 )

that are characteristic of fatigue failure. Figure 2-29 shows an example of these markings at high magnification. For some fatigue failures, the overload area may exhibit a crystalline appearance, and the falure is sometimes attributed to the metal having "crystallized." As will be noted in Chapter 3, engineering metals are almost always crystalline matenals.The final overload fracture srmply propagated along the ~ntercrystalline surfaces (grain houndaries) and revealed the already-existing crystalline nature of the material The conclnsion that the material crystalhzed is totally erroneous, and the term is a definite misnomer. Another common error is to classlfy all progressive-type failures as fatigue failures. Other progressive failure mechanisms,such as creep failure and stress-corrosion cracking, will also produce the charactenstic two-regon fracture. In addttion, the same mechanism can produce fractures with different appearances depending on the magmtude of the load, type of loadlng (torsion, bending, or tension), temperature, and operating environment Correct lnterpretatlon of a metal failure generally requires far more information than that acquired by avlsual examination of the fracture surface. A final msconception regarding fatigue falures is t o assume that the failure is tune dependent.The failure of matenah under repeated loads below their static strength is primarily a function of the magnitude and number of loadimg cycles. If the frequency of loading is increased, the time to failure should decrease proportionately. If the time does not change, the failure is dominated by one or more environmental factors, and fattgue is a secondary component.

2.4 TEMPERATURE EFFECTS (BOTHHIGH AND LOW)

The test data used in design and englneerlng decisions should always be obtained under conditions that simulate those of actual service.A number of engineered structures,such as aircraft, spaee vehicles, gas turbines, and nuclear power plants, are required to operate under temperatures as low as -130UC (-200F) or as high as 1250C (2300F).To cover these extremes,the designer must consider both the short- and long-range effects of temperature on the mechanical and physical properties of the material being considered. From a manufacturing mewpoint, the effects of temperature are equally important. Numerous manufacturing processes involve heat, and the elevated temperature and processing may alter the material properties in both favorable and unfavorable ways. A material can often be processed successfully,or economically,only because heating or cooling can be used to change its properties. Elevated temperatures can be q u t e useful 1n moddymg the strength and ductllity of a material. Figure 2-30 summarizes the results of tensile tests conducted over a wide range of temperatures using a medmm-carbon steel. Similar effects are presented for magnesium in Figure 2-31. As expected, an increase in temperature wdl typically Induce a decrease in strength and hardness and an increase in elongation For manufacturing operations such as metalforming, heating to elevated temperature may be extremely attractive because the material is now both weaker and more ductile.
MPa

PSI x

lo3
200
I

Temperature PC)
400 600
1
1

0 120,1

,120 100 80

Tensile strength

Yield

point

3 ',

8 c Z
2
W

40
%URE

40 20
0

% c

2-30 The effectsof Wnperature on the tensile gperties of a medium-carbon

20 200 400 600 800

0 1000 1200 1400

eL

Temperature ("F)

'roperties of Materials
MPa

PSI X

lo3 50 - 1

Temperature PC)
0 100 200

3y,

100

FIGURE 2-31 The effects o f temperature on the tensile properties of magnesium.

Temperature { O F ) F~gure 2-32 shows the comblned effects of temperature and stran 'ale (speed of testing) on the ultlmate tensile strengthof copper.For a given temperature, the rate of deformat~on can also have a strong influence on mechan~cal properties Room-temperature standard-rate tens~le test data will be of little value if the applicatton involves a material being hot-rolled at speeds of 1300 d m i n (5000 ftlrmn). The effect of temperature on Impact properties became the subject of intense study m the 1940s when the Increased use of welded-steel construction led to catastrophicfdures of ships and other structwes while operatrng in cold environments. Welding produces a monolithic (smgle-piece) product where cracks can propagate through a joint and conttnue on to other sections of the structure! Figure 2-33 shows the effect of decreasing temperature on the Impact properties of two low-carbon steels. Although similar in form, the two curves are significantly diierentThe steel rndicated by the solid line becomes brittle (requires very little energy to fracture) at temperatures below -4C (25F) whlle the other steel retains good fracture resistance down to -26C (-15%). The temperature at which the response goes from high energy absorption to low energy absorption is known as the ducttle-to-bnttle transition temperahdre. W e all steels tend to exhibit this transition, the temperature at which it occurs varies with carbon content and alloy. Special cautlon should be taken, therefore, when selecting steels for low-temperature applications.

50
.-

40
0 ,

m m

0
0 0

-40 70

40

80

OF

50

5 m
c

30

." , r :10
0
1 1o2

. 20 z

YI

40

30 20

&

> 30

. ?

20

--

10 2 0 0 20 40"C

'c40

Strain rate, sec-'

Temperature

FIGURE 2-32 The effectsof temperature and strain rate . on the tensile strength of copper. (From A. Nadai and M. 1 Manjoine, Journal of Applied Mechanics, Vol. 8, 194 1, p. A82, courtesy of A5ME.)

FIGURE 2-33 The effect of temperature on the impact properties of two low-carbon steels.

SECTION 2.4 Temperature Effects (Both H~gh and Low)

49

FIGURE 2-34 Lony~tud~nal and transverse notch toughness impact data: steel from the 7itan1cversus modern steel plate, with both longitud~nal and transverse spectmens. (Courtesy l&SM, September 1999, p 33, Iron and Steel 50cfet~ Warrendole, PA )

Test temperature fC)

Figure 2-34 shows the ductile-to-brittle transition temperature for steel salvaged from the ntanic compared to currently used ship plate material. While both are quality materials for their era, the Etancc steel has a much higher transition temperature and is generally more brittle. Recalling that the water temperature at the time the Tctamc struck the iceberg was -2"C, the results show that the steel would have been quite brittle. Two curves are provided for each material, reflectmg specimens in d~fferent orientation with respect to the direction of product rolhng Heie we see that processing features can further affect the properties and performance of a mater~al.

CREEP
Long-term exposure to elevated temperatures can also lead to failure by a phenomenon known as creep If a tensile-type specimen is subjected to a constant load at elevated temperature, it w~ll elongate cont~nuously until rupture occurs, even though the applied stress is below the y~eld strength of the materjal at the temperature of testmg. While the rate of elongation is often quite small, creep can be an important consideration when designing equpment such as steam or gas turbines, power plant bo~lers, and other devices that operate under loads or pressures for long periods of t ~ m e at high temperature If a test specimen is subjected to conditions of f~xed load and fixed elevated temperature, an elongation-versus-hme plot can be generated, similar to the one shown in F~gure 2-35. The curve contains three distinct stages: a short-lived initla1 stage, a rather long secondstage where the elongation rate 1s somewhat linear, and a short-lived third stage leading to fracture.Two significant pieces of engineering data are obtained from this curve the rate of elongation in the second stage, or creep rate, and the total elapsed Fracture

.Slope= -=creep rate

FIGURE 2-35 Creep curve for

a s~nyle specmen at a f~xed temperature, show~ny the three stages of creep and reported creep rate. Note the nonzero straln at time zero due to the ~ n ~ tappl~cat~on ~al of the load.

T
Eo

Time, r

50

CHA~TER 2 Pmperttes of Materials

MPa
Ksi

10

100

1,000

10,000 100,000

0 00001 0 0001

0 001

0 01

0.1

10

Rupture life, hr.

Creep rate. %ihr.

FIGURE 2-37 Creep-rate properties of solutionannealed lncoloy alloy 800 (Courtesy of Huntington Alloy Products Diwsron, The lnternafional Nlckel Company, Inc., Toronto, Canada)

FIGURE 2-36 Stress-rupture d~agram o f solut~onannealed lncoloy alloy 800 (FsNlCr alloy). (Courtesy of Huntrngton Alloy Products Diwsron, The International Nrckel Company, Inc , Toronto, Canada.)
rime to rupture.These

results are unique to the material being tested and the specific condit~ons of the test.Tests conducted a t higher temperatures or with higher applied loads would exh~bit higher creep rates and shorter rupture times When creep behavlor 1s a concern, multiple tests are conducted over a range of temperatures and stresses, and the rupture time data are collected into a single stressrupture diagram, like the one shown m F~gure 2-36. This simple engineering tool provides an overall picture of material performance at elevated temperature. In a sim~lar manner, creep-rate data can also be plotted to show the effects of temperature and stress F~gure 2-37 presents a creep-rate diagram for a high-temperature nickel-based alloy.

11 2.5 MACHINABILITY, FORMABILITY, AND WELDABILITY

Wh~le ~t1.s common to assume that the various "-abilityn terms also refer to specific material properties, they actually refer to the way a material responds to spec~fic processmg techn~ques. As a result, they can be quite nebulous. Machinability, for example, depends not only on the material being machined but also on the specific machin~ng process; the conditions of that process, such as cutting speed; and the aspects of that process that are of greatest interest. Machinabllity ratings are generally based on relative tool life. In certain applications,however, we may be more Interested in how easy a metal is to cut, or how it performs under high-speed machining, and less interested in the tool l i e or the resulting surface finish. For other applicat~ons, surface finlsh or theformation of fine chips may be the most desirable feature. As a result, the term machinabillty may mean different thmgs to different people, and it frequently lnvolves multiple properties of a material interacting with the condit~ons of a process. In a s~milar manner. malleabil~ty, workabzlity, and formability all refer to a material's suitabilityfor plastic deformation processing Since a material often behaves lfferently at different temperatures, a mater~al \nth good "hot formabdity" may have poor deformatlon character~st~cs at room temperature. Furthermore, materials that flow nicely at low deformation speeds may behave in a brittle manner when loaded at rapid rates Formabihty, therefore, needs to be evaluated for aspecific combination of material,process,and process conditions. The results cannot be extrapolated or transferred to other processes or process conditions.Likewise, the weldability of a matenal may also depend on the specific welding or jolning process and the specific process parameters.

A discuss~on of the mechanical properties of materials would not be complete w~thout mention of the many tests and design concepts based on the fracture mechanrcs approach. Instead of treating test specimens as flaw-free materials, fracture mechan-

SECTION 2.6

Fracture Toughness and the Fracture Mechan~cs Approach

51

ics begins with the premise that all materials contain flaws or defects of some given size.These may be material defects,such as pores, cracks, or inclusions; manufacturing defects, in the form of machinimg marks, arc strikes, or contact damage to external surfaces; or design defects, such as abrupt section changes, excessively small fillet radii, and holes. When the specimen is subjected to loads, the applied stresses are amplified or intensified in the vicinity of these defects, potentially causing accelerated failure or failure under unexpected conditions. Fracture mechanics seeks to identify the conditions under which a defect will grow or propagate to failure and, if possible, the rate of crack or defect growth.The methods concentrate on three principal quantities: (1)the size of the largest or most critical flaw, and (3)the fracture toughness, usually denoted as a; (2) the applied stress, denoted by u ; a quantity that describes the resistance of a material to fracture or crack growth, which is usually denoted by K with subscripts to signify the conditions of testing. Equations have been developed that relate these three quantities (at the onset of crack growth or propagation) for various specimen geometries, flaw locations, and flaw orientations. If nondestructive testing or quality control methods have been applied, the size of the largest flaw that could go undetected is often known. By mathematically placing this worst possible flaw in the worst possible location and orientation, and coupling this with the largest applied stress for that location, a designer can determine the value of fracture toughness necessary to prevent that flaw from propagating during service. Specifying any two of the three parameters allows the computation of the third. If the material and stress conditions were defined, the size of the maximum permissible flaw could be computed. Inspection conditions could then be selected to ensure that Paws greater than this magnitude are cause for product rejection. Finally, if a component is found to have a significant flaw and the material is known, the maximum operating stress can be determined that WIU ensure no further growth of that flaw. In the past, detection of a flaw or defect was usually cause for rejection of the part (detection = rejection). With enhanced methods and sensitivities of inspection, almost every product can now be shown to contain flaws. Fracture mechanics comes to the rescue. According to the philosophy of fracture mechanics, each of the flaws or defects in a material can be either dormant or dynamic. Dormant defects are those whose size remains unchanged through the lifetime of the part and are indeed permissible. A major goal of fracture mechanics, therefore, is to define the distinction between dormant and dynamic for the specific conditions of material, part geometry, and applied loading! Alternative efforts to prevent material fracture generally involve overdesign, excessive inspection, or the use of premium-quality materials-all of which increase cost and possibly compromise performance. Fracture mechanics can also be applied to fatigue,which has already been cited as causing as much as 90% of all dynamic failures. The standard method of fatigue testing applies cyclic loads to polished, flaw-free specimens, and the reported lifetime consists of both crack initiation and crack propagation. In contrast, fracture mechanics focuses on the growth of an already-existingflaw.Figure 2-38 shows the crackgrowth rate (change in size per loading cycle denoted as dddN) plotted as a function of the fracture mechanics parameter, AK (where AK increases with an increase in either the flaw size andlor the magnitude of applied stress). Since the fracture mechanics approach begins with an existing flaw, it provides a far more realistic guarantee of minimum service life. Fracture mechanics is a truly integrated blend of design (applied stresses), inspection (flaw-size determination), and materials (fracture toughness).The approach has proven valuable in many areas where fractures could be catastrophic.

m e basic equation of fracture mechanics assumes the form of K r a c 6 .where K is the fracture toughness of the material(a material property),^ is the maximum applied tensile stress,a is the Sire of the largest or most critical flaw, and a is a dimensionless factor that considers the flaw location and tlaw shape. The left side of the equation considers the maferial and theright side describes ths usage condition (a combination of flaw and loading).The relationship is usually described as a greaterthnn or equal.W'hen the material number is greater than the usage condition. the flaw is domant.When equ&ty is achieved, the flaw becomes dynamic, and crack growth or fracture occurs

Stress lntenslty factor range, AK, ksl, In

FIGURE 2-38 Plot of the fatlyue crack growth rate vs AK for a typlcal steel-the fracture mechanics approach. Sim~larshaped curves are obta~ned for most enyineer~ng metals. (Courtesy of ASM Internatronal)

Stress Intensity factor range, AK, MPa Grn

For certain applications, thephysicnlprope~t~s of an engmeeriagmaterial may be even more Important than the mechanical ones These include the thermal,elect~ical, magnehc, and optical charactenst~cs. We have already seen several ways in which the mechan~cal propernes of materials change w ~ t h variations in temperature. In addition to these effects, there are some truly thermulproperties that should be connde~ed.The heat cupacrty or speccfic heat of a materlal is the amount of energy that must be added to or removed from a givenmass of material to produce a 1"change In temperature.Thrs property IS exlremely Important in processes such as castlng, where heat must be extractod rapidly to promote solldifrcation, or heat treatment, where large quant~ties of materlal are heated and cooled. Thermal conductlvrty measures the rate at whlch heat can be transported through a materul. While this may be tabulated separatelvin reference texts. lt is heloful to remember that for mefals, thermal conductlv~;~ is dlriclly proportional to electrical conduct~wty. Metals such as copper, gold, and alumlnum that possess good eleclrical wnductlvity are also good transporters of thermal energy. Thermal expunslonls another Important thermal property. Most materials expand upon heahng and contract upon cooling, but the amount of expansion or contraction will vary with the matenal For components that are machmed at room temperature but put in service at elevated temperatures, or castings that solid~fy at elevated temperatures and then cool, the manufactured dimensions must be adjusted to compensate for the subsequent changes.

Electrical conductivity and electrical resistivity may also be important design considerations. These properties will vary not only with the material but also with the temperature and the way the material has been processed. From the standpoint of magnetic response. materials are often classified as diamagnetic, paramagnetic, ferromagnetic. antiferromagnetic, and ferrimagnetic. These terms refer to the way in which the material responds to an applied magnetic field. Material properties, such as saturation strength, remanence, and magnetic hardness or softness, describe the strength, duration, and nature of this response. Still other physical properties that. may assume importance include weight or density, melting and hoilingpoints, and the various opticalproperties, such as the ability to transmit, absorb, or reflect light or other electromagnetic radiation.

2.8 TESTING STANDARDS AND CONCERNS

When evaluating the mechanical and physical properties of materials, it is important that testing be conducted in a standardized and reproducible manner. ASTM International, formerly the American Society of Testing and Materials, maintains and updates many testing standards, and it is important to become familiar with their contents. For example, ASTM specification E370 describes the "Standard Test Methods and Definitions for Mechanical Testing of Steel Products."Tensile testing is described in specifications E8 and E83, impact testing in E23, creep in E139, and penetration hardness inE10. Other specifications describe fracture mechanics testing as well as the procedures to evaluate corrosion resistance, compressive strength, shear strength, torsional properties, and corrosion-fatigue. In addition, it is important to note not only the material being tested but also the location from which the specimen was taken and its orientation. Rolled sheet, rolled plate, and rolled bars, for example,will have different properties when tested parallel to the direction of rolling (longitudinal) and perpendicular to the rolling direction (transverse).This variation of properties with direction, known as anisotropy, may be crucial to the success or failure of a product.

L1 Key Words
endurance limit engmeeting strain engmeerlng str%s fat~gue fatigue strength fgtigue striations formab~hty Gacture toughness gage length hardness heat capacity Hooke's law Impact test lzod test Knoop hwciness medunability manufacturing defects material defects mechanical propettles metal microhardness modulus of elastmty modulus of resilience neching nonmetal notch-sensitive notch-insensitive offset yield strength percent elongation percent reduction in area performance physical properties plastic deformation processing properties proportional limit resilience Rockwell hardness test S-N curve scleroscope shear specific heat static properties stiffness strain strain hardening strain-hardening exponent strain rate stress stress-rupture diagram structure tensile strength tensile impact test tensile test tension thermal conductivity thermal expansion time to rupture (rupture time') toughness transition temperature true strain true stress ullimate tensile strength uniaxial tensile lest uniform elongation Vickershardness test weldabiiity yield point Young's modulus

h number test Charpy teSL compresston crack growth rate (doidN) creep crew rate

CHA~E 2 R Propert~es of Materials

I I Review Questions
1. A knowledge . of what four aspects IScnhcal to the suocessful apphcation of a matenal m an englneenng des~gn? 2. What are somepropertles commonly assoc~ated wth metallic materials? 3. What are some of the more common nonmetallicengineering materials? 4. What are some of the important physical properties of materials? 5. Why should caution be exercised when applying the results from any of the standard mechanical property tests? 6. What are the standard units used to report stress and strain in the English system? In the metric or SI system? 7. Whatare staticproperties? 8. What is the most common static test related to mechanical properties? 9. Why might Young's modulus or stiffness be an important material property? 10. What are some of the tensile test properties thaf are used to describe o r define the elastic-to-plastic transition in a material? 11. Why is it important to specify the "offset" when providing yield strength data? 12. What are two tensile test properties that can b e used to desmibe the ductility of a material? 13. Is a brittle material a weakmaterial? What does "brittleness" mean? 14. What is the toughness of a material? 15. What is the difference between true stress and engineering stress? True strain and engineering strain? ~ ? mrght t h ~ s 16. What 1s strain hardening or work h a r d e n ~ n How phenomenon be measured or reported? How might rt he used m manufacturing? 17. How mght tenslle test data be mlrlead~ng for a "stram rate sensltlve" matenal? 18. What are some of the different mdterial characteristics o r responses that have been associated w ~ t h the term hardnesr? 19 What are the slmllarihes and d1fferenct-s between the Bnnell and Rockwell hardness tests? 20 Why are there d~fferent Rockwell hardness scales? 21. Whenmlght a mlcrohardness test be preferred over the more standard Brlnell or Rockwell rests? 22 Why m~ght the varlous types of hardness tests fall to agree w~th one another? 2 3 . What 1s the relat~onsh~p between penetration hardness and the ult~mate tensde strength for steel?
24. Describe several types of dynamic loading. 25. Why should the results of standardized dynamic tests be applied with considerable caution? 26. What are the two most common types of bending impact tests? How are the specimens supported and Loaded in each? 27. What aspects or features can significantly alter impact data? 28. What is "notch-sensitivity" and how might it be important in the performance of a product? 29. What is the endurance limit? What occurs when stresses are above it? Below it? 30. Are the stresses applied during a fatigue test above or below the yield strength (as determined in a tensile test)? 31. What features may significantly alter the fatigue Lifetime or fatigue behavior of a material? 32. What relationship can be used to estimate the endurance limit of a steel? 33. What material, design, or manufacturing features can contribute to the initiation of a fatigue crack? 34. What are fatigue striations and why do they form? 35. Why is it important for a designer or engineer to know a material's properties at aU possible temperatures of operation? 36. Why should one use caution when using steel at low temperature? 37. How might we evaluate the long-term effect of elevated temperature on an engineering material? 38. What is a stress-rupture diagram, and how is one developed? 39. Why are terms such as machinability, formability, and weldability considered to be poorly defined and therefore quite nebulous? 40. What is the basic premise of the fracture mechanics approach to testing and design? 41. What three principal quantities docs fracture mechanics attempt to relate? 42. What are the three most common thermal properties of a material, and what do they measure? 43. Describe an engineering application where the density ol the selected material would h e an important material consideration. 44. Why is it important that property testing be performed in a standardized and reproducible manner? 45. Why is it important to consider the orientation of a test specimen with respect to the overall piece of material?

I I Problems
1. Select a product or component for which physical properties are more important than mechanical properties. a. Describe the product or component and its function. h. What are the most important properties or characteristics? c. What are the secondary properties or characteristics that would also be desirable? 2. Repeat Problem 1 for a product o r component whose dominant required properties are of a static mechanical nature. 3. Repeat Problem 1 for a product or component whose dominant requirements are dynamic mechanical properties.
4. One of the important considerations when selecting a material for an application is to determine the highest and lowest operating temperatures along with the companion properties that must be present at each extreme. The ductile-to-brittle transition temperature, discussed in Section 2.4, has been an important factor in a number of failures. An article that summarized the features of 56 catastrophic brittle fractures that made headline news between 1888 and 1956 noted that low temperatures were present in nearly every case. The water temperature at the time of the sinking of the Titanic was above

Case Study the freezing point for salt water but helow the transition point for the steel used in construction of the hull of the ship. a. Which of the common engineering materials exhibits a ductile-to-brittle transition? b. For plain carbon and low-alloy steels, what is a typical value (or range of values) for the transition temperature? c. What type of material would you recommend for construction of a small vessel to transport liquid nitrogen within a building or laboratory? d. Figure 2-34 summarizes the results of impact testing performed on hull plate from the Titanic and similar material produced for modern steel-hulled ships. Why should there be a difference between specimenscut longitndinally (along the rolling direction) and transversely (across the rolling direction)? What advances in steel makiig have led to the significant improvement in low-temperature impact propertles? 5. Several of the property tests dcscnhed m t h ~chapter r produce results that are q u ~ t e sensit~ve to the presence or absence o i

55

notches or other flaws. The fracture mechanics approach to materials testing incorporates flaws into the tests and evaluates their performance.The review article mentioned in Problem 4 cites the key role of a flaw o r defect in nearly all of the headline-news fractures. a. What are some of the various "flam or defects" that might be present in a product? Consider flaws that might be present in the starting material, flaws that might be introduced during manufacture, and flaws that might occur due to shipping, handling, use, maintenance, o r repair. b. What particular properties might be most sensitive to flaws o r defects? f c. Discuss the relationship of flaws to the various types o loading (tension vs. compression, torsion, shear). d. Fracture mechanics considers both surface and interior flaws and assigns terms such as crock injfkfor,c-nckpropngator, and crack arrestor. Briefly discuss why location and orientation may be as important as the physical size of a flaw.

ecause of the amount of handling that occurs during manufacturing operations, along with the handling of loading, unloading, and shipping, material mix-ups and mixed materials are not an uncommon occurrence. Mixed materials also occur when industrial scrap is collected or when discardedproducts a r e used as raw materials through recycling. Assume that you have equipment to perform each of the tests described in this chapter (as well as access to the full spectrum of household and department store items and even a small machine shop). For each of the following material combinations, determine a procedure that would permit separation of the mixed materials. Use standard datasource references to help identify distinguishable properties.

I. Steel and aluminum cans that have been submitted for

2. Stainless steel sheets of Type 430 ferrltic stainless and Type 316 austenitic stainless. 3. 6061-T6 aluminum and A291 magnesium that have become mixed in a batch of machine shop scrap. 4. Transparent bottles of polyethylene and polypropylene (both thermoplastic polymers) that have been collected for recycling. 5. Hot-rolled bars of AlSl 1008 and 1040 steel. 6. Hot-rolled bars of AlSl 1040 (plain-carbon) steel and 4140 steel (a molybdenum-containing alloy)

CH.

NATURE OF METALS AND ALLOYS

1 3.1 STRUCTURE-PROPERTY-PROCESSING-PERFORMANCE

RELATIONSHIPS

As drscussed 1n Chapter 2, the success of many engineering actrvities depends on the selection of engineering materials whose properties match the requuements of the application. Pnmitlve cultures were often limited to the naturally occurring materials in their environment. As civilization developed, the Spectrum of engineering materrals expanded. Materials could be processed and therr properbes altered and possibly enhanced. The alloying or heat treatment of metals and the firing of ceramics are examples of techniques that can substant~ally alter the properties of a materral. Fewer compromises were required, and enhanced deslgn possrb~ties emerged. Products, m turn, became more sophstrcated.Whrle the early successes in altermg materials were largely the result of trial and error, we now recognize that theproperttes andperformance of amateriai are adirect result of its ytructure andproceumg. If we want to change the properbes, we will most likely have to Induce changes in the material structure. Srnce all materrals are composed of the same basic components-parucles that include protonu, neutrons, and electrons-rt is amazing that so many different materials exist with such widely varying properties. This variation can be explained, however, by the many possible combinatrons these unrts can assume in a macroscopic assembly The subatomic particles combine rn d~fferent arrangements to form the various elemental a t o m , each having a nucleus o f protons and neutrons surrounded by the proper number of electrons to mamtain charge neutra11tyThespecific arrangement of the electrons surrounding the nucleus affectsthe electrical,magnetic, thermal, and optical propertles as well as thc way the atoms bond to one another. Atomic bonding then produces a hlgher level of structure, whlch may he in the form of a molecule, crystal, or amorphous aggregate This structure, along w ~ t h the imperfections that may be present, has a profound effect on the mechanical propertles.The size, shape, and arrangement of multiple crystals, or the mixture of two or more different structures withm a materia1,produces a higher level of structure, known as mitrostructure. Variatrons 1n microstructure further affect the matenal properties. Because of Lhe ab~lity to control structures through processing and the ability to develop new structures through techniques such as composite materials, engineers now have at their disposal a wide variety of materrals with an almost unlimited range of properties The specific propertles of these materials depend on all levels of structure, from subatomic to macroscopic (Figure 3-1). This chapter will attempt to develop an understanding of the basic structure of engineermg materials and how changes in that structure affect their properties and performance

SELTION 3.3
Structure

Atomic Bonding

57

Subatomic
-Nuclear-

Atomlc

Molecule

M~cro

Macro

-Electr~cal -Chemical
FIGURE 3-1 General
-Thermal
+

re,at:onshp between structural (eve anfl the varous woes of engineering propertie;.

-Mechanical Propert~es

3.2 THE STRUCTURE OF ATOMS

Experiments have revealed that atoms consist of a relatively dense nucleus composed of positively charged protons and neutral parhcles of nearly identical ma%,known as neutrons. Surrounding the nucleus are the negatively charged electrons, which appear in number7 equal to the protons, sa as to maintains neutral charge balane. Disttact groupings of these basic particles produce the known elements,ranging from the relatively simplk hydrogen atom to the &stable transuraninnl atoms ovei2% times a6 heavy. Except for density and specific heat, however,the weight of stolll has very little influence on their engineering properties. The light electrons that surround the nucleus play an extremely significant role in dexermining material properties. These electrons are arranged in a characteristic sfructure consisting of shells and subshells,each of which can contain only a limited number of electrons.The first she& nearest the nucleus, can contain only two.The second she11 can convain eight, and the third, 32. Each shell and subshell is most stable when it is completely filled. For atoms containing electrons in the third shell and beyond, however, relative stability 1 s achieved with eight electrons in the outermost layer or subshell. If an atom has slightly less than the number of outer-layer electrons required for stability, it will readily accept an electron from another source. IT will then have one eIectron more than the number of protons and becomes a negatively charged atom, or negaffveLon. Depending on the number of additional electrons, ions can have negative charges of 1,2,3,or more. Conversely,ifan atom has a slight excess of electrons beyond the number required for stastaby (such as sodium, with one electron in the third shell), it will readily give up the excess electron and become apasltrve ton The rema~ning electrons become more strongly bound, so further removal of electrons becomes progressively more difficult. The number of electrons surrounding the nucleus of a neutral atom is called the atomic nwnber. More important, however, are those electrons in the outermost shell or subsheI1, which are known as valence electrons. These are influential in determ~nmg chemical propertieq electrical conductivity, some mechanical properties, the nature of interatomc bonding, atom size, and opttcal characteristics.Elements with simrlar electron configurations in their outer shells tend to have similar properties

Atoms are rarely found as free and independent units; they are usually linked or bonded to other atoms in some manner as a result of interatomic attraction.73~ electron structure of the atoms plays the dominant role in determining the nature of the bond. Three types of p r r m y bonds are generally recognized, the simplest of wluchis the rorPic bond. If more than one type of atom is present, the outermost electrons can break free from atoms with excesses in their valence shell, transforming them hto positive ions The electrons then transfer to amms w~th deficlenwes in their outersheU,eonverting them into negative ~ons.The positive and negative Ions have an electrostatic attraction for each other, resulting in a strong bonding force. F~gure 3-2 presents a crude schematie of the ionic bonding process for sodium and chlorine. Ionized atoms do not usually unite in

58

CWAPTER 3 Nature of Metals and Alloys

@ --@ @ @
.--

-------

Na

CI

FIGURE 3-3 Three-

FIGURE 3-2

Na+-

C I

lon~zat~on of sod~um and chlorine, producing stable outer shells by electron transfer.

dimensional structure of the sodlum chlor~de crystal Note how the various lons are surrounded by ions of the opposlte charge.

@ J @ j $
FIGURE 3-4 Formation of a chlorine molecule by the electron sharing of a covalent bond.

0@fTJ
la)
(bl

&
ic)
FIGURE 3-5 Examples of covalent bonding in (a) nitrogen molecule, (b) HF, and (c) diamond.

slmple pair%however.AU posrtively charged atoms attract all negauvely charged atoms Therefore,eachsodium ion will attempt to surround Itself wth negative chlorine ions, and positive sodium ions Smce the ateach chlorine ion will attempt to surround itself w~lh l a three-dimensional structure, lrkethe traction is equal in all directions, the result w ~ l be one shown in Figure 3-3 Since charge neutrality must be maintained withm the structure, equal numbers of positive and negative charges must be present in each nerghborhood. General characteristics of materials loined by ionic bonds mclude moderate to high strength, high hardness,brittleness,h~gh melting pomt, and low electrical conductivity (since electrons are captive to atoms, charge transport requrres atom-r ion-movement). A second typeof primary bond is the coval~nt type Here the atoms in the assmbly find it impossible to produce completed shells by electron transfer but achieve the same goal through electron shanng.Ad]acent atoms sharc outer-shell electrons so that each achieves a stable electron wnfiguration.The shared (negat~vely charged) electrons locate between the pos~tive nucle~, forming a positive-negative-positsve bondrng link. Figure 3-4 illustrates this type of bond for a pair of chiorme atoms, each of which contarns seven electrons in the valence she1l.De result is a stable two-atom n~olecule,CI~. Stable molecules can also form from the sharing of more than one electron from each atom, as m the case of nitrogen (Figure 3-5.3). The atoms in the as~cmbly need not be identical (as in HF, Fsgure 3-5b), the sharing does not have to be equal, and a s~ngle atom can share electrons with more than one other atom For atoms such as carbon and srhcon,with four electrons in the valence shel1,one atom may share its valence electrons wrth each of four neighboring atoms. The resulting structure is a three-drmonional network of bonded atoms,liie the one shown in Figure3-5c Like the ~onic bond, the covalent bond tends to produce materials with high strength and high melting point. S~nce atom movement withm the three-dimensional structure (plastlc deformat~on) requires the breakrng of discrete bonds, covalent materials are characteristically brittle. Electrical conductivity depends on bond strength, ranging from conductive tsn (weak covalent bonding), through semrconductive silicon and germanium, to insulating diamond (carbon). Ionic or covalent bonds are commonly found in ceramic and polymerrc mate~sals A thrrd type of primary bond is possible when a complete outer shell cannot be formed by either electron transfer or electron sharing.This bond is known as the metallic bond (Figure 3-6). If each of the atoms in an aggregate contains only a few valence electrons (one, two,or three), these electrons can be easily removed to produce "stable" Ions. The positive ions (nucleus and inner, nonvalence electrons) then arrange in a threednnensianal penodrc array and are mrrounded by wandering, un~versally shared valence electrons,sometimes referred to as an electron cloud 01 electron gas These highly mob~le, free electrons account for the h~gh electrical and thermal conductivityvalues as well as the opaque (nontransparent) characteristicobserved in metal$ (the free electrons are able to absorb the various discrete energies of light radiation). They also provide the "cement" required for the positive-negative-positive attractions that result in bonding Bond strength, and therefore materral strength and melting temperature, vanes over a wrde range.More sigtuficant, however, is the ohservatlon that the positwe ions can now move within the structure without the breaking of discrete bonds Materials bonded by metallic bondscan be deformed by atom-movement mechanisms and produce an altered shape that IS every hit as strong as the origsnal This phenomenon is the bas~s of metal plasticity, enabling the wide vanety of forming processes used in the fabr~cation of metal products.

SscrloN 3.6

Crystal Structures of Metals

59

FIGURE 3-6 Schematic of the metallic

-

bond showing the positive ions and free electrons for copper. Each positivecharged ion contains a nucleus with 29 protons and stable electron shells and subshells containing the remaining 28 electrons.

Weak or secondary bonds, known as vun der Waalsforce$, can form between molecules that possess a nonsymmetrical distribution of electr~cal charge, Some molecules, such as hydrogen fluor~de and water,' can bev~ewed as electric dipoles. Certa~n portlons of the molecule tend to be more positive or negative than others (an effect referred to aspolarzzation).The negahve part of one molecule tends to attract the positive region of another, forming a weak bond.Van der Waals forces contribute to the mechan~cal properlles of a number of molecular polymers, such as polyethylene and polyv~nyl chloride (PVC).

ATOMARRANGEMENTS IN MATERIALS
As atoms bond together to form aggregates, we find that the particular arrangement of the atoms has a significant effect on the material properties. Depending on the manner of atomic grouping, materials are classified as having molecular structures, crystal structures, or amorphous structures. Molecular structures have a rlistinct number of atoms that are held together by primary bonds. There is only a weak attraction, however, between a glven molecule and other similar groupings Typical examples of molecules include 02, H20,and GH, (eththese materials exhibit ylene). Eachmolecule is free to act more or less independently,~~ relatively low melting and boiling polnts Molecular materials tend to be weak, since the molecules can move easily with respect to one another. Upon changes of state from sol~d to liquid or l~quid to gas, the molecules remain as dlstinct entities. Solid metals and most minerals have a crystalline strncture. Here the atoms are arranged in a three-d~mens~onal geometric array known as a lattice Latt~cesare descnbable through a unit bulldig block, or unit cell, that is essentially repeated throughout space. Crystalline structures will be discussed more fully in Section 3.6. In an amorphous structure, such as glass, the atoms have a certain degree of local order (arrangement withrespect to neighboring atoms), but when viewed as an aggregate, they lack the periodically ordered arrangement that is characteristic of a crystalhne solid.

From a manufacturing viewpoint, metals are an extremely important class of matenals. They are frequently the materials bemg processed and often form both the tool and the machinery performing the processing. More than 50 of the known chemical elements are classified as metals, and about 40 have commercial importance They are character~zed by the metall~c bond and possess the d~stmguish~ng characteristics of strength, good electrical and thermal condnctivity,luster, the ability to be plastically deformed to a fair degree without fracluring, and a relatively high specific gravlty (density) compared to
'The H,O molecule can be vlaved as a lO9* boomerang 01 elbow w ~ t h oxygen m the mlddle and the two hvdrozens o n the extendme a m s T h e emht valence electrons (SIX from oxveen and two from hvdroeen)

location of adjacent molecules

60

CHAPTER 3 Nature of Metals and Alloys

Metal
Alum~nurn Copper Gold

Iran Lead Magnesium

Silver

T i n Titanium

nonmetals.The fact that some metals possess properties d~fferent from the general pattern simply expands thelr engineering utility. When metals solidify, the atoms assume a crystalline structure; that is, they arrange themselves tn a geometr~c lattlce Many metals exist in only one lattice form. Some, however, can exist in the sohd state in two or more lattice forms, with the particular form depending on the condihons of temperature and pressure. These metals are said to be allotrop~c orpolymorphzc (poly means "more than one"; morph means "structure"), and the change from one lattice form to another is called an aliotroplc tran~furmatton.Themost notable example of such a metal is iron, where the allotropic change makes possible heat-treating procedures that yield a wide range of final properties It is largely because of its allotropy that iron has become the basis of our most unportant alloys. Lattrcc Type There are 14 basic types of crystal structuresor lattices.ForFace-centered cuhic thmately,however,nearly d l of the commerciallylmportant metFacc centered cubic als solidify into one of three lattice types: body-centered cubic, ~a~e-centcrcd cub^ face-centered cubic, or hexagonal close-packed Table 3-1 lists Body centered c u h ~ the room-temperature structure for a number of common metFace-centered cuh~c ale. Figure 3-7 compares the structures to one another as well Hexagonal as to the easily visualized, but rarely observed, simple cubic Face centered cubic structure Body-centered tetragonal To begin our study of crystals,consider the simple cublc Hexagonal structure illustrated in Figure 3-7a. This crystal can be constructed by placing single atoms on all comers of a cube and then linking identmal cube units together Assuming that the atoms are rigtd spheres with atomic radit touching one another, computation reveals that only 52% of available space is occupied. Each atom is in direct contact with only SIX neighbors (plus and minus d~rections along thex,y,and z axes). Both of these observations are unfavorable to the metallic bond, where atoms desire the greatest number of nearest neighbors and highefficiency packing.

structure

Lattice

Un~t cell schematic

Prng-Pong ball

Number of

model

nearest ne~ghbors

Packing Typical ffficiency metals - -

Simple c u b ~ c

52%

None

- -

Body-centered
cub~c

:b, W, Ta, i, V, Na, I

-

e, Cr, Mn

Face-centered cub~c

li, Ca, Au

e, Al, Cu,

ig, Pb, Pt

FIGURE 3-7 Comparison of crystal structures. slrnple cub~c, body-centered cub~c, facecentered cubtc, and hexagonal close-packed

Hexagonal close-packed

le, Cd,
4% Zn. r
-

SLcrroh. 3.7

Development of a Grain Structure

61

FIGURE 3-8 Close-packed atomic plane showing three d,rections of atom touching or close pack~ng.

BS 3.7 DEVELOPMENT OF A GRAIN STRUCTURE

When a metal solidifies, a small particle ot solid forms lrom the llquld with a latbce structure characteristic of the given material. This particle then acts hke a seed or nucleus and grows as other atoms attach themsclves.The basic crystalline unit,or unit cell, 1s repcaled, as illustrated m the examples of F~gure 3-9. In actual solidlficatlon, many nuclel form independently throughout the hquid and have random orientat~ons wlth respect to one another. Each then grows untll it begin? to interfere with its neighbors. Since the adjacent lattice structures have diflerent ahgnments or orientations,growth cannot produce a single continuous structure, and a polycrystalline solid is produced Figure 3-10 prov~des a two-dimcns~onal illustration of this phenornenon.The small,continuous volumes of solid are known as crystals orgram, and the surfaces that dtvide them (i.e., the surfaces of crystalline discontinuity) are known as gram boundaria.The process of solidification 1s one of crystal nucleatzon and growth.

V T g
/

The largest region of unoccupied space is in the gcometrlc center of the cube, where a sphere of 0.732 times the atom diameter could be inserted? if the cube is expanded to permit the tnsertion of an entire atom, the body-catered-cuhcc (BCC) structure results (Figure 3-7b). Each atom now has eight nearest neighbors, and 68% of the space is occupied.This structure e more favorable to metals and is observed in roomtemperature iron, chromium, manganese, and Lhe other metals l~sted in Figure 3-7b Compared to materials wrth other structures, body-centered-cubic metals tend ro be high strength. In seeking efficient packlng and a large number of adjacent neighbors, consider maximizing the number of spheres ln a single laycr and then stacking those layers. The layer of maximized packing is known as a close-packedplane and exhiblts the hexagonal symmetry shown ~nF~gure 3-8. The next layer is powtioned wlth spheres occupying either the "point-up" or "point-down" recesses in the original layer. Depending on the sequence in which the various layers are stacked, two divtlnclly different structures can the orlglnal plane and be produced. Both have twelve nearest neighbors (SIX with~n three from each of the layers above and below) and a 74% efficiency of occupying space If the layers are stacked in sets of three (original location, point-up recess ot the ongmal layer, and point-down recess of the origmal layer), the resulting structure can also be viewed as an expanded cube with an atom Inserted in the center of each of the six cube faces.This is the face-cmiered-cubic (FCC) structure shown in Flgure 3-7c. It is the preferred structure for many of thc engineering metals and tends to provide the exceptionally hlgh ducttlity (ablhty to be plastically deformed w~thout fracture) that n characteristic of aluminum, copper,stlver, gold, and elevated-temperature iron. A stackmg sequence of any two alternating layers results in a structure known as hexagonal close-packed (HCPJ,where the individual close-packed planes can be clearly Identified (F~gure 3-7d). Metals hawng this structure,such as magnesium and zinc, tend to have poor ductility,fall in a brittle manner,and often require s p a a l processingprocedures

---

- I -- - -

(a)

(bl

1CURE 3-9 Growth of crystals ) produce a n extended lattice: ,d)'line schematic; (b) Ping-Pong ball model.

FIGURE 3-10 Schematic representation of the growth of crystals to produce a polycrystall~ne material.
'The diagonal of a cubeis equal to 6 times the length of thecube edge, and the cube edgc is hereequal to 1wo atomic radii or one atomic diarneter.Thus thc diagonal iszqual lo 1.732times the arom diameter and is made up of an atomic radius, open space, and another alomic radius. Since two radii equel one diameter, the open spacemust be equal in size to 0.732 times the atomic diameter.

62

C H ~3 RNature of Metals and Alloys

FIGURE 3-11

Photomicrograph of alpha ferrlte pure iron) showing (essent~ally gram and gram boundarles; (Courtesy of Un~ted States Steel Corp , P~ftsburgh,PA.)

Grains are the smallest unit of structure in a metal thatcan be observed with an ordinary light microscope. If a piece of metal is polished to mirror finish with a series of abrasives and then exposed to an attacking chemical for a short time (etched), the grain structure can be revealed.The atoms along the grain boundarles are more loosely bonded and tend to react with the chemical more readily than those that are part of the grain interior. When viewed under reflected light, the attacked boundaries scatter light and appear dark compared to the relatively unaffected (still flat) grains (Figure 3-11). In some cases, the individual grains may be large enough to be seen by the unaided eye, as with some galvanized steels, but usually magnification is required. The number and size of the grains in a metal vary with the rate of nucleation and the rate of growth.The greater the nucleation rate, the smaller thcresulting grains. Conversely, the greater the rate of growth, the larger the grams. Because the resulting gram structure will Influence certain mechanical and physical properties, it is an important property for a n engineer to contrd and specify. One means of specification is through the ASTM grain size number, defined in ASTM specification El12 as

where Nis the number of grains per square inch visible in a prepared specimen at l00X magnification, and n is the ASTM grain size number. Low ASTM numbers mean a few massive grains, while high numbers refer to materials with many small grains.

The mechanical properties of a material are highly dependent on its crystal structure.An understanding of mechanical behavior, therefore, begins wlth an understanding of the way crystals react to mechanical loads.Most studles begin with carefully prepared single crystals Through them, we learn that the mechanical behavior n dependent on (1) the type of lattice, (2) the interatomic forces (~.e., bond strength), (3) the spacing between adjacent planes of atoms, and (4) the density of the atoms on the various planes If the apphed loads are relatively low, the crystals respond by simply stretching or compressing the distance between atoms (Figure 3-12). The basic lattice unit does not change, and all of the atoms remain in their original posltlons relative to one another. The apphed load serves only to alter the force balance of the atomic bonds, and the atoms assume new equilibrium positions with the applied load as an additional component of force. If the load is removed, the atoms return to thelr original positions and the crystal resumes its original size and shape. The mechanical response is elastic m nature, and the amount of stretch or compression is directly proportional to the applied load or stress. Elongation or compression in the direction of loading results m an opposzte change of dimensions at nght angles to that direction. The ratio of lateral contraction to axial tensile strain is known as Poisson's ratlo.Thls value is always less than 0.5 and is usually about 0.3.

FIGURE 3-12 Distort~on of a crystal lattice in response to various elastic loadings.

Unloaded

Tension

Compression

Shear

HATION
As the magnitude of applied load becomes greater, distortion (or elastic strain) contrnuzs to maease, and a point is reached where the atoms either (1)break bonds to produce a fracture or (2) slrde over one another in a way that would reduce the load. For metalhc materials, the second phenomenon generally requrres lower loads and occurs preferentially.The atomic planes shear over one another to produce a net displacement or permanent shift of atom positions, known asplastrc defurmatron. Conceptually, this is similar to the distortron of a deck of playing cards when one card slides over another. The actual mechanism, however, is really aprogressive one rather than one rn which all of the atoms rn a plane shift simultaneously More significantly, however, the result is a permanent change inshape that occurs without a concurrent detenoratron inproperties. Recall~ng that a crystal structure is a regular and periodic arrangement of atoms in space, we see that it becomes possible to link the atoms into flat planes in an almost mfinite number of ways Planes having dlfferent orientations with respect to the surfaces of the unit cell wlll have different atomic densities and different spacing between adjacent, parallel planes, as shown ~n Figure 3-13, Grven the chorce of all possibilities, plast~c deformatton tends to occur along planes having the highest atomic density and greatest separation. n t e rationale for this can be seen m the simplified two-d~mensional array of Figure 3-14. Planes A and A' have hrgher density and greater separation than planesB and B'. In visuahzing relative motion, we see that the atoms of B and B' would interfere signrf~cantly with one another, whereas planesA and A' do not experrence this difticulq. Although Figure 3-14 represents the planes of sliding as lules,crystal structures are actually three-dimensional. Within the preferred planes are also preferred drectrons. If sliding occurs in a direction that corresponds to one of the close-packed directrons (shown as dark lines m Figure 3-8),atoms can simply follow one another rather than each having to negotiate its own path. Plastic deformat~on, therefore, tends to occur by the preferential sllding of maximum-density planes (close-packed planes if present) In directions of closest pachng.The specific combination of plane and d~rection is called a slip system, and the resulting shear deformation or sliding is known as slip. The ability to deform a gven metal depends on the easc of shearing one atomlc plane over an adjacent one and the orientation of the plane with respect to the applied load. Consider, for example, the deck of playing cards.The deck will not "deform" when laid flat on the table and pressed from the top or when stacked on edge and pressed unrformly. The cards will shde over one another, however, if the deck is skewed with respect to the applied load so as to induce a shear stress along the plane of sl~drng. With this understanding, consider the deformatlon properties of the three most common crystal structures:

FIGURE 3-13 Schematic diagram showrng crystall~ne planes w~th dlfferent atomlc densities and Interplanar spacings.

FIGURE 3-14

Slrnple schematic ~llustratlng the lower deformatlon resistance of planes wlth higher atomkc densiw and larger interplanar spaclng.

1 . Body-centered cubic. In the BCC structure, there are no close-packed planes. Slip occurs on the most favorable alternatrves, which are those planes with the greatest interplanar spacing (SIX of which are illustrated in Figure 3-15).Wtthin these planes,slip occurs along the directions of closest packing, whrch are the cube diagonals If each specific combination of plane and direction is conadered as a separate slip system, we find that the BCC materials contarn48 attractive ways to slip (plastically deform).The probability that one or more of these systems will be oriented in a favorable manner is great, but the force requred to produce deformation is extremely large since there are no

FIGURE 3-15 Slip planes within the BCC, FCC, and HCP

BCC

FCC

HCP

I!

64

CHA!TER 3

Nature of Metals and Alloys close-packed planes. Materials with this structure generally possess high strength with moderate ductlhty (Refer to the typical BCC metals in F~gure 3-7.)

2. Face-centered cubrc. In the FCC structure, each unit cell contains four close-packed planes, as illustrated in Fignre 3-15. Each of those planes contains three close-packed directions, or face diagonals,g~ving12 poss~ble means of slip. Again, the probability that one or more of these will be favorably oriented is great, and t h ~ time, s the force required to induce slip is qulte low. Metals with the FCC structure are relatively weak and possess excellent ductility,as can be confirmed by a check of the metals llsted in Figure 3-7.
3. Hexagonal close-packed The hexagonal lattice also contams close-packed planes, but only one such plane exists w~thln the lattlce. Although this plane contams three close-packed dlrecttons and the force required to produce s l ~ p is agaln rather low, the probability of favorable orientation to the applied load n small (especially if one cons~ders a polycrystalline aggregate). As a result, metals with the HCP structure tend to have low ductility and are often classifred as brittle.

I I i 3.10 DISLOCATION THEORY OF SLIPPAGE

A theoretical calculation of the strength of metals based on the sliding of entire atomlc planes over one another predicts yield strengths on the order of 3 mdlion pounds per square inch,or 20,000 MPa.The observed strengths in actual testing are typ~cally 100 to 150 times less than thls value. Extremely small laboratory-grown crystals, however, have been shown to exh~bit the full theoreticalstrength. An explanation can be provided by the fact that plastic deformation does not occur by all of the atoms m one plane shpp~ng simultaneously over all the atoms of an adjacent plane. Instead, deformation is the result of the progressive slippage of a localized disruption known as a dislocation. Consider a simple ana1ogy.A carpet has been rolled onto a floor, and we now want to move it a short dlstance in a given direction. One approach would be to pull on one end and try to "shear the carpet across the floor," simultaneously overcoming the fr~ctional resistance o f the entire area of contact. This would require a large force actrng over a small distance.An alternative approach might he to form a wrinkle at one end of the carpet and walk the wrinkle across the floor to produce a net shift in the carpet as a whole-a low-force-over-large-distanceapproach to the same task. In the region of the wrmkle, there is an excess of carpet with respect to the floor beneath it, and the movement of this excess is relahvely easy. Electron microscopes have revealed that metal crystals do not have all of their atoms in perfect arrangement but rather contain a variety of localized imperfections Two such imperfections are the edge dislocation and screw dlslocatlon (Figure 3-16), Edge d~slocations are the edges of extra half-planes of atoms. Screw d~slocations correspond to parha1 tearlng of the crystal plane In each case the d~slocation 1s a disruption to the regular, periodic arrangement of atoms and can be moved about with a rather low applied force. It is the motion of these atormc-scale dislocations under applied load that is responsible for the observed macroscopic plastic deformation. All engineering metals contain dislocations, usually in abundant quantltles. The ease of deformation depends on the ease of inducing dislocation movement. Barriers to dislocat~on motion tend to Increase the overall strength of a metal.These barriers take the form of other crystal imperfect~ons and may be of the point type (missing atoms or vacanc~es, extra atoms or mtershtzals,or substitution atom7 of a different variety, as may occur in an alloy), line type (another d~slocation), or surface type (crystal grain boundary or free surface). To increase the strength of a material, we can either remove all defects to create a perfect crystal (extremely difficult) or work to impede the movement of exlstlng dislocations by adding other crystalline defects.

FIGURE 3-16 Schernat~c

representat~on of screw and edge d~slocat~ons.

I I i 3.11 STRAIN HARDENING OR WORK HARDENING

As noted in our discuss~on of the tensile test m Chapter 2,most metals become stronger when they are plastically deformed, a phenomenon known as strain hardenzng or work hardenmg. Understanding of this phenomenon can now come from our knowledge of

JGURE 3-17

Schematic representation of slip and crystal rotation resulting from deformation.

dtslocat~ons and a further extension of the carpet analogy. Suppose that this time our goal is to move the carpet diagonally The best way would be to move a wnnkle in one direction, and then move a second one perpendicular to the first. But suppose that both wrinkles were started simultaneously. We would find that wrinkle 1 would impede the 2, and vice versa. In essence, the feature that makes deformation easy motion of wr~nkle can also serve to impede the motion of other,similar dislocations. In metals plastic deformation occurs through dislocation m0vement.A~ dislocations move, they are more likely to encounter and interact with other dislocations or crystalline defects, thereby producingresistance to further motion. In addition,mechanisms exist that markedly increase the number of dislocations in a metal during deformation (usually by several orders of magnitude), thereby enhancing the probability of interaction. The effects of strain hardening become attractive when one considers that mechanical working (metalforming) is frequently used in the production of metal products. Since strength can be increased substantially during deformation, a strain-hardened (deformed), inexpensive metal can often be substituted for a more costly, stronger one that is machined to shape.As the product shape is being formed, the material is simultaneously becoming stronger. Experimental evidence has confirmed the dislocation and slippage theory of deformati0n.A transmission electron microscope can reveal images of the individual dislocations in a thin metal section, and studies confirm both the increase in number and the interaction during deformation. Macroscopic observations also lend support. When a load is applied to a single metal crystal, deformation begins on the slip system that is most favorably oriented.The net result is often an observable slip and rotation,like that of a skewed deck of cards (Figure 3-17). Dislocation motion becomes more difficult as strain hardening produces increased resistance and rotation makes the slip system orientation less favorable. Further deformation may then occur on alternative systems that now offer less resistance, a phenomenon known as cross slip.

Commercial metals are not single crystals but usually take the form of polycrystalline aggregates. Within each crystal, deformation proceeds in the manner previously described. Since the various grains have different orientations, an applied load will produce different deformations within each of the crystals.This can be seen inFigure 3-18, where a metal has been polished and then deformed.The relief of the polished surface reveals the different slip planes for each of the grains. One should note that the slip lines do not cross from one grain to another. The grain boundaries act as barriers to the dislocation motion (i.e., the defect is confined to the crystal in which it occurs). As a result, metals with a finer grain structure-more grains per unit area-tend to exhibit greater strength and hardness, coupled with increased impact resistance. This near-universal enhancement of properties is an attractive motivation for grain size control during processing.

FIGURE 3-18

Slip lines in a polycrystalline material. (From Richard Hertrberg, Deformation and Fracture Mechanics of Engineering Materials; courtesy of lohn Wiley h Sons, lnc.)

66

CHA~E 3 R Nature of Metals and Alloys

1 3.13 GRAIN SHAPE AND ANISOTROPIC PROPERTIES

When a metal is deformed, the grains tend to elongate in the diiecbon of metal flow (Figure 3-19). Accompanying the nonsymmetnc structure are nonsymmetr~c or dlrectlonally varylng properties. Mechanical properties (such as strength and ductility), as well as physical propert~es (such as electrical and magnetlc characteristics), may all exhibit directional differences. Properties that vary with direction are said to be are zsotropzc. ancsotropcc. Properties that are urxform in all d~rections The directional variation oi properties can be harmful or beneficial, and therefore such variation assumes importance to both the part designer and the part manufacturer. If the metal flow is controlled in processes such as forging, enhanced strength or fracture resistance can be imparted to certain locations. Caution should be exercised, however, since an improvement in one direction 1s generally accompanied by a decllne m another. Moreover, directional variation in properties may create problems during further processmg operations, such as the further formmg of rolled metal sheets.

I,

FIGURE 3-19 Deformed grams in a coldworked 1008 steel after 50% reduction by roll~ng. (From Metals Handbook, 8th ed. ASM Internat~onal,Mater~olz Park, OH, 1972.)

If too much plastic deformation is attempted,the metal may respond by fracture. When plastlc deformation precedes the break, the fracture 1s known as a ductile fracture. Fractures can also occur before the onset of plastic deformation.These sudden,catastrophic failures, known as brlttle fractures, are more common in metals having the BCC or HCP crystal structures. Whether the fracture ia ductile or brittle, however, often depends on the specific conditions of mater~al, temperature, state of stress, and rate of loading.

During plastic deformation, a portlon of the deformahon energy is stored withm the mater~al m the form of addibonal d~slocations and increased grain boundary surface area.3 If a deformed polycrystalline metal n subsequently heated to a high enough temperature, the material will seek to lower its energy. New crystals nucleate and grow to consume and replace the original structure (Figure 3-20).This process of reduclng the internal energy through the formation of new crystals is known as recrystallizatzon.The temperature at which recrystallization occurs is different for each metal and also vartes with the amount of prior deformation. The greater the amount of prior deformation,

the applied load is removed, the balloon snaps back to its original shape, the one involving the lsast surface energy. Metalr behave in an analogous manner. During deformation, the distortion of the crystals increases the mergy of the material. Given the opportunity, the material will try to lower its energy by returning to spherical grains.

FIGURE 3-20

Recrystall~zat~on

of 70-30 cartridge brass: (a) cold-worked 33%; (b) heated at 580C (1075F) for three

seconds, (c)four seconds, and (d) e~ght seconds; 45X. (Courtesy o f / E Burke, General Electr~c Company, Farrfreld, CT.) the more stored energy and the lower the recrystallization temperature.There is a lower limit,however, below which recrystallization will not take place in a reasonable amount of time.Table 3-2 gives the lowest practical recrystallization temperatures for several materials.This temperature can often be estimated by taking 0.4 times the melting point of the metal when the melting point is expressed as an absolute temperature (Kelvin or Rankinelmis is also the temverature at which atomic diffusion (atom movement within l h ? % c ~ l i I~ hc.onic\ l ~i$nii~c~rlt. indic.~t~ng 1l1;1t cl~ifus~un is 311~mport:~nl I Cn CI , hIli/:~i~o~i. llization ~ i i c : h ~ ~ ~in ii> ~ \\ l~cn nir.t:~lsar- ~,l.,.tic.~ll\ Jc,L>rn~cJ :11 trrnpzraturc\ bclo\v lhcir I ~ I I I I ~ ~ I i Ii p r t rh . J i I I r . 111: n i c t ~ \trcngthcns l In. sir,~i~i h;lrJ<nind.. I I ~ thc rdsultant s1ru;turc c,,n" Temperature ["F("C)] sists of distorted grains. As deformation continues, the metal decreases 3(H)(150) in ductility and may ultimately fracture. It is a common practice, there390 ( 2 0 ~ ) fore, to recrystallize the material after a certain amount of cold work. 390 (200) Through this recrystallization anneal, the structure is replaced by one of 840 (450) new crystals that have never experienced deformation. All strain hardBelow room lrmpcrature ening is lost, but ductility is restored, and the material is now capable of 300 (150) further deformation without the danger of fracture. 1100 (590) If the temperature of deformation is sufficientlyabove the recrys390 (200) tallization temperature, the delormation process becomes hot working. Below room temperature Recrystallization begins as soon as sufficient driving energy is created, Room temperature (i.e., deformation and recrystallization take place simultaneously), and extremely large deformations are now possible. Since a recrystallized grain structure is constantly formin& the fmal product will not exhibit strain hardening. Recrystallization c a n also be used to control or improve the grain structure of a material. A coarse grain structure can be converted to a more attractive fine grain structure through recrystallization.The material must first be plastically deformed to store sufficient energy to provide the driving force. Subsequent control of the recrystallization process then establishes the final grain size.

Metal
Alum~num Copper Gold

lron
,I ,
,

Lead

Magnesium
Nickel Silver

'I

Ti" Zinc

,
A

I,

68

CHA~. 3R Nature of Metals and Alloys

Recrystallizatron is a continuous process in which a material seeks to lower its overall energy Ideally,recrystallizatlon wdl result in a structure of un~fornl crystals wlth a comparatively small grain s~ze. If a metal is held at or above its recrystalliuation temperature for any appreciable tlme, however, the grains in the recrystallized structure willcontinue to Increase in size. In effect, some of the grains become larger at the expense of theil smaller neighbors as the material seeks to further lower its energy by decreasing the amount of grain boundary surface area. Slnce engineering properties tend to diimrsh as the size of the grams increase, control of recrystallization n of prlme importance. A deformed materlal should be held at elevated temperature just long enough to complete the recrystalliiat~on process.The temperature should then be decreased to stop the process and avoid the property changes that accompany gram growth

1 1 3.17 ALLOYS AND ALLOY TYPES

Our discussion thus far has been directed toward the nature and behavior of pure metals. For most manufacturing applications, however, metals are not used In t h e ~ r pure form Instead, engineering metals tend to be alloy~,mater~als composed of two or more dlfferenl elements, and they tend to exhibit their own characterlstlc properties. There are three ways in which a metal might respond to the add~tion of another eIement.De fist, and probably the simplest,response occurs when the two mntercals are ~nsoluble m one another m thesolrdstute In this case the base mela1 and the alloying addition each maintains its individual identity, structure, and properties.The alloy in effect becomes a composite structure, consisting of two types of bullding blocks in an intlmate ~nechantcal mixture. The second possibility occur5 when the two elementr cxhihrr some degree of soluwhere the alloy element hilcty m the solid stute.The two materials form a solid ~olution, dissolves in the base metal.The solutions can be (1)substztuhonul or (2) interstitral. In the substitutional solution, atoms of the alloy element occupy lattice sites normally filled by atoms of the base metal. In an interstitial solution, the alloy element atoms squeeze into the open spaces between the atoms of the base metal lattlce. A thlrd possrbility exists where the elements combcne to form intermetall~c compounds. In this case, the atoms of the alloying element interact w~th the atoms of the base metal in definite proportions and in detinite geometric relationships The bonding is primarlly of the nonmetallic variety (i.e ionlc or covalent), and the lattice structures are often qurte complex. Because of the type of bonding, intermetall~c compounds tend to be hard, but brlttle, high-strength materials Even though alloys are composed of more than one type of atom, their structure IS still one of crystalline lattices and grains.Their behavior in response to applied loadlngs is simdar to that of pure metals, wrth some features reflecting the Increased level of structural complexity Dislocation movement can be further unpeded by the presence of unllke atoms. If neighboring grains have different chemistries andlor structures, they may respond differently to the same type and magnitude of load.

In addltion to mechanical properties, the structure of a material also influences its conducrrvity refers to physical properties, such as its electrical behavior. Ele~trtcal the net movement of charge through a material. In metals, the charge carriers are the valence electrons The more perfect ihe atomic arrangement, the greater thefreedom of electron movement and the htgher the electrical conductivity.Latt~ce Imperfections or ~rregularitresprovide impediments to electron transport, and lower conductlvtty. The electrical resistance of a metal, therefore, depends largely on two factors: (1)lattice lmperfechons and (2) temperature.Vacant atomlc sites, mterst~tial atoms, subrtitutional atoms, dislocations, and grain boundaries all act as disruptions to the regularity

of a crystalline lattice. Thermal energy causes the atoms to vibrate about their equilibrium position. These vibrations cause the atoms to be out of position, which further interferes with electron travel. For a metal, electrical conductivity will decrease with an increase in temperature.As the temperature drops, the number and type of crystalline imperfections become more of a factor.The best metallic conductors, therefore, are pure metals with large grain size, at low temperature. The electrical conductivity of a metal is due to the movement of the free electrons in the metallic bond. For covalently bonded materials, however, bonds must be broken to provide the electrons for charge transport.Therefore. the electrical properties of these materials is a function of bond strength. Diamond, for instance, has strong bonds and is a strong insulator. Silicon and germanium have weaker bonds that are more easily broken by thermal energy. These materials are known as intrinsic semiconductors, since moderate amounts of thermal energy enable them to conduct small amounts of electricity. Continuing down Group IV of the periodic table of elements, we find that tin has such weak bonding that a high number of bonds are broken at room temperature, and the electrical behavior resembles that of a metal. The electrical conductivity of intrinsic semiconductors can be substantially improved by a process known as doping. Silicon and germanium each have four valence electrons and form four covalent bonds. If one of the bonding atoms were replaced with an atom containing five valence electrons, such as phosphorus or arsenic, four covalent bonds would form, leaving an additional valence electron that is not involved in the bonding process. This extra electron would be free to move about and provide additional conductivity. Materials doped in this manner are known as n-type extrinsic semiconductors. A similar effect can be created by substituting an atom with only three valence electron&such as aluminum.An electron will be missing from one of the bonds, creating an electron hole. When a voltage is applied, a nearby electron can jump into this hole, creating a hole in the location that it vacated. Movement of electron holes is equivalent to a countermovement of electrons and thus provides additional conductivity.Materials containing dopants with three valence electrons are known as p-type semiconductors. The ability to control the electrical conductivity of semiconductor material is the functional basis of solid-state electronics and circuitry. In ionically bonded materials, all electrons are captive to atoms (ions). Charge transport, therefore, requires the movement of entire atoms, not electrons. Consider a large block of salt (sodium chloride). It is a good electrical insulator until it becomes wet, whereupon the ions are free to move in the liquid solution and conductivity is observed.

111 Key Words

allbtropic allq amarphollstitnctuw anisotropic ASTM grain sgze number body-centered cubic brittle fracture dose-packed planes cold wotk favalent bend ~qrss~ip crystal Structure dislmatiaa edge dislMation elastin ddormetion electrical eondudivity extn~sic ~c~nductor tam-qnteted* c gram grain boundary grain gtowth grain srze hexagonal close-packed hot work mtennetalbc compound inmtitial intrimic semiwnductar ion ionic bond isotropic lattice metallic bond microstructure molecular structure negative ion nucleation and growth plastic deformation Poisson's ratio polarization polymorphic positive ion primary bond recrystallization screw &slocatlon secondary bonds simple cublc sllp system solld solution straln harden~ng Structure substitutional atom umt cell vacancy valence electrons van der Waals forces work hardemng
S ~ P

~OP% ductile fra-re

70

CHAITER 3 Nature of Metals and Alloys

Review Questions
1. Why might an engineer be concerned with controlling or altering the structure of a material? 2. What are the next levels of structure that are greater than the atom? 3. What is meant by the term microstructure? 4. What is an ion and what are the two varieties? 5. What properties or characteristics of a material are influenced by the valence electrons? 6. What are the three types of primary bonds,and what types of atoms do they unite? 7. What are some general characteristics of ionically bonded materials? 8. What are some general properties and characteristics of covalently bonded materials? 9. What are some unique property features of materials bonded by metallic bonds? 10. For what common engineering materials are van der Waals forces important? 11. What is the difference between a crystalfinematerial and one with an amorphous structure? 12. What is a lattice? A unit cell? 13. What are someofthe general charaeteristicsofmetallic materials? 14. What is an allotropic material? 15. Why is the simple cubic crystal structure not observed in the engineering metals? 16. What are the three most common crystal structures found in metals? 17. What is the efficiency of filling space with spheres in the simple cubic structure? Body-centered-cubic structure? Facecenteted-cubic structure? Hexagonal-close-packed structure? 18. What is the dominant characteristic of hody-centered-cubic metals? Face-cenlered-cubic metals? Hexagonal-close-packed metals? 19. What is a grain boundary? 20. What is the most common means of describing or quantifying t h e grain size of a solid metal? 21. What is implied by a low ASTM grain size number? A large ASTM grain size number?

22. How does a metallic crystal respond to low applied loads? 23. What is plastic deformation? 24. What is a slip system in a material? What types of planes and directions tend to be preferred? 25. What structural features account for each of the dominant properties cited in Question 18? 26. What is a dislocation? What role d o dislocations play in determining the mechanical properties of ametal? 27. What are some of the common barriers to dislocation movement that can he used to strengthen metals? 28. What are the three major types of point defects in crystalline materials? 29. What is the mechanism (or mechanisms) responsible for the observed deformation strengthening or strain hardening of a metal? 30. Why is a fine grain sue often desired in an engineering metal? 31. What is an anisotropic property7Why might anisotropy be a concern? 32. What is the difference belween brittle fracture and ductile fracture? 33. How does a metal increase its internal energy during plastic deformation? 34. In what ways can recrystallization be used to enable large amounts of deformation without fear of fracture? 35. What is the major distinguishingfeature between hot and cold working? 36. Why is grain growth usually undesirable? 37. What types of structures can he produced when an alloy element is added to a base metal? 38. As a result of ionic or covalent bonding, what types of mechanical properties are characteristic of intemetallic compounds? 39. How is electrical charge transported in a metal (electrical conductivity)? 40. What features in a metal structure tend to impede or reduce electrical conductivity? 41. What is the differencebetween an int~sicsemiconductor and an extrinsic semiconductor?

II Problems
1. It is not uncommon for subsequent processing to expose manufactured products to extremely elevated temperatures. Zinc coatings can be applied by immersion into a bath of molten zinc (hot-dip galvanizing). Welding actually melts and resolidifies the crystallir~emelals. Brazing deposits molten filler metal.How might each of the structural features listed below, and their associated properties, be altered by an exposure to elevated temperature? a. A recrystallized polycrystalline metal b. A cold-worked metal c. A solid-solution alloy such as brass, where zinc atoms dissolve and disperse throughout copper
2. Polyethylene consmts of fibrous molecules of covalently bonded atoms tangled and lnteractlng hke the fibers of a cotton ball. Weaker van der Waals forces act between the molecules w ~ t h a strength that isinversely related to separabon d~stance. a What properties of polyethylene can be attnbuted to the covalent bondlngl b. What properties are most likely the result of the weaker van der Waals forces? e If we pull on the ends of a cotton hall, the cotton fibers go from a random arrangement to an array of somewhat ahgned fibers Assummg we get a similar response from de formed polyethylene, how might properhes change? Why?

EQUILIBRIUM PHASE DIAGRAMS AND THE IRON-CARBON SYSTEM

4.1 INTRODUCTION 4.2 PWES 4 3 EQUILIBRIUM PHASE DIAGRAMS Temperature-Composxtion 4.5 STEELS AND ?FIE SIMPLLFIED IRON-CARBON DLAGRAM 4 6 CAST IRONS Utsiation of Diagams Types of Cast Iron Solidficatlon of Alloy X The Role of Processing an ThreePhase Reaclions Propert~es Iatermetatlic Compounds Case Study. THEBLACXS~I~~H Comple%~iagams 4.4 IRON<A$SON EQ~JILI~IUUM A ~ v m DIA~RAM

pama1 Sokd Solubility

Insolubility

Cooling Curves Solubibty Studrs Complete Solubility in Both Liqu~d and Solid States
iie

rge old

l i 4.1 INTRODUCTION
As our study of engineering materials becomes more focused on spec~bc metals and alloys, 11 is increasingly important that we acqulre an understand~ng of their natural characterislics and properties What is the baslc structure of the material? 1 % the material unlCorm throughout, or 1s it a mixture of two or more distinct components? If there are multiple components, how much of each is present, and what are the different chemistries? Is there a component that may impart undesired properties or characteristics? What will happen if temperature is Increased or decreased, pressure is changed, or chemistry 1s varied? The answers to these and other important questions can be obtained through the use of equmlrbrrumphase dmagrams.

uce

and

1 4.2 PHASES
Before we move to a discussion of equilibrium phase diagrams,it 1s important that we first develop a worhng definition Of the term phase. As astarting definition, a phaseis simply a form of material possessing a characteristic structure and characteristic prope~ties Uniformity of chemistry, structure, and properties is assumed throughout a phase. More rigorously, a phase has a definable structure, a uniform and idc~~tmfzabLe chemlsny (also known as composrt~on), and distinct boundar~es or interfaces that separate it from other differentphases. A phase can be contmuous (like the air in a room) or discontinuous (like grains of salt in a shaker). A phase can be solid, hquid, or gab. i n addition, a phase can be apure substance or a solution,provided that the structure a i ~ d composition are uniform through. therefore form a s~ngle phase wher out.Alcoho1 and water rnlx in all proportions and WIII combined.lhere are no boundaries across wh~ch structure and/or chemistry changes. Oi and water, on the other hand, tend to formisolated regions with distinct boundaries a ~ ( must be regarded as two dishnct phases. Ice cubes m water are another two-phasesystem since there are two distinct structures w ~ t h interfaces between them.

ded )alL mth

the

S 60

that ,deby?

1 4 3 EQUILIBRIUM PHASE DIAGRAMS

An equtl&hnum phase d~agram is a graphic mapping of the natural tendencies of material or a material system, assuming that equilibrium has been attained for all poss ble conditions~There are three primary variables to be considered:temperature, pressur

(water)

(vapor)

Trlple polnt
I
I

100 Temperature- ' C (not to scale)

FIGURE 4 1 Pressure-temperature equil~brlum phase diagram for water

and composztzon.The simpled phase diagram 1s a pressure-temperature (P-T) diagram for a fixed-composition material. Areas of the diagram are assigned to the varlous phases, with the boundaries mdicating the equihbrium conditions of transit~on. As an introduction, consider thc pressure-temperature diagram for water, presented as Figure 4-1. W ~ t h the composition fixed as H,O, the diagram maps the stable form of water for varlous conditions of temperature and pressure. If the pressure is held constant and tempe~ature is varied, the region boundaries denote themelting and bolling points. For example, at 1atmosphere pressure, the dlagram shows that water melts at 0C and boils at 100C.Still other uses are possible. Locate a temperature where the stable phase is liquid at atmospheric pressure. Maintaining the pressure at 1atmosphere, drop the temperature until the material goes from llquid to solid (i.e., Ice). Now, maintain that new temperature and begm to decrease the pressure. A transition will be encountered whcre sohd goes directly to gas without melting (subhmation). The combined process just described, known asfreeze drying, 1s employed in the manufacture of an appropriate phase diagram, process condit~ons numerous dehydrated products. W ~ t h can be determined that might reduce the amount of required cool~ng and the m a p t u d e of pressure drop requred for sublimation. A process operating about the triple point would be most efficient.

TEMPERATURE-COMPOSITION DIAGRAMS While the P-T diagram for water is an excellent introductionto phase diagrams,P-Tphase diagram are rarely used for engmeering applications.Most enpneenng processes are conducted at atmospheric pressure, and variations are more likely to occur in temperature and composition. The most useful mapping, therefore, is usually a temperature-composition phave diagram at 1atmosphere pressure. For the remainder of the chapter, this will be the form d phase &agram that will be considered. For mapping purposes, temperature is placed on the vertical axis and composition on the horizontal. Figure 4-2 shows the axes for mapping the A-B system, where the left-hand vertical corresponds to pure material A, and the percentage of B (usually expressed m weight percent) increases as we move toward pure matenal B at the right side of the diagram.The temperature range often includes only solids and liquids, since few processes involve engineering matenals in the gaseous state. Experimental investigations that provide the details of the diagram take the form of either verbcal or horlzontal scans that seek to locate the transitions betweenphases. COOLING CURVES Conwderable information can be obtained from vertical scans through the d~agram where a fixed-composition material n heated and slowly cooled. If the cooling history

is platted in the farm of a tempatwe-versus-time plot, known as a cooling cum@, t h transithas in structurevvill appear as characteristic p h i s , such a$ slope changa or isothermal (constmt-tmpe~atw~) holds; Consider thesysem composed of sodium chloride (common tablesalt) and water. Five different aooling curves are presenhd in Agure 4 3 . C a v e (a) is fa^ pure water be-ing emled from the l@M hat&A desreasing-temperature limis observed for the L i q uid where the removal of heat prrtduce;sa concurrent drop in temperature. When the freezing point af0"Cis re$ctUed (poinr a ) , h material begins to c h w e state a h d s e i e w hear energy as partuf the liquid-to-solidVansition. Heat is beingcontinuo&ly extracted fram the system, but since its source is now the cbange in state> there isno campauioa deaease in tempaaturaAn ignShermal. ar eonstant-temperature hold (neb)is observed until the aolidiBcaZion is complrste, From this point, as heat extraction continue% the newly fwmed solid -experiences a steady drop in temperature. This type of curve is cbaract@rLstk of pure meals and other substsnces with a distinctmelting point. Crwe [b) in R@rs 4-3 pfments the cooling curve for a solution of 10% salt in warer.The liquid region unaerg~es wntinww cooling dawn to pointo,where the slap absuptly decreases At this temperature, mall partides of ice (i.6,solid) begin t o farm

Time + (a1 0% NaCl

Tlms

Time

(b) 10% NaCl

(c) 23.5% NaCl

(dl 50% NaCl

Time ie) 100% NaCl

I
I
I

I
+ .

E 3

Constant-compos~t~on scan icooling curve)

m m

I I
I

I I

I I

Isothermal scan

FIGURE 4-2

Mapping axes for Composition (weight percent

B)

a temperature-composit~on
equil~br~urn phase diagram.

is plotted in the form of a temperature-versus-time plot, known as a cooling curve, the transitions in structure will appear as characteristic points, such as slope changes or isothermal (constant-temperature) holds. Consider the system composed of sodium chloride (common table salt) and water, Five different cooling curves are presented in Figure 4-3. Curve (a) is for pure water being cooled from the liquid state.A decreasing-temperature line is observed for the liquid where the removal of heat produces a concurrent drop in temperature. When the freezing point of OC is reached (point a), the material begins to change skate and releases heat energy as part of the liquid-to-solidtransition. Heat is being continuously extracted from the system, but since its source is now the change in state, there is no companion decrease in temperature. An isothermal or constant-temperature hold (a-b) is observed until the solidification is complete. From this point, as heat extraction continues, the newly formed solid experiences a steady drop in temperature. This type of curve is characteristic of pure metals and other substances with a distinct melting point. Curve (b) in Figure 4-3 presents the cooling curve for a solution of 10% salt in water.The liquid region undergoes continuous cooling down to point c, where the slope abruptly decreases.At this temperature, small particles of ice (i.e., solid) begin to form

[a) 0% NaCl

(b) 10% NaCl

(c)23.5% NaCl

CURE 4 3

Coollng curves five different solut~ons of salt d water. (a) 0% NaCI; 10% NaCI; (c) 23.5% NaCI; ) 50% NaCI; (e) 100% NaCl

Time

id) 50% NaCl

Time -4
ie) 100% NaCl

74

CHAPTER 4 Equ~l~br~um Phase D~agrams and the Iron-Carbon System

FIGURE 4-4 Partial equilibrium diagram for NaCl and H 2 0 derived from cooling-curve information.

0 (HzO)

10

255

50

100 (NaCI)

% NaCl

and tlte reduced slope is attributed ro the energy releasedin this transition.The formation of these ice particles leaves the remainmg solution r~cher in salt and imparts a lower freezing temperature. Further cooling results in the formation of additional solid, which cont~nues to enrich the solution and further lowers the freezing pomt of the remaining liquid. Instead of possessing a distmnct melting point or freezmg point, this material is s a ~ d to have a freeringrange.When the temperature of point d is reached, the remaining 11quid tmdergoes an abrupt reaction and solidifies into an intimate mixture of solid salt and solid water (discussed later), and an isothermal hold is observed. Further extrachon of heat produces a drop in the temperature of the fully sohdified mater~al For a solution of 23.5% salt in water, a dist~nct freeling point 1s again observed, as shown in curve (c). Compositions with richer salt concentration, such as curve (d),show phenomena similar to those in curve (b), but with salt being the first solid to form from the liquid. Fmally, the curve for pure salt, curve (e). exhibits behavior similar to that of pure water If the observed transition points are now transferred to a temperaturecomposition diagram, such as Flgure 4-4, we have the beg~nniugs of a map that summarizes the behavior of the system. Line ax-f-h-1, denoting the lowest temperature at 1 s totally liquid,ls known as the liqurdus line. Line d-f-j denotes a parwhich the rnater~al ticular three-phase reaction and will be discussed later. Between the line&two phases coexist, one being a hquid and the other a sohd The equilibrium phase diagram, therefore, can be v~ewed as a collective presentation of cooling-curve data for an entlre range of alloy compositions. Our coohng-curve studies have provlded some key information regarding the salt-water system, mcluding some might into the use of salt on htghways in the winter. With the admtion of salt, the freenng point of water can be lowered from P C (32F) to as low as -22C (-7.6"F).

SOLUBILITY STUDIES
The observant reader will note that the ends of the d~agram still remain undetermined. Both pure materials have a distinct melting po~nt, below wh~ch they appear as a pure solid Can ice retain some salt m a single-phase solid solut~on? Can solid salt hold some water and remarn a single phase? If so, how much, and does the amount vary w ~ t h temperature? Completion of the diagram, therefore, requires 9everal horizontal scans to determine any solubllrty limzt~ and their possible variation with temperature These so thermal (constant-temperature) scans usually require the preparation o f specnnens over a range of composition and theusubsequent examination by X-ray technlques, microscopy, or other methods to determine whether the structure and chemistty are uniform or ~ndicate a two-phase mixture.As we move away from a pure matenal, we often encounter a single-phase solid solution, m which a small amount of one component is d~ssolved and dispersed throughout the other. If there is a limit to this solubihty there wll be a hne in the phase diagram, known as a solvus line, denoting the conditions where the single-phase sohd solution becomes a two-phase mixture. Figure 4-5 presents the equ~hbrium phase diagram for the lead-tin system, using the conventional notation in

Weight percent tln

which Greek letters are used to denote the vanous single-phase sohds. The upper portion of the diagram closely resembles the salt-water diagram, but the partial solubil~ty of one material in the other canbe observed on both ends of the diagram.'

COMPLETE SOLUBILITY I N BOTH LIQUID AND SOLID STATES Having developed the basic concepts of equilibrium phase diagrams, we now consider a series of examples 1 n which solubility changes. If two materials are completely soluble in each other in both the liquid and solid states, a rather simple diagram results,Sie line, the lowest the copper-nickel diagram of Figure 4-6. The upper lme is the l~quidus temperature for which the matenal is 100% 1iquid.Above the liquidus, the two materials form a uniform-chemistry Liquid solution. The lower line, denoting the highest temperature at which the material is completely solid, is known as asol~dusline. Below the solidus, the materials form a solid-state solution in which the two types of atoms are uniformly distributed throughout a smgle crystalline lattice. Between the liquidus and solidus is afrrezlng range, a two-phase reglon where Liquid and solid solutions coexist. PARTlAL SOLID SOLUBILITY Many materials do not exhibit complete solub~lity in the sohd state, Each is often soluble in the other up to a certan limit or saturation pomt, which varies wlth temperature. Such a diagram has already been observed for the lead-hn system in Figure 4-5. At the pomt of maximum solubility, 183OC, lead can hold up to 19.2 wt% tin in B single-phase solution and tin can hold up to 2 5% lead w ~ t hais structure and still be a single phase. If the temperature is decreased, however, the amount of solute that can be held in solution decreases in a continuous manner. If a saturated solutionof tin in lead

We~ght percentage copper Lead-tin solders have had a long history in joining electronic components. W ~ t h the miniaturization of components and the evolution of the circuit board to multitudes of circuits on single chips, exposure to the potentially damaging temperatures of the soldering operation became an increasing concern. F i u r e 4-5 reveals why 60-40 solder (60 wi% tin) became the primary joining material in the lead-tin system. Of all possible alloys, it has the lowest (all-liquid) melting temperature.

'

76

CHAPTER 4 Equilibrium Phase Diagrams and the lron-Carbon System

is cooled from 183"C,the material will go from a single-phase solution to a two-phase mixture as a tin-rich second phase precipitates from solution.This change in structure can he used to alter and control the properties in a number of engineering alloys

INSOLUBILITY If one or both of the components are totally insoluble in the other, the diagrams will also reflect this phenomenon. Figure 4-7 illustrates the case where component A is completely insoluble in component B in both the liquid and solid states. UTILIZATION OF DIAGRAMS Before moving to more complex diagrams,let us first return to a simple phase diagram, such as the one in figure 4-8, and develop several useful tools. For each point of temperature and composition, we would like to obtain three pieces of information:
1. The phasespresent. The stable phases can be determined by simply locating the point of consideration on the temperature-composition mapping and identifying the region of the diagram in which the point appears. 2. The composition of eachphase. If the point lies in a single-phase region, there is only one component present, and the composition (or chemistry) of the phase is simply the composition of the alloy being considered. If the point lies in a two-phase region, a tieline must be c0nstructed.A tie-line is slmply an isothermal (constant-temperature) line drawn through the point of consideration, terminating at the boundaries of the singlephase regions on either side.The compositions where the tie-line intersects the neighboring single-phase regions will be the compositions of those respective phases in the two-phase mixture. For example, consider point a ~nFigure 4-&The tie-line for this temperature runs from S, to L2.The tie-lie intersects the solid-phase region at point S,. Therefore, the solid in the two-phasemixture at point a has the composition of point S2. Since the other end of the tie-line intersects the liquid reglon at &, the liquid phase that is present at point a will have the composition of point L2.

3. The amount of each phase present. If the point lies in a single-phase region, all of the material, or 100%. must be of that phase. If the point lies in a two-phase region, the relative amounts of the two components can be determined by a lever-law calculation using the previously drawn tie-line. Consider the cooling of alloy X in Figure 4-8 in a manner sufficiently slow so as to preserve equilibrium. At temperatures above t , the material ism a single-phase liquid state.Temperature t i , therefore, is the lowest temperature at which the alloy is 100% liquid. If we draw a tie-line at this temperature, it runs from S, to L, and lies entirely to the left of composition X. At temperature t3, the alloy is

I
Liquid A

Liquidus

I
8 .

Liquid solution

+ Liquid B

Freezing.

temp.^ 1
Sol~d A+Liquid B

Freezing temp. B

Solid A + Solid B

I
A

I
B

Composition (weight percent 6)

FIGURE 4-7

Composition - % B
FIGURE 4-8

Equilibrium diagram o f two materials that are completely insoluble in each other in both the liquid and solid states.

Equilibrium diagram showing the changes that occur during the cooling of alloy X.

completely sohd, and the tie-line lies completely to the right of compositlon X. As the alloy cools from temperature t , to temperature t,, the amount of solld goes from zero to 100% while the segment of the tie-lrnc lhat Ires to the rrght of compos~tion X a190 goes from zero to 100%. S~milarly, the amount of liquid goes from 100"' to zero as the segment of the tie-line lying to the left of compositlon X undergoes a s~mdar change. Extrapolating these observations to intermediate temperatures, such as temperature fa, we pred~ct that the fraction of the tie-lme that lies to the left of polnt a corresponds to the fraction of the material that 1s liqu1d.This fractlon can be computed as:

r values in we~ght percent B. In a where the values of a, S,, and L2 are t h e ~ composition similar manner, the fraction of solid corresponds to the fraction of the tte-line that hes to the right of point a . (Nute:?hese mathemat~cal relations could be r~gorously denved from the conservation of either A or B atoms, as the material dlvides into the two difthe calculat~ons consider the t~e-line as a lever ferent compositions of S z and L, ) S~nce with the phases at each end and the fulcrum at the composit~online, they are called lever-law calculations. Equilibr~um phase diagrams can also he used to provide an ovcrall picture of an alloy system or to Identify the transition points for phase changes in a given alloy. For example, the temperature required to redissolve a second phase or melt an alloy can be eas~ly determ~ned.The varlous changes that will occur dnrlng the slow heat~ng or slow cooling of a materlal can now be predicted. In fact, most of the questrons posed at the beginn~ng of this chapter can be answered.

SOLIDIFICATION OF ALLOY X Let us now apply the tools that we have just developed, tie-lines and lever laws, to follow the solidification of alloy X in Figure 4-8. At temperature rl, the first minute amount of solid fonns with the chemistry of point S, . As the temperature drops, more solid forms, but the chemistries of both the solid and liquid phases shift to follow the tie-line endpoints.The chemistry of theliquid follows the liquidus line, and the chemistry of the solid follows the solidus. Finally, at temperature t 3 , solidification is complete. and the composition of the single-phase solid is now that of alloy X, as required. The composition of the first solid to form is different from that of the final solid. If the cooling is sufficiently slow, such that equilibrium is mainpained or approximated, the composition of the entire mass of solid changes during cooling and foUows the endpoint of the tie-line.These chemistry changes are made possible by diffusion, the process in which atoms migrate through the crystal lattice given sufficient time at elevated temperature. If the cooling rate is too rapid, however, the temperature may drop before sufficientdiffusion occnrsTheresultant material will have a nonunifom chemistry.The initialsolid that formed will retain a chemistry that is different from the solid regions that form later. When these nonequilibrium variations occur on a microscopic level, the resultant struclure is referred to as being cored. Variation on a larger scale is called mcruseg~-egation. THREE-PHASE REACTIONS Several of the phase diagrams that were presented earller contam a feature in which phase regions are separated by a horizontal (or constant-temperature) liie.These lines are further characterized by e~ther a V intersecting from above or an inverted-V intersecting from below The intersect~on of the V and the hne dcnotes the location of a threephase equilthrium reactron One common type of three-phase reaction, known as a eutecnc,has already been observed in Figures4-4 and 4-5. It is posstble to understand these reactions through use of the tie-line and lever-law concepts that have been developed. Refer to the lead-tin dlagram of Frgure 4-5 and consider any alloy containmg between 19 2 and 97.5 wt% tm at a temperature just above the 183Chorizontal line Tie-line and lever-law computations reveal that the materlal contains elther a lead r~ch or tin-rich sohd and remaimng liquid.At thrs temperature, any liquid that 1s present will have a composition of 61 9 wt% tm. regardless of

78

C H A ~4 R Equlltbrium Phase Dlagrams and the lron-Carbon System

the overall composition of the alloy. Iwe now focus on this liquid and allow it to cool to just below 183"C,a transition occurs in which the liquid of composition 61.9% bn transforms to a murture of lead-nch solid with 19.2% tin and tin-rich solid containing 97.5% tin. The three-phase reaction that occurs upon coohng through 183Ccan be written as. Iiquid619%
~n

IR3"C

Note the similarity to the very simple chemical reaction in whlch water dissociates, or separates, into hydrogen and oxygen: H,O H, f oz Since the two solids ~nthe lead-tin eutectic reaction have chemistries on either side of the original hquid, a similar separation must have occurred. Any chem~cal separation requlres atom movement, but the distances lnvolved in a eutectic reaction cannot be great. The resulhng rtructure, known as eutectic structure, will be an ~ntimate mlxture of the two single-phase solids, with a multitude of interphase boundaries. For a given reaction, the eutectic structure always forms from the same chemistly at thesame temperature and therefore has i t 9 own characteristicset of physical and mechanical properties.Alloys \nth the eutectic compowtion have the lowest meltmg point of all neighboring alloys and generally possess relatively high strength. For these reasons, they are often used as casting alloys or as filer materlal m boldering or brazing operations. Theeutectic reaction can be written m the general form of:

ru192ah /,sn

+ PY~s./,s~

liquid ,solid,

+ solld,

F~gure 4-9 summarizes the various forms of thrce-phase reactions that may occur in engineering systems, along with the generic description of the reaction shown below the figures These include the peritectic, monotectcc, and syntectrc reactions, where the suffix -LCdenotes that at least one of the tluee phases in the reaction is a llquid If the same prefix appears with an -oid suffix, the reaction is of a similar form but all phases lnvolved are solids.Two such reactions are the eutectord and the peritectoid. The separation eutectoid produces an extremely fine two-phase mixture, and the combination peritectoid reaction is very sluggish since all of the chemistry changes must occur withim (usually crystalltne) solids.

IL, ,S ,

Monotectlc + Lz)

Syntectic
IL,+L*iS1)

IS,

Peritectoid + 5 -B $3)

Sto~ch~ometric ~nterrnetallic compound

Nonstoichiometric
lnterrnetalllc compound

FIGURE 4-9

Schemattc summary of three-phase reacttons and intermetallic compounds.

determine rhe snecific form n E a three-nhme

reaclioo.locate its horizontal line and the V intersectinp.

to their &n&ic forms presented below the figures,remcmberiogthat Greek letters denote single-phase 8alids.

INTERMETALLIC COMPOUNDS A final phase diagram feature occurs in alloy systems where the bonding attraction between the component materials is strong enough to form stable compounds.These compounds are single-phase solids and tend to break the diagram into recognizable subareas. If components A and B form a compound, and the compound cannot tolerate any deviation from its fixed atomic ratio, the product is known as a stoichiomerric intermetallic compound and it appears as a singlevertical line in the diagram. (Note:%s will be seen for t h e F e C iron carbide at 6.67 wt.% carbon in the upcoming iron-carbon equilibrium diagram.) If some degree of chemical deviation is tolerable, the vertical lme expands into a single-phase region, and the compound is known as a nofisioichiometric intermetallic compound. Figure 4-9 shows schematic representations of both stoichiometric and nonstoichiometric compounds. Intermetallic compounds appear as single phases in the middle of equilibrium diagrams, with locations consistent with whole-number atomic ratios, such as AB, A2B, AB2, A3B, AB,, and so In general, they tend to be hard, brittle materials, since these properties are a consequence of their ionic or covalent bonding. If they are present in large quantities or lie along grain boundaries in the form of a continuous film, the overall alloy can be extremely brittle. If the same compound is dispersed throughout the alloy in the form of small discrete particles, the result can be a considerable strengthening of the base metal. COMPLEX DIAGRAMS The equilibrium diagrams for actual alloy systems may be one of the basic types just discussed or some combination of them. In some cases the diagrams appear to be quite complex and formidable. However, by focusing on a particular composition and analyzing specific points using the tie-lie and lever-law concepts, even the most cnmplex diagram can be interpreted and understood. If the properties of the various components are known, phase diagrams can be used to predict the behavior of resultant structures.

1 4.4 IRON-CARBON EQUILIBRIUM DIAGRAM

Steel, composed primarily of iron and carbon, is clearly the most important of the engineering metals. For this reason, the iron-carbon equilibrium diagram assumes special importance. The diagram most frequently encountered, however, is not the full iron-carbon diagram but the iron-iron carbide diagram shown in Figure 4-10. Here, a
"C
"F

FIGURE 4-10 The iron-carbon equilibrium phase diagram. Single phases are a, ferrite; y,

Weight percent carbon

'The use of "weight percent" along the horizontal axis tends to mask the whole~numberatomic ratio of intemetaUic compounds. Many equilibrium phase diagrams now include a second horizontal scale to reflect "atomicpercent? Intemetaiiic compounds then appear at atomic percents of 25,33,50.67.75. and similar values that reflect whole-number atomic rsfias.

80

CHAPITR 4 Equilibrrum Phase Diagrams and the Iron-Carbon System

stoichiometr~c intermetallic compound, Fe3C, is used to terminate the carbon range at 6.67 wt% carbon.The names of key phases and structures, and the specific notation5 used on the diagram, have evolved historically and will be used m their generally accepted form. There are four single-phase solids wlthin the d~agam.Three of these occur m pure iron, and the fourth 1s the Iron cartnde mtermetalllc that forms at 6.67% carbon. Upon coohng,pure Iron solidifies into a body-centered-cnlncsolid that is stable down to 1394C (2541F). Known as delta-fernte, this phase 1s present only at extremely elevated temperatures and has llttle engineering importance. From 1394Oto 912C (2541" to 1674F) pure iron assumes a face-centered-cubic structure known as austenzte m honor of the famed metallurgist Roberts-Austen of England Designated by the Greek letter y , austenite exhiblts the h ~ g h formability that is characteristic of the face-centered-cubic structure and is capable of mssolving over 2% carbon m smgle-phase sohd solution. Hot forming of steel takes advantage of the low strength, high ductility, and chemical nnlformlty of austenite. Most of the heat treatments of steel begin by forming the hlghtemperature austenite structure. Alpha-fernte, or more commonly just ferrrte, 1s the stable form of Iron at temperatures below 912C (1674C). This body-centcred-cub~c structure can hold only 0.02 wt% carbon in solid solution and forces the creauon of a two-phase mlxture m most steels. Upon further coollng to 770C ( 1 4 1 8 " F ) , iron undergoes a transition from nonmagnetic to magnetic.The temperature of this transition w known as the Curie point, but because it IS not assoc~ated with any change in phase (but 1s an atomic-level transibon), it does not appear on the equilibrium phase diagram. The fourth single phase is the stoichiometrlc ~ntermelallic compound Fe,C, which goes by the name cementite,or iron-carbide. L i e most mtermetaIlics, it is quite hard and brittle, and care should be exercised In controlling the structures in which it occurs Alloys with excessive amounts of cementite, or cementite in undesirable form, tend to have brittle characteristics Because cementite dissociates prior to melting, its exact melting polnt is unknown, and the hquidus line remains undetermined in the high-carbon region of the diagram. Three d~shnct three-phase reactions can also be identified. At 14YjbC(27U0F),a peritectlc reaction occurs for alloys with a low weight percentage of carbon. Because of its high temperature and the extensive smgle-phase austemte reglon immediately below it, the pentectic reaction rarely assumes any engineer~ng wgnificance. A euteckc is of4.3% carbon All alloys conohserved at 1148'C (209SF),with the eutectic compos~tion taining more than 2.11 % carbon w~ll experience the eUteChc reaction and are classified by the general term cast irons. The final three-phase reactlon is a eutectoid at 727'C (1341F) w ~ t h a eutectoid composition of 0.77 wt% carbon.Alloys w~thless than 2.11% carbon miss the eutectic reachon and form a two-phase mixture when they cool through the eutectoid.These alloya are known as steels. The point of maximum soluhlity of carbon m iron, 2 1 1wt%, therefore,forms an arbitrary separation between steels and cast irons

If we focus on the materials normally known as steel, the phase diagram of Figure 4-10 can be snnplified considerably. Those portions near the delta phase (or pentectic) regon are of little significance, and the h~gher-carbon region of the eutectic reactlou only applies to cast irons. By deleting these segments and focusing on the eutectoid reactlon, we f Frgure 4-1 1 to provide an understand~ng of the propcan use the scmpl~fied diagram o erties and processing of steel. Rather than beginning with liquid, our cons~derations highgenerally begin w ~ t h temperature, face-centered-cub~c, single-phase austen~te. The key translhon will be the conversion of austenlte. to the two-phase ferrite plus carbide mixture as the temperature I drops. Control of this reaction, which arises as a result of the drastically d~fferent carbon solubilities of the face-centered and body-centered structures, enables a wide range o f properties to be achieved through heat treatment. To begin to understand these processes, consider a steel of the eutectoid compc- 1 sltion, 0.77% carbon, berng slow cooled along hne x-x' m Flgure 4-ll.At the upperl

Liquid

cementite

he
IY, bit tot
n 1 -

bhe
lic fa
311

0.77

1.0

is
FIGURE 4-11

%Carbon

lUt

Simplified iron-carbon phase diagram with labeled regions. Figure 4-10 shows the more standard Creek letter notation.

FIGURE 4-12

Pearlite; 1000X. (Courtesy o f United States Steel Corp., Pittsburgh, PA.)

temperatures, only austenite is present, with the 0.77% carbon being dissolved in solid solution within the face-centered structure. When the steel cools through 727'C (1341mF), several changes occur simultaneously. The iron wants to change crystal structure from the face-centered-cubic austenite to the body-centered-cubic ferrite, but the ferrite can only contain 0.02% carbon in solid solution. The excess carbon is rejected and forms the carbon-rich intermetallic known as cementite. The net reaction at the eutectoid, therefore, is: austeniteon,"/.c:FCC

ferrite,,msh c:BCC

+ cementite,,,,

Since the chemical separation occurs entirely within crystalline solids, the resultant structure is a fine mixture of ferrite and cementite. Specimens prepared by polishing and etching in a weak solution of nitric acid and alcohol reveal a lamellar structure composed of alternating layers or plates, as shown in Figure 4-12. Since it always forms from a fixed composition at a fixed temperature, this structure has its own set o C characteristic properties (even though it is composed of two distinct phases) and goes by the name penrlite because of its metallic luster and resemblance to mother-of-pearl when viewed at low magnification. Steels having less than the eutectoid amount of carbon (less than 0.77%) are called hypoeutectoid steels (hypo means "less than"). Consider the cooling of a typical hygure 4-1 1. At high temperatures the material is er, it enters a region where the stable phases are law calculations show that the low-carbon ferrite nucleates and grows, leaving the remaining austenite richer in carbon. At 727C (1341F), the remaining austenite will have assumed the eutectoid composition (0.77% carbon); and further cooling transforms it to pearlite.The resulting structure,therefore, is a mixture of primary or proeutectoid ferrite (ferrite that forms before the eutectoid reaction) and regions of pearlite as shown in Figure 4-13, Hypereutectoid steels (hyper means "greater than") are those that contain more than the eutectoid amount of carbon. When such a steel cools, as along line Z-z' in Figure 4-11, the process is similar to the hypoeutectoid case, except that the primary or proeutectoid phase is now cementite instead of ferrite. As the carbon-rich phase nucleates and grows, the remaining austenite decreases in carbon content, again reaching the

8;

Cmmm 4

Equlllbrium Phase D~agrams and the Iron-Carbon System eutectoid composition at 727C (1341F).As before this austenite transforms to pearlite upon slow coolinl through the eutecto~d temperature Figure 4-14 is : photonucrograph of the resulting structure, whch con sists of primary cementite and pearlite, In this case t h ~ continuous network of primary cementite (an inter metallic) will cause the material to be extremel! brittle. It should be noted that the tram~tions just de scribed are for equilibrium conditions, which can b~ approximated by slow cooling Upon slow heating the transitions will occur in the reverse manner When the alloys are cooled rap~dly, however, entirely different results may be ob tained, since sufficient time may not be provided for the normal phase reactions t c occur. In these cases, the equ~hhrmm phase diagram is no longer a v a l ~ d tool for en gineering analysis. Since the rapid-cool processes are important in the heat treatmen of steels and other metals, their characteristics will be discussed in Chapter 5, and nev tools will be introduced to aid our understanding. Steels and other ferrous metals, in cludlng stainless steels and tool steels, will be further developed in Chapter 6.

FIGURE 414 Photomicrograph o f a hypereutectold steel showlng primary cementite along graln boundaries, 5OOX. (Courtesy of United States Steel Corp, Pittsburgh, PA)

Iron-carbon alloys with more than 2.11% carbon experience the eutcctic reaction dur lug cooling and are known as cast zrons.The term cast tron applies to an entire famil: of metals with a wide variety of properties Being relatwely inexpensive, with good flu idity and rather low liquidus temperatures, they are readily cast and occupy an impor tant place in engineering application5 Most commercial cast irons also contain a signiticant amount of si1icon.A typica cast iron contams 2.0 to 4 0% carbon, 0 5 to 3.0% s~licon, less than 1.0% manganese and less than 0.2% sulfur. Silicon produces several major effects. First, it partially sub stitutes for carbon, so that use of the equilibrium phase diagram requires replacing t h ~ weight percent carbon scale with a carbon equivulmt. Several formulat~ons exist to com pute the carbon equivalent, with the simplest being the weight percent carbon plus one thud the weight percent sillcon, carbon equivalent (CE) = (wt % carbon)

+ f (wt % silicon)

The high sll~con enhances the oxidation and corrosion resistance of cast irons b. promoting the formation of a tightly adhering surface oxide. Silicon also tends to pro mote the formation of graphite as the carbon-rich single phase instead of the Fe,( mtermeta1lic.The eutectic reaction now has two dlstinct possibilities, as indicated in thi modified phase diagram of Figure 4-15: liquid liquid austenite + austenite
+

+ Fe3C

+ graphite

1 The final microstructure of cast lron,thereforq has two possible extremes: (1) a of the carbon-rich phase being intermetalllc Fe,C and (2) all of the carbon-rich phast being gruphrte. In practice, both of these extremes can be approached by controlling th~ chemistry and other process variables. Graphite formation is pl-omoted by slow coolinp high carbon and silicon contents, heavy or thick section sizes, inoculation practices, an1 the presence of sulfur, phosphorus, aluminum, magnesium, antimony, tin, copper, nick el, and cobalt. Cementite (Fe,C) formation is favored by fast cooling,low carbon and s11 icon levels, thin sections, and alloy additions of titanium, vanadium, zirconium, chromum manganese, and molybdenum.

TYPES OF CAST IRON

Various types of cast iron can be produced, depending on the chemical composition cooling rate, and the type and amount of inoculants that are used (Inocnlants ant

SF.CTION 4.6 Cast Irons

83

- - -. FB-graphite

Fe-Fe3 C diagram diagram

------------------y + graphite

a + carbide

738"
727"

An iron-carbon diagram showing two possible high-carbon phases. Solid lines denote the iron-graphite system; dashed lines denote iron-cementite (or iron-carbide).
FIGURE 4-15

600

2 3 4 Weight percent carbon or carbon equivalent

= 6

can usually be obtained if one rubs a finger across the fracture. On a more positive

The size, shape, and distribution of the graphite flakes have a considerable effect on the overall properties of gray cast iron. When maximum strength is desired, small, uniformly distributed flakes with a minimum amount of intersection are preferred. A more effective means of controlling strength, however, is through control of the metal matrix structure, which is in turn controlled by the carbon and silicon contents and the cooling rate of the casting. Gray cast iron is normally sold by clms,with the class number corresponding to the minimum tensile strength in thousands of pounds per square inch. Class 20 iron (minimum tensile strength of 20,000 psi) consists of highcarbon-equivalent metal with a ferrite matrix. Higher strengths, up to class 40, can be obtained with lower carbon equivalents and a pearlite matrix. To go above class 40, alloying is required to provide solid solution strengthening, and heat-treatment practices must be performed to modify the matrix. Gray cast irons can be obtained up through class 80, but regardless of strength the presence of the graphite flakes results in extremely low ductility. Gray cast irons offer excellent compressive strength (compressive forces do not promote crack propagation, so compressive strength is typically 3 4 times tensile strength), excellent machinability (graphite acts to break up the chips and lubricate contact surfaces),good resistance to adhesive wear and galling (graphite flakesself-lubricate),and outstanding sound- and vibration-dampingcharacteristics (graphite flakes absorb transmitted energy). Table 4-1 compares the relative damping capacities of various engineering metals, and clearly shows the unique characteristic of the high-carbon-equivalent gray cast

84

CHAPILR 4 Equll~brlum Phase Dragrams and the Iron-Carbon System

, ... .
%

. . .. ...:,, ,,. , . . . . Relative GrniincjCabacity of Var;ous Me1a.s ,%'..:. . .. . . .

.
'

Material
Gray iron (high carbon equivalent) Gray iron (low carhon equivalent) DucLile iron Malledble iron White iron
Stcel

Dam~ine Ca~acitv"
10&500
20-1 00

Aluminum
"Natural log of the ratio olrucccsaiue amplitudes

.S20 %I5 2-4 4 0.4

FIGURE 4-17

of

iron showing the irregular graphite spheroids, here in a ferrite matrix. (Courtesy of Iron Castings Society, Rocky River, OH.)

irons (20-25 limes better than steel and 250 times better than aluminum!). High silicon contents promote good corrosion resistance and the enhanced fluidity desired for casting operations. For these reasons, coupled with low cost, high thermal conductivity, low rate of thel-ma1expansion, good stiffness, resistance to thermal fatigue, and 100% recyclability, gray cast iron is specified for a number of applications, including automotive engine block& heads, and cylinder liners: transmission housings: machine tool bases; and large equipment parts that are subjected to conlpressive loads and vibrations. White cast iron has all of its excess carbon in the [orm of iron carbide and receives its name from the white surface that appears when the material is fractured. Features promoting its formation are those that favor cementite over graphite: a low carbon equivalent (1.8 to 3.6% carbon, 0.5 lo 1.9% silicon. and 0.25 to 0.8% manganese) and rapid cooling. Because the large arnounl of iron carbide dominates the microstructure, white cast iron is very hard and brittle, and finds applications where high abrasion resistance is the dominant requirement. For these uses it is also common to pursue the hard, wearresistant marterzsite structure as thc metal matrix. (Note: This structure will be described in Chapter 5.) In this way, both the metal matrix and the high-carbon second phase contribute to the wear-resistant characteristics of the malerial. White cast iron surfaces can also be formed over a base of another material. For example: mill rolls that require extreme wear resistance may have a white cast iron surface over a steel interior. Accelerated cooling rates produced by tapered sections or metal chill bars placed in the molding sand can be used to produce white iron surfaces at selected locations o l a gray iron casting. Where regions of white and gray cast iron occur in the same component, there is generally a transition region comprised of both whitc and gray irons, known as the mottled zone. When white cast iron is exposed to an extended heat treatment at temperatures in the range of 900C (1650"F), the cementite will dissociate into its component elements, and some or all of the carbon will be converted into irregularly-shaped nodules of graphite (also referred to as clump or popcorn graphite).The product, known as malleable cast iron, has significantly greater ductility than that of gray cast iron because the more favorable graphite shape removes the internal notches.The rapid cooling required to produce the starting white iron structure restricts the size and thickness of malleable iron products such that most weigh less than 5 kg (10 lb). Various types of malleable iron can be produced, depending on the type of heat treatment that is employed. If the white iron is heated and held for a prolonged time just below the melting point, the carbon in the ccmentite converts to graphite (first-stage graphitization). Subsequent slow cooling through the eutectoid reaction causes the carboncontaining austenite to transform to ferrite and more graphite (second-stage graphitizaon).The resulting product, known as ferritic malleable cust iron, has astructure of irregular particles of graphite dispersed in a ferrite matrix (Figure 4-17), Typical properties would be: 10%elongation;35-ksi (240-MPa) yield strength: 50-ksi (345-MPa) tensile strength: and excellent impact strength, corrosion resistance, and machinability. The heat-treatment times, however,are quite lengthy, often involving over 100 hours at elevated temperature. If the material is cooled more rapidly through the eutectoid transformation, the carbon in the austenite does not form additional graphite but is retained in a pearlite or

hECTIoN 4.6 Cast Irons

a>

martensite matrix. The resulting pearlitic malleable cast iron is characterized by higher 1 '0 strength and lower ductility than its ferritic counterpart. Properties range from 1 to 4 elongation, 45- to 85-ksi (310- to 590-MPa) yield strength, and 65- to 105-ksi (450- to 725-MPa) tensile strength, with reduced machinability compared to the ferritic material. The modified graphite structure of malleable iron provided quite an improvement in properties compared to gray cast iron, but it would be even more attractive if it could be obtained directly upon solidification rather than throueh a oroloneed heat treatment terials can promote graphite formation and change the morphology (shape) of the graphite product. If ferrosilicon is injected into the melt (inoculation),it will promote the formation of graphite. If magnesium (in the form of an MgFeSi or MgNi alloy) is also added just prior to solidification, the graphite will form as smooth-surface spheres.The latter addition is known as a nodulizer, and the product becomes ductile or nodular cast iron. Subsequent control of cooling can produce a variety of matrix structures, with ferrite or pearlite beingthe most common (Figure 4-18). By controlling the matrix structure, properties can be produced that span a wide range from 2 to 18% elongation, 40to 90-ksi (275- to 620-MPa) yield strength, and 60- to 120-ksi (415- L o 825-MPa) tensile strength. The combination of good ductility, high strength, toughness, wear resistance, machinability, low-melting-point castability, and up to a 10% weight reduction compared to steel makes ductile iron an attractive engineering material. High siliconmolybdenum ductile irons offer excellent high-temperature strength and good corrosion resistance. Unfortunately, the costs of a nodulizer, higher-grade melting stock, better furnaces, and the improved process control required for its manufacture combine to place it among the most expensive of the cast irons. Amtempered ductile iron (ADI), ductile iron that has undergonth special austempering heat treatment to modify and enhance its properties,4has emergeil as a significant engineering material. It combines the ability to cast intricate shapes witdstrength, fatigue, and wear-resistance properties that are similar to those of heat-treated steel. Compared to conventional as-cast ductile iron, it offers nearly double the strength at the same level of ductility.Compared to steel, it also offers an 8 to 10% reduction in density (so strengthto-weight ratio is excellent) and enhanced damping capability,both due to the graphite nodules, but generally poorer machinability and with about a 20% lower elastic modulus. Table 4-2 compares some typical mechanical properties of malleable and ductile irons with the five grades of austempered ductile cast iron that are specified in ASTM Standard A-897. Compacted graphite cast iron (CGI) is also attracting considerable attention. Produced by a method similar to that used to make ductile iron (an Mg-Ce-Ti addition is made), compacted graphite iron is characterized by a graphite structure that is intermediate to the flake graphite of gray iron and the nodular graphite of ductile iron, and it tends to possess some of the desirable properties and characteristics of each.Table 4-3 shows how the properties of compacted graphite iron bridge the gap between gray and ductile. Strength, stiffness, and ductility are greater than those of gray iron, while castability, machinability, thermal conductivity, and damping capacity all exceed those of ductile. Impact and fatigue properties are good. ASTM Specification A842 identifies five grades of CGI-250,300,350,400, and 450-where the numbers correspond to tensile strength in megapascals. Areas of application tend to be those where the mechanical properties of gray iron are insufficient and those of ductile iron, along with its higher cost, are considered to be overkill. More specific,compacted graphite iron is attractive when the desired properties include high strength, castability, machinability,thermal conductivity,and thermal shock resistance.

'The austempering process begins by heating the metal to a temperature between 1500eand 1750F (815" to 955C) and holding for sufficient time to saturate the austenite with carbonThe metal is then rapidly cooled to an austempering temperature between 450" and 750F (230"-400"C), where it is held until allcrystal structure changes have completed, and then cooled to room temperature. High austempering temperatures give good toughness and fatigue properties, while lower austempering temperatures give better strength and wear resistance.

86

GHAPI-ER 4

Equilibrium Phase Dlagrams and the Iron-Carbon System

Class or Grade Malleable Irona

M3210 M4504 M5003 M5503 M7002 M8501

M~nimum Y~eld Strength

ksi

M~nlmum Tens~le Strength

ksl

MPa

MPa
50 65 75 75 90 105 345 448 5 17 517 621 724

M~rumum

Percentage Elongation

Brine11 Hardness Number

32 45 50 55 70 85

224 310 345 379 483 586

10

156 max 163-217 187-241 187-241 229-269 269-302

Ductile Ironb
6030-18 6545-12 80-5046 100-?M)3
12Q-9M12

40 45 55 70

90
80
100 125 155

276 310 379 483 621

60 65 80 100 120

414

18
12 F
6

448
552 689 827

149-187 170-207 187-248 217-269 240-300

Austempered Duct~le Iron'

1 2 3 4 5

550
700 850 1100 1300

185

125 150 175 200 230

850 1050 1200 1400

1600

10 7 4 1 -

269-321 302-363 341444 388-477 444-555

'ASTM SpeclflcahonAh02 (Also SAE 1158)

%STM SoeaficatlonA536 'ASTM Specrf~~atlon~897

Property
Tens~le strength (MPa) Elastic modulus (Gpa) (%) Elangat~on Thermal conductmty (wlmk) Relative darnp~ng capaclty (Gray = 1)

Gray
250 205 0 48 1

CGI
450 145 15 37 035

Ductrle
750 I60 5 28

0 22

THE ROLE OF PROCESSING O N PROPERTIES

While typical propertles have been presented for the various types of cast iron, it should be noted that the properties of all metals are influenced by how they are processed. For cast materials, propertles wdl vary wlth the manner of solidification and cooling. Because cast components often have complex geometries, the cooling rate may vary from locatlon to location, with compamon varlation in properties. To assure comphance with industry specificat~ons, standard geometry test bars are often cast along with manufactured products so the material can be evaluated and properties ensured independent of product geometry.

Review Questions

I IKey Words
austcmpered ductile iron austenite carbon equivalent cast iron cementite class compacted graphite complete solubility composition cooling curve cored structure ductile cast iron equilibrium eutectlc eutectlc structure cutectoid felrlte freezing range graphlte gray ca$t iron hypereutectoid hypoeutecto~d moculat~on ~nterfaccs mtermetall~c compound lever law llqurdus macrosegregatlon malleable cast iron martensite monotectrc mottled zone nodular cast Iron noduluer nonsto~ch~ometric pearhte perltectic per~tecto~d phase p h a ~ dragram e primary phase solidus solub~l~ty 11mt solute soIvus steel stoichiometr~c syntectlc three-phase reaction he-lme wh~te cast Iron

IReview Questions
1. What are some features that are useful in defining a phase? 2. Supplement the examples provided in tho text with another example of a single phase that is each of the following: continuous, discontinuous, gaseous, and a liquid solution. 3. What is an equilibrium phase diagram? 4. What three primary variables are generally considered in equilibrium phase diagrams? 5 . Why is a prssure-temperature phase diagram not that uscful for most engineering applications? 6. What is a cooling curve? 7. What features in a cooling curve indicate some form of change in a material's structure? 8. What is a solubility limit, and how might it be determined? 9. In general, how does the solubility of one material in another s increased? change as temperature i 10. Describe the conditions of complete solubility, partial solubility,and insolubility. 11. What types of changes occur upon cooling through a liquidus line'? A solidus line? A solvus line'? 12. What three pieces of information can be obtained for each point in an equilibrium phase diagram? 13. What is a tie-line? For what types of phase diagram regions would it be uselul? 14. What points on a tie-line are used to determine the chemistry (or composition) of the component phases? 15. What tool can be used to compute the relative amounts of the component phases in a two-phase mixture? How does this tool work? 16. What is a cored structure? Under what conditions is it produced'! 17. What features in a phase diagram can be used to identify threephase reactions? 18. What is the gencral fonn of a eutectic reaction? 19. What is the general form of the eutectic structure? 20. Why are alloys of eutectic composition attractive for casting and as filler metals in soldering and brazing? and how would 21. What is a stoichiometricintemetalliccompound, it appear in a temperaturecomposition phase diagram? How would a nonstoichiometric intermetallic compound appear? 22. What type o i mechanical properties would be expected for intermetallic compounds? 21. In what lorm(s) might inlermetallic compounds be undesirable in an engineering material? In what form(s) might they be 24. What are the four single phases in the iron-iron carbide diagram? Provide both the phase diagram notation and the assigned name. 25. What feature in theiron-carbon diagramis used to distinguish between cast irons and steels? 26. What features of austenite make it attractive for forming operations? What features make it attractive as a starting strncture for many heat treatments? 27. Which of the three-phase reactions in the iron-carbon diagram is most important in understanding the behavior of steels? Write this reaction in terms of the interacting phases and their composition. 28. Describe the relative ability of iron to dissolve carbon in solution when in the form af austenite (the elevated temperature phase) and when in the form of ferrite at room temperature. 29. What is pearlite? Describe its structure. 30. What is a hypoeutectoid steel, and what structure will it assume upon slow cooling? What is a hypereutectoid steel and how will its structure differ from that of a hypoeutectoid? 31. I n addition to iron and carbon,what other eleme$t is present ', in rather large amounts in cast iron? 32. What is a carbon equivalent and how is it comput~d? 33. What are the two possible high-carbon phases in cast irons? What features tend to favor the formation of each? 34. Describe the microstructure of gray cast iron. 35. Which of the structural units is generally altered to increase the strength of a gray cast iron? 36. What are some of the attractive engineering properties of gray cast iron? 37. What aresome of the key limitations to the engineering use of gray cast iron? 38. What is the dominant mechanical property of white cast iron? 39. What structural feature is responsible for the increased ductility and fracture resistance of malleable cast iron? 40. How is malleable cast iron produced? 41. What is unique about the graphite that forms in ductile cast iron? 42. What requirements of ductile iron manufacture are responsible for its increased cost over materials such as gray cast iron? 43. What are some of the attractive features of austempered ductile cast iron? 44. Compacted graphite iron has a structure and properties intzrmediate to what two other types of cast irons?

88

CHAPTER 4

Equilibrium Phase Diagrams and the Iron-Carbon System

II Problems
1. Obtain a binary [two-co~npgnent) phase diagram (or a system
not discussed in this chapter. Identify each: a. Single Phase b. Three-phase reaction c. Intermetallic compound 2. Identify at least one easily identified product or component that i s currently being produced from each of the following types of cast irons: a. Gray cast iron b. White cast iron c. Malleable cast iron d. Duclile east iron e . Compacted graphite cast iron 3. Find an example where one of the types of cast iron has been used in place of a previous material. What feature or features might have prompted the substitution?

I ( The Blacksmith Anvils

s an off~cer in the Western-America Blacksmdh Assockatlon, you have determined that a number of your members would like to have a modern equivalent of an 1870-vlntageblacksmith anvll. Your objective is to replicate the des~gn but utillre the advantageous features of today's engineering materials You hope ultimately to identify a producer who will make a limited number of these itemsfor sale and distribution through your monthly maganne.The proposed design Is a largeforglng anvil that has a total length of 20 inches The top surfaces must be resistant to wear, deformation, and chipping Estimated mechanical properties call for a yield strength in excess of 70 ksi, an elongation greater than 2%, and a Brlnell hardness ot200 or more on the top surface.You feel confident fhat you will be able to secure a minimum of 500 orders

1 Discuss the varlous properties that this part must possess to adequately perform its intended task

2. Discuss the various concerns that would influence the proposed method of fabricating the anv~is. 3 Assuming that the anvils be made from some form of ferrous metal, consider the properties of the various types of cast and steels regard to this appll. which material would you recommend? 4 How would you propose that a production run of 500 replica anvils be produced?

I
5.1 INTRODUCTION 5.5 S T ~ G T H E ~ HEAT NG 5.2 PROCESSING HEATTREATMENTS TREATMENTS
FOR STEEL

Processing Heat Peatments Heat Treatments for Nonferrous Metals 5 3 HEAT TREATMENTS USED TO INCREASE STRENGTH 5.4 STRENGTHENING HEAT TREATMENTS FOR NONFERROUS Preclp~tation or Age

Isothermal Transformation Diagram Tempering of Martensite Continuous Cooling

ans sf or mat ions

Jormny %st Tor Hardenability Hardmabihty Cons~deratlon~ Quench Med~a Design Concem, Residual Stresses,Distortion, and Crachng

~ e c h i ~ uto es Reduce Cracfung and D~stortlon Ausforming 5.6 SURFACE HARDENING OF STEEL Selective Heatlug Techniques Techniques Involving Altered Surface Chermstry 5 7 FURNACES FurnaceTypes and Furnace Atmospheres Furnace Contlols 5.8 HEAT TREATMENT AND ENERGY CLAW Case Study. A CARPENTER'S HAMMER

11 5.1 INTRODUCTION
In the prevlous chapters, you have been introduced to the rnterrelatlonship among the structure,properties,processing, and performance of engineering materials. Chapters 3 and 4 considered aspects of structure, while Chapter 2 focused on properties. In this chapter, we begin to expand on and Incorporate processing so that the structure and companion properties can be manipulated and controlled not by a single set of properties Many engineering materials can be characteri~ed but by an entire spectrum of possibilities that can be selected and varied at will. Heat treatment i u the term used to descrlbe the controlled heating and coolmg of nzaterzals for the purpose of altering thew ytructures and propcrtie~.The same matenal can he made weak and duct~le for ease in manufacture, and then retreated to provide high strength and good fracture resistance for use and application. Because both physical and mechanical properties (such as strength, toughness, machinahihty, wear resistance, and corrosion resistance) can be altered by heat treatmenl and these changes can be induced wlth no concurrent change in product shape,heat treatment is one of the most Important and widely used manufacturing processes, Technically, the term heat rreutment applies only to processes where the heatmg and cooling are performed for the specific purpose of altering properties, but heatmg and cooling often occur as incidental phases of other manuEacturing proceyses, such as hot forming or welding.The matenalproperties will be altered, however, just as though an intentional heat treatment had been performed, and the results can he either beneficial or harmful.For this reason, both the individual who selects materlal and the person who specdies its processmg must be fully aware of Lhe possible changes that can occur during heatlng or cooling actrvitie.s.Heat treatment should be fully integrated with other manufacturing processes IE effective results are lo he obtained.To provide a baslc understanding,this chapter wrllpresent both the theory of heat-treatment and a surveq of the more common heat-treatment processes. Bnce more than 90% of all heat treatment I? performed onsteel and other ferrous metals, these materials will receive the bulk of our attention.

M 5.2 PROCESSING HEAT TREATMENTS

The term heat treatment is often associated with those thermal processes that ~ncrease the strength of a matenal, but the broader definition permits inclusion of another set of processes that we will call procersing heat treatments.These are often performed as a means of

89

90

C H m5

Heslt Treatment preparing rhr. material for tabricallo~i.Specific ol$ccti\.cs mil! ht: t h e i~npro\erncnt of machiningchnracteristi~~ rhc reduction 01'fomting i ~ ~ r ~01. . cchi: s . rc>to~.ati~)n of ductility 10 enable further processing,

EQUILIBRIUM DIAGRAMS AS AIDS

&lost of the processing hcat treatments in\ol\'c r:~thcrslou cooling or csrcndcd timcs 31 eIc\.atcd temper;lrures. Thcsc condition5 tend t o approxi~uarc equilibrium. aliJ rhc resulting structurt?&tm\eieforeacan be reasonably predicted &ough the use of an aquilibriwraphme diagram (presented in Chapter 4). These diagrams can be used to determine the temperaturas that mud be attained to produce a desired starting structure, and to describe the changes that will then occur upon ~ubsequent cooling. It should he noted, however, that these diagrams are for true equil~briumcoudittons, and any departure from equilibrium may lead to substantially different results.

FRCJCESSING HEAT TREATMENTS FOR S.TEEL

Because many of the processmg heat treatments are applied to plain-carbon and lowalloy steels, they will be presentell here with the simpl~fied iron-carbon equiltbrium diagram of Figure 4-11s e m g as. a reference guideFigure 5-1 shows this diagram wrth the key transitim lines labeled in standardnotation. The eutectoid line is designated by the symbol Al, and A3 dmignates the boundary between austenite and ferrite + auatenite.' The tramition from austenite to austenite cementite. is designated as the A,, line. A number of process heat-treating operations have been classified under the general term of ann&alitng.Thesemay be emplayed to reduce strength or hardness, remove residual stresves, improve toughness, restore ductility, refine grain size, reduce segregation, ar alter the electrical or mapeticproperties of the material. By producing a certain desired stnreture,charactcristics can be imparted that will be favorable to subsequent operations (sueh as machining or forming) or apphcations. Because of the variety of anneals,it is impartant to designate the specific treatment,which is usually indicated by a precedmg adjective.The specific temperatures, cooling rate, and detatls of the process will depend on me material being treated and the objectives of the treatment. In the process of fudlannealing,hypoeutectoidsteels (less than 0.77% carken)are heated to 30" to 60% (50- to 10OL1F) above the A3 temperature, held for sufficient time to convert the structure to homogeneous sinde-ph-ase austenite of uniform composition and temperature, and then slowly aooled at a bonlrolled rate t h r o u ~ h the A , temperature. cooling 'ts usually done inthe furnace by decreasing the temperature b ; 10" to 30C (20" to 50F) per hour to a t least 3UWC (50'F) below the A , temperature.

Liquid

1148

Ferrite

Ausrenite

*
cementite

Slmpl~f~ed ~ron-carbonphase d~agram for steels wlth transttlon llnes labeled in standard notatlon as A,, A,, and A,,.
FIGURE 5-1

0 . 0 2

0.77 1.0
% Carbon

: Ilinuricall!, an :l; line uo-2 dppc?md between thz A . .snJ I - Ihi, It,:; du\~~,n.iled i n ? m;l~,etic pr,pertv chsn*~ k n u o n i ~ lhc s Curl? [r,ull Sinre lhls transition us?\nt:r slto\r,! 10 h : I n al.>mic chdnsc iuot a chdnge ~n phhw,thc line sar d;leled i:~hlt l l i ;qu:llhriuni ph.1-r. d u g - n l ui!h.,ut ;I ;rlmpnnlon ~~labelinq

SECTION 5.2

Processing Heat Treatments

At this point all structural changes are complete, and the metal can be removed from the furnace and air cooled to room temperature. The resulting structure is one of coarse pearlite (widely spaced layers or lamellae) with excess ferrite in amounts predicted by the equilibrium phase diagram. In this condition, the steel is quite soft and ductile. The procedure to full-anneal a hypereutectoid alloy (greater than 0.77% carbon) is basically the same, except that the original heating is only into the austenite plus cementite region (30' to 60C above the A,). If the material is slow cooled from the allaustenite region, a continuous network of cementite may form on the grain boundaries and make the entire material brittle. When properly annealed, a hypereutectoid steel will have a structure of coarse pearlite with excess cementite in dispersed spheroidal form. While full anneals produce the softest and weakest properties, they are quite time consuming,and considerable amounts of energy must be spent to maintain the elevated temperatures required during soaking and furnace cooling. When maximum sofmess and ductility are not required and cost savings are desirable, normalizing may be specified. In this process, the steel is heated to 60C (100F) above the A, (hypoeutectoid) or A,, (hypereutectoid) temperature, held at this temperature to produce uniform austenite, and then removed from the furnace and allowed to cool in still air.The resultant structures and properties will depend on the subsequent cooling rate. Wide variations are possible,depending on the size and geometry of the product, but fine pearlite with excess ferrite or cementite is generally produced. One should note a key difference between full annealing and normalizing. In the full anneal, the furnace imposes identical cooling conditions at all locations within the metal, which results in identical structures and properties. With normalizing, the cooling will be different at different locations. Properties will vary between surface and interior, and different thickness regions will also have different properties. When subsequent processing involves a substantial amount of machining that may be automated, the added cost of a full anneal may be justified, since it produces a product with uniform machining characteristics at all locations. If cold working has severely strain-hardened a meta1,it is often desirable to restore the ductility,either for service or to permit further processing without danger of fracture. This is often achieved through the recrystallization process described in Chapter 3.When the material is a low-carbon steel (<0.25% carbon), the specific procedure is known as a process anneal. The steel is heated to a temperature slightly below the A,, held long enough to induce recrystallization of the dominant ferrite phase, and then cooled at a desired rate (usually in still air). Since the entire process is performed at temperatures within the same phase region, the process simply induces a change in phase morphology (size, shape, and distribution).The material is not heated to as high a temperature as in the full-anneal or normalizing process, so a process anneal is somewhat cheaper and tends to produce less scaling. A stress-relief anneal may be employed to reduce the residual stresses in large steel castings, welded assemblies, and cold-formed products. Parts are heated to temperatures below the A, (between 550" and 650C or 1000" and 1200F),held for a period of time, and then slow cooled to prevent the creation of additional s t r e s ~ e s . ~ m and e s temperatures vary with the condition of the component, but the basic microstructure and associated mechanical properties generally remain unchanged. When high-carbon steels ( ~ 0 . 6 0 % carbon) are to undergo extensive machining or cold forming, a process known as spheroidization is often employed. Here the objective is to produce a structure in which all of the cementite is in the form of small spheroids or globules dispersed throughout a ferrite matrix.This can be accomplished by a variety of techniques,including (1) prolonged heating at a temperature just below the A, followed hy relatively slow cooling, (2) prolonged cycling between temperatures slightly above and slightly below the A,, or (3) in the case of tool or high-alloy steels, heating to 750" to 800C (1400" to 1500F) or higher and holding at this temperature for several hours, followed by slow cooling. Although the selection of a processing heat treatment often depends on the desired objectives, steel composition strongly influences the choice.Process anneals are restricted to low-carbon steels, and spheroidization is a treatment for high-carbon material.

92

CHAFER 5

Heat Treatment

Process anneal Craph~cal summary of the process heat treatmens for steels on an equ~l~br~urn dtagram.
FIGURE 5-2

500

02

04

06

08

10

12

14

16

% Carbon

Normallzing and lull annealing can be apphed to all carhon contents, but even here, preferences are noted Slnce d~fferent cooling rates do not produce a w d e variation of propertie? in low-carbon steels, the a n cool of a normali~ing treatment often produces acceptable uniform~ty. For higher carbon contents, such as the 0 4 to 0.6% range, different coolmg rates can produce w~der property variations, and the unilorm furnace cooling of a full anneal is often preferred. F~gure 5-2 provldes a graphical summary of the process heat treatments.

HEAT TREATMENTS FOR NONFERROUS METALS

Most of the nonferrous metals do not have the sign~ficant phase transif~ons observed in thc iron-carbon system, and for them, the process heat treatments do not play such a significant role. Aside irom the strengthening treatment of preclpitat~on hardening, whlch 1s discussed later,the nonferrous metals are usually heat treated for three purposes: (1) to produce a nn~form, homogeneous structure, (2) to provide stress relief, or (3) to brmng aboutrecrystall~zation. Castings that have been cooled too rap~dly can possess a segregated sohdlt~cation structure known as coring (d~rcussed more fully in can be achieved by heatmg to moderate temperatures and Chapter 4). ~Iomogenrzahun then holdlng for a sufficient tlme to allow thorough d~ffuslon to take place. S~m~larly, heat~ng for several hours at relatively low temperatures can reduce the Internal stresses that are often produced by forming, weldmg, or brazing. Recrystnllczutcon /discussed In Chapter 3) is a function of the particular metal, the amount of prior straining, and the desired rec~ystalllzation tlme. In general, the more a metal has been strained, the lone1 the recrystall~zal~on temperature or the sl~orter the time. W~thout pnor stralnIng, however, recrystallization w~ll not occur and heating will only produce undes~rable grain growth.

Six major mechanisms are available to increase the strength of metals: 1. Solid-solut~on strengthemng 2. Stram harden~ng 3. Grain slze refinement 4. Precipitat~on harden~ng 5. Dlaperslon hardening 6. Phase transformahons

SECTION 5 . 4

Strengthen~ngHeat Treatments for Nonferrous Metals

93

All of these can be Induced or altered by heat treatment, but all may not be apphcable to a specificmetal or alloy. In ~olzd-~olution strengthenmg, a base metal dissolves other atoms, either as r ~ l h stztutiunal solutcons, where the new atoms occupy site? In the host crystal latt~ce, or as rnterstrtralsolut~ons,where the new atoms squeeze into "holes"hetween the atom? of the base latt~ce.The amount of strengthemng depends on the amount of dissolved solute and the size d~fference of the atoms mvolved. Since distortion of the host Ttructure makes dlslocat~on movement more d~fficult, the @cater the size d~fference, the mole effective the addition Stratn hardening (discussed in Chaptei 3) produces an increase In strength by means of plastic deformation under cold-workmg condit~ona. Because grain boundar~es act as barr~ers to d~slocat~on motlon, a metal wlth small grains tends to be stronger than the same metal with larger grains, Thub gram size refinement can be used to Increase strength, except at clevated temperatures, where grain growth can occur and gram boundary diffusion contributes to weep and fa~lure It is important to note that gram s ~ z e refinement is one of the few processes that can Improve 5irength withoot a companion loss of duct~lity and toughnes5 Inpreclpitatmon hardening, or age hardenmg, strength 1s obtained from a nonequllibrium structure that is produced by a three-step heat tleatment. Detads of t b ~ method s will be provided in Section 5.4 Strength obtamed by dispersing second-phase particle? throughout a base material is known as dispersion hardentng. To be effective, the disper~cd particles should be stronger than the matrix, adding strength through both their reinforcing action and the ddditlonal ~nterfac~al surfaces that prcsent barriers to dislocat~on movement. Phase transformat~on strengthening involves those alloys that can be heated to form a singlc phase at elevated temperature and subsequently hansform to one or more low-temperature phases upon cooling, When t h ~ feature s is used to increarc strength, the cooling is usually rapid and the phases that are produced are usually of a nonequiltbrium nature.

5.4 STRENGTHENING HEAT TREATMENTS FOR NONFERROUS METALS

All six of the mechanisms just described can he uscd to increase the strength of nonferrous metals Solid-solution strengthening can impart strength to single-phase materials. Strain hardening can be quite useful if sufficient ductility 1s present. Alloys containmy eutectic structure exhiblt conslderahle disper~lon hardening. Among all of the posslb~lities, however, the most effective strengthen~ng mechanism for the nonferrous metals tends to be precipltatlon hardening.

PRECIPITATION OR AGE HARDENING

I

To be a candidate for precip~tation hardening, an alloy system must exhibit solubihty that decreases with decreasing temperature, such as the aluminum-rich portion of the alummum-copper system shown in Figure 5-3 and enlarged in Flgurc 5-4. Conslder the alloy with 4% copper,and use the phase d~agram to determine its equilibrium structure, Liquid metal solid~fies Into a single-phase sohd (a phase).At 1OOO0F,the full 4 % of copper ir dissolved and d~stributed throughout the alpha crysta1s.A~ the temperature drops, the maximum solubility of copper in aluminum decreases Irom 5.65% at 10ISFto less than 0 . 2 % a t room temperature. Upon cooling through the solvus (or solub~hty hmrt) line at 930F, the 4% copper alloy enters a two-phase region, and copper-rich thetainterphase precipitate? form and grow (Note:Theta-phase is actually a hard, br~ttle metalhc compound w ~ t h the chemical formula of CuAI,.) The equ~librium structure, therefore, would be an aluminum-rich alpha-phare structure wilh coarse theta-phase precipitates, generally lying along alpha-phase gram boundaries where the nucleation of second-phase particles can benefit from the exlstlng interfacial surface. Whenever two or more phases are present, the material exhibits d~sperslon strengthening. Dislocations are confined to thelr own crystal and camlot cross interfac~al

CHA~E 5 R Heat Treatment

Al

Weight % coppet
I

10

Atomic % copper

20

30

100% Al

Welght

percentage copper

Hlgh-aluminum section of the aluminum-copper equilibrium phase diagram.

FIGURE 5-3

FIGURE 5-4 Enlargement o f the solvus-line region of the aluminum-copper equil~br~um d~agram of Figure 5-3.

boundaries.Therefore, each interface between alpha-phase and the theta-phase preapitate is a strengthening boundary.Take a particle of theta precipitate and cut it into two halves. Forming the two half-size precipitates has just added two additional interfaces, corresponding to both sides of the cut. If the particle were to be further cut Into quarters, eighths, and sixteenths, we would expect strength to increase as we continually add interfackal surface. Ideally, we would like to have rmllions of ultra-small particles dlspersed throughout the alpha-phase structure. When we try to form this more desirable nonequilibnum configuration, however, we gain an unexpected benefit that adds significant strength. This new nonequillbrium treatment is known as age hardening or
preclpllakon hardenmng.

The process of precipitation hardening is actually a three-step sequence,The first step, known as solution treatment, erases the room-temperature structure and redissolves any existing preciprtate.The metal is heated to a temperature above the solvus and held in the single-phase region for sufficient time to redissolve the second phase and uniformly distribute the solute atoms (in this case, copper). If the alloy were slow cooled, the second-phase precipitate would nucleate and the material would revert back to a structure similar to equilibrium. To prevent this from happening, age-hardening alloys are quenched from their solution treatment temperature. The rapid-cool quenching, usually in water,suppresses diffusion,trapping atoms in place.The result is a room-temperature supersaturated solid solution. In the alloy discussed above, the alpha phase would now be holding 4% copper in solutlon at room temperature-far in excess of its equilibrium maximum of <0.2%. In this nouequilihrium quenched cou&tion,the material is often soft and can be easily straightened, formed, or machined If the supersaturated material were now reheated to a temperature where atom movement (diffusion) could occur, the alloy would attempt to form its equ~librium structure. If the reheating temperature remained w i t h the two-phase reglon, the excess solute atoms would precipitate out of the supersaturated matrii.This stage of the process, known as agrng, is actually a continuous transition. Solute atoms begin to cluster at

SEC~L 5.4 ~ N Strengthening Heat Treatments for Nonferrous Metals

95

FIGURE 5-5 Two-dimensional illustrations depicting (a) a coherent precipitate cluster where the precipitate atoms are larger than those in the host structure, and (b) its companion overaged or discrete secondphase precipitate particle. Parts (c) and (d) show equivalent sketches where the precipitate atoms are smaller than the host.

Ids or rst isnd nind his

m-

i
8

~
I

locations within the parent crystal, still occupying atom sites within the original lattice. Various transitions may then occur,leading ultimately to the formation of distinct secondphase particles with their own characteristic chcmistly and crystal structure. A key concept in the agingsequenceis that of coherency or crystalline continuity. If the clustered solute atoms continue to occupy lattice sites within the parent structure, the crystal planes remain continuous in all directions, and the clusters of solute atoms (which arc of different size and possibly different valence from the host material) tend to distort or strain the adjacent lattice for a sizable distance in all directions, as illustrated in Figures 5-5a and 5-52 For this reason, each small cluster appears to be much largrrwith respect to its ability to interfere with dislocation motion (i.e., impart strength). When the clusters reach a certain size, however, the associated strain becomes so great that the clusters can lower their energy by breaking free from the parent structure to form distinct second-phase particles with their own crystal structure and well-defined interphase boundaries, as shown in Figures 5-5b and 5-5d. Coherency is lost and the strengthening reverts to dispersion hardening, where dislocation interference is limited to the actual size of the particle. Strength and hardness decrease, and the material is said to be overaged. Figure 5-6 presents a family of aging curves for the 4% copper-96% aluminum alloy. For higher aging temperatures, the peak properties are achieved in a shorter time,

I ,
,

1
mi

m s lis~rn lri:d, Jc:ss

~ '
I
i

,
(

'

l
I !

at
'*

the AI4%Cu alloy at various temperatures showing peak , strengths and times of attainment. (Adapted from ; : Journal of the Institute for Metals, Val. 79, p. 321, 1951.)

' FIGURE 5-6 Aging curves for '

0.001

0.01

0.1

10

100

1000

10.000

Time, days

CHAFER 5

Heal Treatment

the finer but the peak hardness (or strength) IS not as great as can be achieved w ~ t h prec~pitates and larger amounts that form at lower aging temperatures. Selection of the aglngcondit~ons (temperature and time) is a decision that is made on the basls of desired strength, available equipment, and product~on constra~nts. The aging step can be used to divide precipitation-hardening materials into two types. (1) naturally agmg materrals, where room temperature is sufficient to move the unstable supersaturated solution toward the stable two-phase structure, and (2) art@crally agng malerials, where elevated temperatures are required to provide the necessary diffusion.With natural aging materials, such as alummum alloy rivets, some form of refr~gerat~on may be requred to retam the after-quench condition of softness Upon removal from the refrigeration, the rivets are eas~ly headed but progress to full strength after several days at room temperature. Since artificial aging requires elevated temperature to provide d~ffusion, the aging process can be stopped at any tlme by simply dropping the temperature (quenchmg).D~ffus~on is halted, and the current structure and propertles are "locked-m," prov~ded that the matenal 1s not subsequently exposed to elevated temperatures that would reactivate diffusion. When diffusion is possible, the material will always attempt to revert to its equihbrium structure! According lo Figure 5-6, if the 4% copper alloy were aged lor one day at 375F and then quenched to prevent overagmg. the metal would attain a hardness of 94 Vickers (and the associated strength) and retain these properrla throughout its useful lifetime provided subsequent diffus~on did not occur. If a h~gher strength is required, a lower temperature and longer tlme could be selected. Prec~pitation hardening is an extremely effective strengthen~ng mechanism and is responsible for the attractive engineering properties of many alum~nnm, copper, magnesium, and htanium alloys In many cases, the strength can more than double that observed upon conventional coolmg. Whileolher strengtheningmethods are traditionally used with steels and cast irons, those methods have been comb~ned w~th age hardening to produce some of the highest-strength ferrous alloys, such as the maraging steels and precipitation hardenable stainless.

tNING

HEAT TREATMENTS FOR STEEL

Iron-based metals have been heat treated for centuries, and today over 90% of all hcal-treatment operations are performed on steel. The str~king changes that resulted from plunging red-hot metal into cold water or some other quenching medium were awe-~nsp~ring to the ancients.Those who performed these acts in themakmg of swords and armor were looked upon as possessing unusual powers, and much superst~tion arose regarding the process Because quality wasdirectly related to the act of quenching, great importance was placed on the quenching medium that was used. Urine, for example, was found to be a superior quenching medium, and that from a red-haired boy was deemed particularty effective, as was that from a 3-year-old goat fed only ferns

ISOTHERMAL TRANSFORMATION DIAGRAM It has only been within the last century that the art of heat treatmg has begun to turn into a science One of the major barriers to understanding was the fact that the strengthening treatments were noneqmlibrium ln nature. &or varlatlons in coolmg often produced major variations in htructure and propertles. A useful aid to understanding nonequilibr~um heat treatment processes 1s the ~.~othermal-transformation (I-T) or time-temperature-transformatron ('I-T-T) dlugram. The mformahon in (his diagramis obtamed by heating thin specimens of aparticular steel to produce elevated-temperature uniform-chemistry austenite, "instantaneously" quenching to a temperature where austenite is no longer the stable phase, holding for variable periods of time at t h s new temperature, and observing the resultant structures vla metallographic photomicrographs (l.e, optical microscope exammation). For s~mplic~ty, consider a carbon steel of eutectoid compos~tion (0.77% carbon) and its T-T-T diagram shown as Figure 5-7. Above the A, temperature of 1341CF (727'C), austenrte is the stable phase and will pers~st regardless of the time Below this temper-

CHAPTER 5 Heat Treatment but the peak hardness (or strength) is not as greal as can be achteved w ~ t h the finer prec~pitates and larger amounts that form at lower aging temperatures. Selection of the agng conditions (temperature and time) is a declsion that 1s made on the has~s of deslred strength, available equipment, and production constralnts. The aging step can be used to div~de precip~tation-hardenmgmaterlals Into two typec (1) naturally agrng materials, where room temperature is sufficient to move the unstable supersaturated solution toward the stable two-phase structure, and (2) art$cially agrng materials, where elevated temperatures are required to prov~de the necessary diffus~on With natural aging materials, such as alummum alloy nvets, some form of refrigeration may be required to retain the after-quench condition of softness. Upon removal from the refrigeration, the rivets are easily headed but progress to full strength after several days at room temperature. Smce artific~al agingrequires elevated temperature to prov~de diffns~on, the aglng process can be stopped at any time by slmply dropping the temperature (quenchiug).Diffusion n halted, and the current structure and properties are "locked-in," prov~ded that the material is not subsequently exposed to elevated temperatures that would reactivate diffusion. When d~ffns~on is possible, the material w11l always attempt to rcvert to its equilibrium structure! Accordtng to F~gure 5-6, if the 4% copper alloy were aged for one day at 375F and then quenched to prevent overaging, the metal would attain a hardness of 94 Vickers (and the associated strength) and retam these properties throughout its useful lifetime provided subsequent diffusion did not occur. If a htgher strength 1s required, a lower temperature and longer tune could be selected. Prec~pitation hardening is an extremely effective strengthening mechanism and is responsible for the attractive englneerlng properties of many aluminum, copper, magnesium, and titanium alloys. In many cases, the strength can more than double that observed upon conventional cooling.Wlule other strengthening methods are traditionally used w~th steels and cast irons, those methods have been comblned with age hardening to produce some of the h~ghest-strength ferrous alloys, such as the maraging steels and prec~pitation hardenable stainless.

I(1

5.5 STRENGTHENING HEAT TREATMENTS FOR STEEL

Iron-based metals have been heat treated for centuries, and today over 90% of all heat-treatment operations are performed on steel. The strtking changes that resulted from plunging red-hot metal into cold water or some other quenching medium were awe-inspiring to the ancients Those who performed these acts in the maklng of swords and armor were looked upon as possessing unusual powers, and much superstition arose regarding the process. Because quahty was directly related to the act of quenchIng, great importance was placed on the quenching medium that was used. Urine, for example, was found to be a superior quenching medium, and that from a red-haired boy was deemed particularly effective, as was that from a 3-year-old goat fed only ferns.

ISOTHERMAL TRANSFORMATION DIAGRAM It has only been within the last century that the art of heat treating has begun to turn Into a science. One of the major harrters to understanding way the fact that the strengthIn nature. Minor variat~ons In cooling often proening treatments were nonequ~lthrium duced major variations in structure and properties. A useful atd to understanding nonequ~lihrium heat treatment processes is the so thermal-transformat~on (I-T) or tzme-temperamre-hansformatron(T-T-T) dragram. The information in this diagram 1s obtalned by heating thin specnnens of a particular steel to produce elevated-temperature uniform-chemistry austenlte, "instantaneously" quenching to a temperature where austenite is no longer the stable phase, holding for vartable perlods of tlme at this new temperature, and observtng the resultant structures via metallographlc photomicrographs (1 e., optical microscope examination). Mn Simplicity,consider a carbon steel of eutectoid composition (0.77% carbon) and 1t.s T-T-T d~agram shown as Figure 5-7 Above the A, temperature of 1341F (727'C), austenlte is the stable phase and will perslst regardless of the tune. Below thls temper-

SECTION 5.5 Strengthening Heat Treatments for Steel

"C

97

700

600

Isotherms1 transformation diagram

C-0.89% Mn 0.29% aunenitized at 1625*F.~ ~ asire i n6 5 Photornicro~iaphs ortginally X2500

500

400

300

zoo

M ~ 150% O Martensite)
Mso (90% Martensite)
200

1 mi".

1 hour

1 day

I week
I
306

- 0

a
PA.)

1
I

I
10

I I$

I
i o3

I
i o4

1 o5

ature, the face-centered austenite would like to transform to body-centered ferrite and

carbon-rich cementite, Two factors control the rate of transition. (1) the motlvatlon or drivmg force for the change and (2) the ability to form the desired products (i.e., the ability to rearrange the atoms through diffusion). The region below 1341Fin Figure 5-7
can be interpreted as follows. Zero time corresponds to a sample ccinstantaneously~' quenched to its new, lower temperature. The structure is usually unstable austenite. As tune passes (movlng horizontally across the diagram), a line is encountered representing the start of transformation and a second ime indicating completion of the phase change.At elevated temperatures (just below 1341F), atom movement within the solid (diffusion) is rapid, but the rather sluggish drivlng force dominates the kinetics. At a low temperature, the driving force 1 s high but diffusion 1s quite limited The kinetics of phase transformation are most rapid at a comprom~se ~ntermediate temperature, resulting in the charactenstic C-curve shape.The portion of the C that extends farthest to the left is known as the nose of the T-T-T or I-T d~agram. If the transionnation occurs between the A , temperature and the nose of the curve, the departure from equilibr~umis not very great. The austenlte transforms into

98

C~Arrm 5

Heat Treatment altemat~ng layers of ferrite and cementrte, producmg the pearlite structure that was introduced wrth the equilibnum phase diagram description in Chapter 4. Since the diffusion rate is greater at higher temperatures, pearlite produced under those cond~tions has a larger lamellar spacing (separat~on dlstance between srmilar layers).The pearlite formed near the A, temperature is known as conrsepearl~re, while the closer-spaced structures formed near the nose are calledfinepearlite. Since the resulting structures and properties are similar to those of the near-equilibrium process heat treatments, the procedure just descr~bed 1s called an so thermal anneal. If the austenite is quenched to a temperature between the nose and the temperature desrgnated as M,, a different structure is produced.These transformat~on condrtions are a srgnificant departure from equilibnum, and the amount of diffusion requlred to form the continuouslayers within pearlite is no longer available.The metal still has the goal of changing crystal structure from face-centered austenite to body-centered fernte, wrth the excess carbon being accommodated in the form of cementite The resulting structure, however, does not contain cementite layers but rather a dispers~on of discrete cementite particles dispersed throughout a malrix of femte. Electron microscopy may be required to resolve the carbides m this strucrure, which is known as bainite. Because of the fine d~spersion of carbide,it is stronger than fine pearlite, and ductility is retained because the soft ferrite is the continuous matrix. If austenrte is quenched to a temperature bclow the M, line, a different type of transformation occurs. The steel still wants to change its crystal structure from facecentered cubic to body-centered cub~c, but it can no longer expel the amount of carbon necessary to l o r n ferrite. Responding to the severe noncquilibrium conditions, it slmply undergoes an abrupt change in crystal structure with no significant movement of carbon.The excess carbon becomes trapped, distorting the structure into a body-centered tetragonal crystal lattice (distorted body-centered cubic), with the amount of d~stortion belng proportional to the amount of excess carbon. The new structure, shown in Frgure 5-8,s known as marten$ite,and, with sufficient carbon, it is exceptionally strong, hard,and brittle.Tne highly distorted lattlce effectively blocks the d~slocation motion that is necessary for metal deformatron. As shown in Figure 5-9, the hardness and strength of steel wlth the martensttic structure are strong functions of the carbon content Below 0.10% carbon, martensite is not very strong. Since no diffusion occurs durlng the transformation, hrgher-carbon steels form higher-carbon martensite, with an increase in strength and hardness and a concurrent decrease In toughness and ductility. From 0.3 to 0.7% carbon, strength and hardness increase rapidly. Above 0.7% carbon, however, the rlse is far less dramatic and

HGURF. 5-8 Photomrctogtaph bf rnartensfte; 100QX. (Courtesyof UnffedSfmi-8 Sfel CoVpP,

--

0 20

Pittsburph, PAJ

0 40 Carbon, %

0 60

0 so

Effectof carbon on the hardness of martensfte.

FIGURE 5-9

SECTION 5.5 Strengthening Heat Treatments for Steel

99

Temperature
FIGURE 5-10 Schematic representation depicting the amount of rnartensite formed upon quenching to various temperatures from M, through M,.

Weight

percent carbon

FIGURE 5-11 Variation of M, and M,temperatures with carbon content. Note that for high-carbon steels, completion of the martensite transformation requires cooling to below room temperature.

may actually be a decline, a feature related to the presence of retained austenite (to be described below). Unlike the other structure transformationsl the umount of martensite that forms is not a function of time, but rather depends only on the lowest temperature that is encountered during the quench.This feature is shown in Figure 5-10. where the amount of martensite is recorded as a function of temperature. Returning to the C curve of Figure 5-7, there is a temperature designated as M,,, where the structure is 50% martensite and 50% untransformed austenite.At the lower M,, temperature, the structure has become 90% martensite. If no further cooling were to occur, the untransformed austenite could remain within the structure. This retained austenite can cause loss of strength or hardness, dimensional instability, and cracking or brittleness. Since many quenches are to room temperature, retained austenite becomes a significant problem when the martensite finish, or 100% martensite, temperature lies below room temperature. Figure 5-11 presents the martensite start and martensite finish temperatures for a range of carbon contents. Higher carbon contents, as well as most alloy additions, decrease all martensite-related temperatures, and materials with these chemistries may require refrigeration or a quench in dry ice or liquid nitrogen to produce full hardness. It is important to note that all of the transformations that occur below the A, temperature are one-way transitions (austenite to something).The steel is simply seeking to change its crystal structure, and the various products are the result of this change. It is impossible, therefore, to convert one transformation product to another without first reheating to above the A , temperature to again form the face-centered-cubic austenite. T-T-T diagrams can be quite useful in determining the kinetics of transformation and the nature of the products.The left-hand curve shows the elapsed time (at constant temperature) before the transformation begins, and the right-hand curve shows the time required to complete the transformation. If hypo- or hypereutectoid steels were considered, additional regions would have to be added to the diagram to incorporate the primary equilibrium phases that form below the A, or A,, temperatures.These regions would not extend below the nose, however, since the nonequilibrium bainite and martensite structures can exist with variable amounts of carbon, unlike the near-equilibrium pearlite. Figure 5-12 shows the T-T-T curve for a 0.5%-carbon hypoeutectoid steel, showing the additional region for the primary ferrite.

C H A I ~5 ER Heat Treatment

FIGURE 5-12 lsothermaltransformation diagram for a hypoeutectoid steel (1050) showing the additional region for primary ferrite.

M 5 0

Man

QL

0 1 1

'

'

5 1 2

510

loZ lo3 lo4 Ttme - seconds

lo5

lo6

TEMPERING OF MARTENSITE
Despite its great strength, medlum- or high-carbon rnartenslte in its as-quenched form lacks sufficient toughness and ductility to be a useful engineering structure. A subsequent hcating, known as sempmng.is usually requ~red to unpart the necessary ductility and fracture resistance, and relax undes~rahle res~dual stresses As with most propertychanging processes, there is a concurrent drop In other features, most notably strength and hardness Marteus~te is a supersaturated solid solution of carbon in alpha-ferr~te and, therefore, is I$ metastable structure. When heated into the range of 100" to 700C (200" to 1300"F),the excess carbon atoms are rejected from solut~on, and the structure moves toward a mixture of the stable phases of ferrite and cementite. This decomposition of martensite into ferr~te and cementite is a time- dnd temperature-dependent, dlffusioncontrolled phenomenon with a continuous specirum of ~ n ~ e r m e d ~and a t e transitory structures Tahlc 5-1presents a chart-type comparison of the prev~ously discussed precipltationhardening process and the austenitize-quench-and-temper sequence. Both are nonequil~bnum heat treatments that mvolve three distinct stages.In both, the first step is ' a n elevated temperature soaking designed to erase the prtos structure, redissolving material to produce a uniform-chemistry, single-phase starting cond~tltlonBoth treatments follow this soak with a rapid-cool quench. In precipitation hardening, the purpose of the quench is to prevent nucleation of the second phase, thereby producinga superqaturated sol~d solut~on.lIus material is usually SOH, weak, and ductlle,w~th good toughness Subsequent agng (reheating wlthln the temperatures of the stable two-phase reglon) allows the mater~al to move toward the formation of the stable two-phase structure and sacrihces toughness and duct~lity for an increase in strength. When the proper balance is achieved, the temperature is dropped, d~ffusion ceases, and the current structure and properties are preserved, provided that the material is never subsequently exposed to any elevated temperature that would reactivate dlffusion and permit the structure to move further toward equilibrium. For steels, the quench induces a phase transformation a? the matenal changes from the face-centered-cubic austenite to the distorted body-centered structure known as martensite.The quench product is again a supersaturated,single-phase sohd solution, this time of carbon in iron, but the assoc~ated properties are the reverse of precip~tation hardening Martensite is strong and hard but relatively brittle When the matenal 1s

S E C ~ I5.5 ~ N Strengthening.Heat Treatments for Steel

101

HeatTreatment

Step 1
Hear into the stable single-phase region (above the solvus) and hold to form B unifarrn-chemistry single-phase solid solution.
Soiution irearrnent

Step 2
Quench- Rapid cool to form a oonequiiibrium supersaturated single-phase solid solution (crystal structureremains unchanged,malerial is soft and ductile).

Step 3
Age. A contn[>lledreheat in the.siable two-phase region (below the solvus]. The material rnwos toward the fotmarion ofthe,stahle.m-phase structurk, bbecmiitgstronger ahd hafder.Thepropertiesran be "frhaen iu3'bydrappiq the.tempperature to stop further diffusion. Temper. A controlled reheat in the slhblc two-phase region (below &+A,). lhematerial maver tnuard the formation af the stable two-phase structure, becoming weaker but tougher.Rlhe properties cnn be "frosen insby dropping the remperature to stop fwther diffusion,

Austmitine. Heat inlo the stable singlenhase region.(above the A, or A,,) and hold to form a uniform-chemistry singleephase sslid solution (austeniteJ.

Queuch. Rapid cool lo form a

nonequilibrium supersaturated single-phase solid solutiun (crystal stntcture'changes to body-centered martensite, which ishard but brittle).

tempered (reheated to n temperature within the stable two-phase region), strength and hardness are sacrificed for an increase in ductility and toughness. Figure 5-13 shows the final properties of a steel that has been tempered at a variety of temperatures. During tempering, diffusion enables movement toward the stable two-phase structure, and a drop in temperature can again halt diffusion and lock in properties. By quenching steel to form martensite and then tempering it at various temperatures, an infinite range of structures and corresponding properties can be produced.Tbis procedure is known as the quench-and-temperproccvs and the product, which offers an outstanding combination of strength and toughness, is called tempered martensite.

eering Properties of Steel,

Tempering tempstature, "F

FIGURE 5-14 C-C-T diagram for a eutectoid composition steel (bold), with several superimposed cooling curves and the resultant structures. The lighter curves are the T-T-T transitions for the same steel. (Courtesy of United States Steel Corp., Pittsburgh, PA.)

01 0.1

10 Time - seconds

I 02

1o3

Air cooled end

CONTINUOUS COOLING TRANSFORMATIONS While the T-T-T diagrams provide considerable information about the structures, obtained through nonequilibrium thermal processing, the assumptions of instantaneous, cooling followed by constant temperature transformation rarely match reality. Actual' parts generally experience continuous cooling from elevated temperature, and a dia-' gram showing the rewlts of this type of cooling at various rates would be far more useful.l What would be the result if the temperature were to be decreased at a rate of 500Fpen second, 50F per second, or 5F per second? A contmuous-cooling-transformation (C-C-T) dragram, like the one shown Figure 5-14, can provide answers to these questions and numerous others. If the cool ing is suffiaently fast (dashed curve A), the structure wll be martensite The slowes coohng rate that wlll produce a fully martensltic structure is referred to as the crtti cal coolcng rare. Slow cooling (curve D) generally produces coarse pearlite along wi a posslble primary phase. Intermedlaterates usually result in mixed structures, sin the time at any one temperature is usually insufficient to complete the transformation If each structure is regarded as providing a companion set of properties,the wide of possibilities obtainable though of steel become - the controlled heating - and coolmg . even more evident. JOMINYTEST FOR HARDENABILITY The C-C-T diagram shows that different cooling rates produce different structures wi different associated properties. Since the C-C-T diagram wlll change with materia chemistry, we have a general relation of the form:
material

Test bar

water

quenchxBT
structure
--t

+ cooling rate

--

properties

112"

,112';
I I

FIGURE 5-15 Schemat~c d~agram of the jominy hardenabil~ty test.

For a given material, its C-C-T diagram provides the link between cooling rate and st ture. Engineers who focus on use and application, however, are often more lntereste material properties and how they may be achieved. The Jominy end-qumch harderrab test2 and associated diagrams provide a useful tool and expand our understand' nonequilibriumheat treatment. In this test, depicted schematicallyin Figure 5-15,an spectrum of cooling rates are produced on a single four-inch-longspecimen by quenc a heated (i.e., austenitized) cylindrical bar from one end. The quench is standardize specifying the quench medlum (water at 75"F),the Internal nozzle diameter inc

(1

'Thls test n described m deta~l in the fullowmg standards ASTMA255, SAE J406,DIN 50191, andISO64

i l

SECTION 5.5

Strengthening HeatTreatments for Steel

"C
50

Hardness tests

Jominy bar

rr"

I \

dness
8

test

J 0

20

~ I 40

I 60

I 80

1 I 100 mm

Distance from quenched end

water pressure (that producing a 2;-inch vertical fountain), and the gap between the nozzle and the specimen (finch). Since none of the water contacts the side of the specimen, all cooling is directional along the axis of the bar. One end sees rapid cooling, whiie the other is essentially air-cooled.Since the thermal conductivity of steel does not change over the normal ranges of carbon and alloy additions, a characteristiccoolingrate can be assigned to each location along the length of the standard-geometry specimen. After the test bar has cooled to room temperature, a flat region is ground along opposite sides and Rockwell C hardness readings are taken every inch along the bar. The resulting data are then plotted as shown in Figure 5-16. The hardness values are correlated with position, and since the cooling rate is known for each location within the bar, the hardnesses are indirectly correlated with the cooling rate that produced them. Since hardness is also an indicator of strength, we have experimentally linked cooling rate to resultant strength in a simple and efficient manner. For any given material, application of the test assumes that equivalent cooling conditions will produce equivalent results. If the cooling rate is known for a specific location within a part (from experimentation or theory), the properties at that location can be predicted to be those at the Jominy test bar location with the same coolimg rate. Conversely,if specific properties are required, the necessary cooling rate can be determined. Should the cooling rates be restricted by either geometry or processing limitations, various materials can be compared and a satisfactory alloy selected. Figure 5-17 shows the Jominy curves for several engineering steels. Since differences in chemical composition can exist between heats of the same grade of steel, hardenability data are

f
I

,~:
I
8

-17 Jominy hardness

r :

:,i,
..
~

same carbon content ing types and amounts

10 1

I I 24 32 Sixteenths of an loch D~stance from quenched e n d of Jom~ny speclmen

0
I 8

16

. .

104

CHARER5

Heat Treat ment often presented in the form of bands, where the upper curve corresponds to the maxunum expected hardness and the lower curve to Lhe min~mum.The data for actual heats should then fall between these two extremes

HARDENABILITY CONSIDERATIONS Several key effect? must be considered if we are to understand the heat treatment of steel: (1)the effect of carbon content, (2) the effect of alloy additions, and (3) the effect of vanous quenching conditions.The first t%o relate to Lhe materlal being treated and the third to the heat-treatment process Hardness is a mechanical property related to strength and is a strong function of the carbon content of a steel and the particular microstructure. With difterent heat treatments, the same steel can have different hardness values. Hurdenubil~tyis a measure of the depth to whlch full hardness can be obtained under a normal hardenmg cycle and is related primarily to the amounts and types of alloyrng elements. Hardenabil~ty, therefore, is a material property dependent upon chemical composition. In Figure 5-17, all of the steels have the same carbon content, but they drffer m the type and amounts of alloy elementsThe maxunum hardness is the same in all cases, but the depth of hardening (or the way hardness varies wlth distance from the quenched end) varies constderably. Figure 5-18 shows Jominy test results Cor steels containmg the same alloying elements but varlable amountsof carbon.Note the change in peak hardness as carbon content is increased. The results of a heat-treat operatLon depend on both the hardeuabdity of the metal and the rate of heat extraction.The primary reason for adding alloy elements to commercial steels is to Increase their hardenabihty,not to improve then strength. Steels wlth greater hardenabihty can achieve a deslred level of ~Lrength or hardness wtth blower rates of cooling and, for thrs reason, can be completely hardened in th~cker sections. Slower cooling also serves to reduce the amount of quench-lnduced distortion and the likelihood of quench crackmng. An accurate determination of need 1 s requlred if steels are to be selected for specific applications. Strength tends to be assoc~ated wlth carbon content, and a genBecause eral rule is to select the lowesl posslble level that will meet the spec~ficat~ons heat can be extracted only from the surface of a metal, the slze of the plece and the depth of requlred hardening set the conditions for hardenability and quench. For a given quench condit~on, different alloys will produce different results. Because alloy add~tions lncrease the cost of a material, lt is best to sclect only what 1s required to ensure compliance wlth specifications.Money is often wa~ted by specifying an alloy steel
M~lllmeters 20 30 I I

70

10

40 I

50 -

60 -

FIGURE 5-18 Jominyhardn ess curves for engineering steels with identical alloy conditions but variable carbon content.

16 24 32 Sixteenths of an mcll Dtstance from quenched end of Jominy speclmen 8

SECTION 5.5

Strengthening Heat Treatments for Steel

105

for an application where a plain-carbon steel, or a steel with lower alloy content (less costly), would be satisfactory. When greater depth of hardness is required, another alternative is to modify the quench conditions so that a faster cooling rate is achieved. Quench changes may be limited, however, by cracking or warping problems as well as other considerations relating to the size, shape, complexity, and desired precision of the part being treated.

QUENCH MEDIA
Quenchants are selected to provide the cooling rates required to produce the desired structure and properties in the size and shape part being treated. Quench media vary in their effectiveness, and one can best understand the variation by considering the three stages of quenching. Let's begin with a piece of hot metal being inserted into a tank of liquid quenchant.If the temperature of the metal is above the boilingpoint of the quenchant, the liquid adjacent to the metal will vaporize and form a thin gaseous layer between the metal and the liquid. Cooling is slow through this vaporjacket (first stage) since the gas has an insulating effect,and heat transfer is largely through radiation. Bubbles soon nucleate, however, and break the jacket. New liquid contacts the hot metal, vaporizes (removing its heat of vaporization from the metal), forms another bubble, and the process continues. Because large quantities of heat are required to vaporize a liquid, this second stage of quenching (or nucleate boiling phase) produces rapid rates of cooling down to the boiling point of the quenchant. At this point, vaporization can no longer occur. Heat transfer must now take place by conduction across the solid-liquid interface, aided by convection or stirring within the liquid. In this third stage of quenching, the slower cooling by conduction and convection continues from the boiling point to room temperature. Breakdown of the vapor jacket, bubble removal, and convection cooling can all be aided by moving the metal through the liquid or flowing the liquid over the metal surface. Various liquids offer different heats of vaporization, viscosities, and boiling points. Quenches can be further tailored by varying the temperature of the liquid and the degree of flow or agitation. Water is a fairly effective quenching medium because of its high heat of vaporization and the fact that the second stage of quenching extends down to 100C (212*F),usually well into the temperatures for martensite formation. Water is also cheap, readily available, easily stored,nontoxic, nonflammable, smokeless,and easy to filter and pump. Agitation is usually recommended with a water quench, however, since the clinging tendency of the bubbles may cause soft spots on the metal. Other problems associated with a water quench include its oxidizing nature (i.e., its corrosiveness) and the tendency to produce excessive distortion and possible cracking. While brine (salt water) has a similar heat of vaporization and boiling point to water, it produces more rapid cooling because the salt nucleates bubbles, forcing a quick transition through the vapor jacket stage. Unfortunately, the salt in a brine quench also tends to accelerate corrosion problems unless a U residues are completely removed by a subsequent rinse. Different types of salts can be used, including sodium or potassium hydroxide,and various degrees of agitation or spraying can be used to adjust the effectiveness of the quench. (Note: Because of all of the dissolved salts, the urines cited in quenching folklore are actually quite similar to the brines of today.) If a slower cooling rate is desired, oil quenches are often utilized.Various oils are available that have high flash points and different degrees of quenching effectiveness. Since the boiling points can be quite. high, the transition to third-stage cooling usually precedes the martensite start temperature. The slower cooling through the M,to-Mi martensite transformation leads to a milder temperature gradient within the piece, reduced distortion,and reduced likelihood of cracking.Heating the oil actually increases its cooling ability, since the reduced viscosity assists bubble formation and removal. Problems associated with oil quenchants include water contamination, smoke, fumes, spill and disposal problems, and fire hazard. In addition, quench oils tend to be somewhat expensive. Quite often, there is a need for a quenchant that will cool more rapidly than the oils but slower than water or brine. To fill this gap, a number of water-based polymer

ieat Treat1nent

quench solutions (also called synthetic quenchants)have been developed.Tailored quenchants can be produced by varymg the concentrations of the components (such as liquid organic polymers, corrosion inhibitors, and water) and adjusting the operating temperature and amount of agitation.The polymer quenchants provide extremely uniform and reproducible results, are less corrosive than water and bnne, and are less of a fire hazard than oils (no fires, fumes, smoke, or need for air pollution control apparatus). Distortion and crackmg are less of a problem since the boiling point can be adjusted to be above the martensite start temperature. In addition, the polymer-rich film that forms initially on the hot metal part serves to modify the cooling rate, If slow cooling is required, molten salt baths can be employed to provlde a medium where the quench goes directly to the third stage of cooling Still slower coohng can be obtained by cooling in still air, burying the hot matenal m sand, or a variety of other methods. High-pressuregas quenching uses a stream of flowing gas to extract heat, and the cooltng rates can be adjusted by controlling the gas velocity and pressure. Results are comparable to oil quenching, with far fewer environmental and safety conceins. From an environmental perspective, vegetable oils may also be an attractive quenchant.They are biodegradable, offer low toxicity, and are a renewable resource.

DESIGN CONCERNS, RESIDUAL STRESSES, DISTORTION, AND CRACKING Product design and material selection play important roles iu the satisfactory and economical heat treatment of parts. Proper consideration of these factors usually leads to simpler, more economical, and more reliable producb Failure to relate des~gn and materials to heat-treatment procedures usually produces disappo~nting or variable results and may lead to a vanety of service failures. From the viewpoint ot heat treatment,undesirahle design featuresinclude (I)n form sections or thicknesses. (2) sharp interior cornersand (3) sharp exterior corner . these features often find their way into the design of parts, the de~igner should be aware therr effect on heat treatment. Undesirable results may include nonuniform structure and properties, undesirable residual stresses, cracking, warping, and dimensional changes Heat can only be extracted from a piece through its exposed surfaces.The if the plece to be hardened has a nonuniform cross section, any thln reglon w rapidly and may fully harden, whlle thick regions may harden only on the surface,if all.The shape that m~ght he closest to ideal from the viewpoint of quenching a doughnut. The uniform cross sechon with h ~ g h exposed surface area and absence sharp comers would be qulte attractwe. Since most shapes are designed to function, however, compromises are usually necessary. Residual stresses are the often-complex stresses that are presen independent of any applied load.They can be induced in a number of ways, b plex dmensional changes that crtn occur during heat treatment are a pri ma1 expansion during heating and thermal contraction dumg cooling are phenomena, but when these occur in a nonumform manner, the results can b complex. In additLon, the various phases and structures that can exist wlthin a materi usually possess different densities. Volume expauslons or contractions accompany an phase transformation.For example, when au5tenite transforms to martensite,there is avo ume expansion of up to 4%.Transformations to femte, pearhte, or other room-temperat structures also involve volume expansions but of a smaller magnitude. If all of the temperature changes occurred uniformly throughout a part, all of tb associated dimensional changes would occur slmuttaneously and the resultant pro would be free of residual stresses. Most of the parts being heat treated rlence nonumform temperatures during the cooling or quenching oper a block of hot aluminum being cooled by water sprays from top a simplicity, let us model the block as a three-layer sandwich, as in the top sequence Figure 5-19. At the start of the quench, all layers are uniformly hot. As the waters begins, the surface layers cool and contract, but the center layer does not experie the quench and remains hot. Since the part 1s actually one piece, however, the vario

------.----

----------Cold

jLq
Cornpresslon Compression
If separate layers

-------.--. .-----.----

Hot material ,dergoing surface coaling

If separate -butlayers

Since integral piece, the hat center deforms

-bttt-

Final residual stresses, since integral piece

Posstbie
quench crack

---.-------

Cornpresslon ----------Tension

HGURE 5-19 Three-layer model of a plate undergoing cooilng. The upper sequence depicts a material

that contracts upon cool~ng while the bottom sequence deplcts steel, which expands such as alum~num durlng the cooling-induced phase transformation. layers must accommodate each other.The contractrng surface layers exert compressive forces on the hot, weak interior, causing it to also contract,-but by plastic deformation,not thermal contracti0n.A~ time passe$ the interior now cools and wants to contract but finds itself sandw~cbed between the cold, strong surface layers. It pnlls on the surface layerq placlng them in compression, while the surface layers restrict its movement, creating tension in the interior.Whie the net force is zero (since there is no applied load), counterbalancing tension and compress~on stresses exist within the product. Let us now change the material to steel and repeat the sequence. When heated, all threelayers are hot, face-centered-cubic austenite, as shown in the bottom sequence of Figure 5-19. Upon quenching, the surface layers transform to martensite (the structure changes to body-centered tetragonal) and expand! The expanding surfaces deform the .soft, weak, (and s t 1 1 1hot) austemte center, which then cools and wants to expand as it undergoes the crystal structure change to martensite or another body-centered, roomtemperature product.The hard, strong surface layers hold it back, placing the center in compression, wh~le the expanding center tries to stretch the surfacelayers,producingsurface tension. If the tension at the surface becomes great enough, cracking can result, a phenomenon known as quench cracking. One should note that for the rapid-quench conditions just described, aluminum w i l l never quench crack, because the surface is in compression, but the steel might, since the residual stresses are reversed. If the cooling conditions were not symmetrical, therenught be more contraction or expansion on one side, and the block might warp. With more complex shapes and The accompanyingnonuniform coollng, the residual stresses induced by heat treatment can be extremely complex. Various techniques can be employed to minimize or reduce the problems associated with residual stresses. If product cross sections can be made more uniform, temperature differences can be minimized and will not be concentrated at any specific location. If this is not possible, slower cooling may be recommended, coupled with a ovide the desired properties after a slower oil or air quench. Since r hardenability are more expensive,design alternatives should genWhen temperature differences and the resultant residual stresses become severe or localized, cracking or distortion problems can be expected. Figure 5-20a shows an example where a sharp Interior corner bas been placed at a change in cross section. Upon quenching, stresses will concentrate along lime A-B, and a crack is almost certain

e un~form sections.

foresurt, EcbmgesLh~croses~m q n i . F e d , f % q M k e r @ % redesigned version at Figure 5-2Qa.Genetous EIlinets at interior eornem raeiused rior ~ ~ r n eand r s smooth irmsitiotrs rduce problems. A material with pester ha& &a lwmvem gue& would a180 help. Figure 5-ZUb shoes thscrnm seaion da,M&&ww m m k t e d ym & k r h i n g [email protected] and adding a d d i t i d holes to prtwEde a mwe udom c m sectdan t i & @ ? Ylubmfq$ eIimfmted the problem. (3ne of the om~nous features of pws product and process design is the fact that the residual vtrmes may ngt grodp~$ huwdiatsfailrrre but m y cantribute to fail= at a later tiae.A,pplled skeases add to the reidual stresses alreadypresent Within thepart. TherEfore, i t is podble for applied stresses that are wdl within the "safe" desi&n&
Urnit to couple with tesidual stresfen_ and produae a d u e . sufficientto induce fdu16.

am employed.

TECHNIQUES TO REDUCE CRACLClNG AND DIsTORTI~N

Lop rime --t

I*)

to result. If changes in cross sectlon are reqmred, they should be gradual, as in the redesigned version of 5-20a. Generous f~llets at Interior corners, radiused exterlor corners, and smooth trandtions all reduce problems A material with greater hardenability and a less severe quench would also help. E~gure 5-20b shows the cross sectlon of a blanking die that consistently cracked during hardening.Rounding the sharp corners and addimg additional holes to provide a more uniform cross sectlon during quenching eliminated the problem. One of the omlnous features of poor product and process design 1s the fact that the residual stresses may not produce immediate fatlure but may contribute to allure at a later time. Applied stresses add to the res~dual stresses already present within the part. Therefore, it is possible for appl~ed stresses that are well within the "safe" designed llm~t to couple with residual stresses and produce a value sufficient to induce failure. Residual stresses can also accelerate corrosion reactions. Dimensional changes or warp. mg can occur when subsequent machining or grmdmg operations upset the equilibrium balance of the residual stresses.Mter the removal of some material, the remaining piece adjusts its shape to produce a new (sum of forces equalzero) equihbrium balance. When we consider all of the possible difficulties,it 1s apparent that considerable tune and money can be saved if good design, material selectiou,and heat-treatment practices are employed.

TECHNIQUES TO REDUCE CRACKING AND DISTORTION The steel segment of Figure 5-19 has already introduced the phenomena of quench cracking, and F~gure 5-21a further illustrates its cause uslug a T-T-T diagram (a misuse of the diagram, since continuous cooling is employed, but helpful for visuahzation) The surface and center of a quenched product generally have d~fferent cooling rates. As a result, when the surface has cooled and is transforming to martensite with its companion expansion, the center is still hot and remains soft, untransformed austemte. At a later tune, the center cools to the martensite transformation, and it expandsThe result is sigmficant tension in the cold, hard surface and possible cracking. Figure 5-21b deplcts two varlatlonb of rapid quenching that have been developed to produce strong structures while reducing the l~kehhood of cracking. A raprd cool must still be employed to prevent transformation to the softer, weaker pearlihc structure, but instead of quenching through the martensite transformation, the component is rap~dly quenched into a liquid medium that is now several degrees above the martenslte start (M,) temperature, such as hot oil or molten salt.Holding for a period of time in this bath allows the entire piece to return to a nearly uniform temperature.
Surface

3 +

E' ,m m

I -

:
M s Mf
Log time(a)

Log tlme-

representat~on of the cool~ng paths of surface and center during a dlrect quench; FIGURE 5-21 (a) Schemat~c (b) The mod~fied cooling paths exper~enced during the austempering and martempering processes.

Time - seconds

If the material is simply held at t h ~ temperature s for sufficient time, the austen~te will transform to bainite, which usually has sufftc~ent toughness that a subsequent temper is not required Thls process is known as austemperlng. If the materlal IS brought to a uniform temperature and then slow cooled tbrough the martennte transformation, the process is known as martempenng, or marqu~nihlng. The resulting structure is martensite, which must be tempered the samc as the martensite that forms directly upon quenching. In both of these processes, all transformations (and related volume cxpans~ons) occur at the same time, thereby el~minating the residual streshes and tendency to crack.

AUSFORMING A process that is often confused with austempering i q ausforming. Certaln alloys tend to retard the pearlite transformation more than the bainite reaction and produce a T-T-T curve of the shape shown in Figure 5-22 If this mater~al is heated to form austenite and then quenched to the temperature of the "bay" between the pearlite and bainite reactIons,~t can retaln its austenite structure for a useful period of time. Deformation can be performed on an austenite structure ar a temperature where it techn~cally should not exist Benefits include the tncreased ductiltty of the face-centered-cublccrystal structure, the Elner grain slze that forms upon recrystalhzation at the lower temperature, and the possib~lityof some degree of strain hardenmg. Follow~ng the deformat~on, the metal can be slowly cooled to produce bainlte or rapidly quenched to martensite, which must then be tempered. The resulting product has exceptional strength and duct~lity,coupled w~th good toughness. creep resistance, and fat~gue ltfe-propart~es that are superior to those produced if the deformat~on and transformation processes are conducted in thew normal separated sequence. Ausform~ng is an example of a growing class of fhermomrchanrcal processes ln which deformatton and heat treatment are intimately combmed.

this Many products require different properties at d~fferent locations Q u ~ t frequently, e variation takes the form of a hard, wear-resistant surface coupled with a tough, Eractureresistant core.The methods developed to produce the varied properties can be classhed into three basic groups: (1)selective heating of the surface, (2) altered surface chemistry, and (3) deposibon of an addittonal surface layer.l%e first two approaches will be dlscussed In the next sections, whde platings and coatings will be described in Chapter 36.

FIGURE 5-23 Sectlon of gear teeth showing induct~onhardened surfaces. (Courtesy of Ajax Tocco Magnatherm~c Corp., Warren, OH)

SELECTIVE HEATING TECHNIQUES If a steel has sufficient carbon to attain the desired surface hardness, generally greater than 0.3%, the different properties can often be obtamed simply by varying the thermal histones of the various regions Core properties are set by a bulk treatment, with the surface properties bemg established by a subsequent surface treatment. Maximum hardness depends on the carbon content of the mater~al, while the depth of that hardness depends on both the depth of heating and the material's hardenability. The various methods generally differ in the way the surface is brought to elevated temperature. Plume hurdenzng uses an axy-acetylene flame to raise the surface temperature h~gh enough to reform austenlte.The surface is then water quenched3 to producemartensite and tempered to the desired level of toughness.Heat input a quite rapid,leaving the interior at low temperature and free from any srgnificant change. considerable flexlbility is provided slnce the rate and depth of heating can be easily varied. Depth of hardening can range from thin skins to over 6mm. ($inch). Flame hardening is often used on large objeets,slnce alternative methods tend to be lim~ted by both size and shape. Equipment varies from crude handheld torches to fully automated and computenzed units When using znducrion hardenzag, the steel part is placed inside a conductor coil, which is then energized with alternating current. The changlng magnetlc field Induces surface currents in the steel, which heat by electrical resistance.The heatingrates can be extremely rapid, and energy efficiency IS high. Rapid cooling is then provided by either an immersion quench or water spray using a quench rlng that follows the induction cod. Induction heating IS particularly well su~ted to surface hardening slnce the rate and depth of heating can be controlled directly through the amperage and frequency of hardening IS ideal for round bars and cylindrical parts but can the generator. Induct~on also be adapted to more complex geometries. The process offers high quality, good reproducibility, and the posslbil~ty of automation. Figure 5-23 shows a partial cross sectlon of an induction-hardened gear, where hardening haa been applied to those areas expected to see hlgh wear.Distortion during hardening is negligible since the dark areas remain cool and n g d throughout the entire process. Laser-beam hardening has been used to produce hardened surfaces on a wide variety of geometries. An absorptive coating such as zinc or manganese phosphate is often appl~ed to the steel to improve the efficiency of converting light energy Into heat. The surface 1s then scanned with the laser, where beam srze, beam intensity, an scanning speed have been selected to obtain the desired amount of heat input and for the depth of heating. Because of the localized heatrng of the beam, it is poss~ble process to be autoquench~ng (cooled simply by conductive transfer Into the underly Ing metal), but a water or 011quench can also beused.Through laser-beam hardening, a 0.4% carbon steel can attain surface hardnesses as high as Rockwell C 65.Typical depth of h a r d e m g is between 0.5 and 1 mm (0.02 to 0.04 inches).The process operates at high speeds, produces little distortion, induces res~dual compressive stresses on the surface, and can be used to harden selected surface areas while leaving the remaining surfaces unaffected. Computer software and automation can be used to control the process parameters, and conventional mirrors and optics can be used to shape and manipulate the beam Electron-beam hardenzng is similar to laser-beam hardening Here the heat sour is a beam of high-energy electrons rather than a beam of light, with the charged par cles being focused and directed by electromagnetic controls. Like laser-beam treating, the process can be read~ly automated,and production equipment can perform a var' of operations with efficiencies often greater than 90%.Electrons cannot travel ~nair,h ever, so the entire operation must be performed in a hard vacuum, which is the majo limitat~on of thls process. More mformation on laser- and electron-beam techniques,a well as other means of heating material, is provided in Chapters 31 through 34.
'There is no real danger of surface crackmg durlng the water quench When thesurface IS rzausten~t~zed,th soft austenlte adjusts to the colder, stronger underly~ng malenal Upon quenching, the surface austenl expands durrng rransfomatlon.Tne intenor is still cold and restrams the expansron, producinga surface compression w ~ t h no tendency toward crack1ngTn18is just the opposlte of the cond~ttans that occur dun the thmugh-hardenmg of a furnace soaked workplecel

SECTION5.6 Surface Hardening of Steel

11 1

Still other surface-heating techniques employ immersion in a pool of molten lead or molten salt ( l a d p o t or salt bath heating).These processes are attractive for treating complex-shaped products and hardening relal~vely inaccess~ble surfaces.

TECHNIQUES INVOLVING ALTERED SURFACE CHEMISTRY If the steels contain msufficient carbon to ach~eve the desired surface properties, or the difference m surface and mterior properties is too great for a single chem~stry material, an alternative approach 1s to alter the surface chemistry,The most common technique within this category, curburrzing, involves the hffusion of carbon inlo the elevaledtemperature, face-centered-cub~c, austenite structure, at temperatures between 800" and 1050C (1450" and 1950F) When suIficient carbon has d~ffused to the desued depth, the parts are then thermally processed. Direct quenching from the carburizat~on treatment 1s the s~mplest alternative, and the different carbon contents and cool~ngrates can often produce the desired varlation in properties Alternative processes include a slow cool from the high-temperature carburlzing treatment, followed by a lower-temperaturereaustenitizing and quenching, or a duplex process involv~ng a bulk treatment and a separate surface heat treatment The5e latter processes are mole involved and more costly but produce improved product properties.The carbon content of the surface usually vanes from 0 7 to 1.2% depending on the material, the process, and the desiredresults. Case depth may range from a few thousandths of an inch to over; inch The various carburizmg processes T i e r in the source of the carbon.The most common is gm carburizlng,where a hot, carbon-conta~ning gas surrounds the parts.The process is fast,is easily controlIed, and produces accurate andun~formly modiCed surfaces.Inpackcarbunzrg, the steel components are surrounded by a hlgh-carbon solid material (such as carbon powder, charcoal, or cast Iron tnrnings) and heated in a furnace.The hot carburuing compound produces CO gas, which reacts wth the metal, releasing carbon, which is readly absorbed by the hot austenite.Amolten bathsupphes the carbon m lqurd carburizlng At one time, liquid-carburiung baths contained cyan~de, which supphed both carbon and mtrogen to thesurface. Safetyand environmental concerns now dictate the use of noncyanlde hquid compounds that are generally used to produce thin cases on small parts. Nttndmg hardens the surfaceby producing alloy nitride? in special steels that contain nitride-forming elements like aluminum, chrommm,molybdenum,or vanadinm.The parts are first heat treated and tempered at 525" to 675'C (1000" to 1250F).After cleaning and removal of any decarburized surface material, they are heated man atmosphere contain~ng dissociated ammonia (nitrogen and hydrogen) for 10 to 40 hours at 500" to 625C (950" to 115WF). S~nce the temperatures are below the A , temperature, the nitrogen is diffusing into fernte, not austenite, and subsequent cooling will not Induce a phase transformation.The diffused nitrogen forms aUoy nitrides, hardening the metal to a depth of about 0.65 mm (0.025 m.). Extremely hard cases are formed and distortion is low. No subsequent thermal processing rsrequired. In fact,subsequent heating should be avoided because the thermal expansions and contractions will crack the hard nitnded case. Fmshgnnding should also be avoided because the nitrided layer is excephonally thm. lonrtriding is a plasma process that has emerged as an attractwe alternat~ve to the conventional method of nitridmg. Parts to be treated are placed m an evacuated"furnace" and a direct current potential of 500 to 1000 volts is applied between the parts and the furnace walls. Low-pressure nitrogen gas is introduced Into the chamber and becomes ionized.The Ions are accelerated toward the negatively charged product surface,where they impact and generate sufficient heat to promote Inward diiusiou. T h ~ is s the only heat associated w~th the process, the "furnace" acts only as a vacuum container and electrode. Advantages of the process include shorter cycle tlmes, reduced consumption of gases, sign~ficantly reduced energy costs, and reduced space requirements. Product quality is improved over that of conventional nitriding, and the process is applicable to a w~der range of materials Ion carbur~zmg 1s a parallel process m which low-pressure methane is substituted for the low-pressure nitrogen, and carbon diffuses into the surface. Carbon and nltrogen can be added s~multaneously. If ammonia is added to a standard carbunzmg atmosphere, and the steel is heated to a lemperature where the structure is austenite, the process becomes one of carbonrtridtng.The temperature is usually

1 12

CHAPTER 5 Heat Treatment

lower than for standard carburizing, and the treatment time is somewhat shorter. If C02 is added to the ammonia of nitriding and the process is carried out at a temperature below the A,, the process is known as nitrocarbuvizing. The resulting surface resists scuffing, and fatigue resistance is improved. Ion plating and ion implantation are other technologies that permit modification of the surface chemistry. In Chapter 35 we will expand on the surface treatment of materials and compare the processes discussed in the preceding sections to various platings, coatings, and other techniques.

FURNACE TYPES A N D FURNACE ATMOSPHERES To facilitate production heat treatment, many styles o C furnaces have been developed in a wide range of sizes, each having its characteristic advantages and disadvantages.These furnaces are generally classified as batch or continuous type. Batcicfurnuces, in which the workpiece remains stationary throughout its treatment, are preferred for large parts or small lots of a particular part or grade of steel. Continuous furnaces move the components through the heat-treatment operation at r a t s selected to be compatible with the other manufacturing operations. Continuous fumaces are used for large production r u n s where the same or similar parts undergo the same thermal processing.The workpieces are moved through the furnace by some type of transfer mechanism (conveyor belt, walking beam, pusher, roller, or monorail) and often fa11into a quench tank to complete the treatment. By incorporating various zones, complex cycles of heating, holding, and quenching/cooling can be conducted in an exact and repeatable manner with low labor cost. Horizontal batch furnaces, often called box furnaces because of their overall shape, generally use gas or electricity as their source 0 f h e a t . A ~ shown in Figure 5-24, a door is provided on one end to allow the work to be inserted and removed.When large or very

is lowered over the work. An airtight inner shell is often placed over the work@eces

FIGURE 5-25

Car-bottom box furnace. (Courtesy of Sola/Hevi-Dut~Rosemont, IL.)

contain a protective atmosphere during the heatmg and cooling operatlons. After the work is heated, the furnace unit can be lifted off and transferred to another batch, wh~le themner shellmainta~ns a protective atmosphere during cooling. If extremely slow cooling is desired, an insulated cover can be placed over the heated shell An mterestmg modification of the bell dewgn 1s the elevator furnace, where the bell remains statlonary and the workpieces are ralsed into it on a movable platform that then forms the bottom of the furnace.By planng a quench tank below the furnace, t h ~ design s enables the workpleces to first be raised into the furnace and then lowered into the quench tank. It 1s extremely attractive for applications where the workmust be quenched as soon as possible after be~ng removed from the heat. When Ion&slender parts are pos~tioned horizontally, there is little resistance to sagging or warpmg. For these types of workp~eces, a vertccal pia furnace is preferred.These furnaces are usually cylindrical chambers sunk into the floor wlth a door on top that can be swung a s ~ d to e allow suspended workpieces to be lowered lnto the furuace.l%ey can also be used to heat large quantittes of small parts by loading them ~ n t o wxe-mesh baskets that are then stacked w~thin the column. While all of the furnaces that have been described can heat in air, most commercial furnaces can also employ artlficralgas atmospheres. These are selected to prevent scaling or tarnishing, to prevent decarburization, or even to provide carbon or nitrogen for surface modification. Many of the artific~al atmospheres are generated from either the combustion or decomposition of natural gas, but nitrogen-based atmospheres frequently ofler reduced cost, energy savings, increased safety, and environmental attractlveness Other common atmospheres include a~gon, dissociated ammoma, dry hydrogen, helium, steam, and vacuum. The heating rates of gas atmosphere furnaces can be slgnificantly increased by incorporating thefluidized-bed concept.These furnaces consist of a bed of dry,~nert parucles, such as aluminum oxide (a ceram~c), which are heated and fluidxed (suspended) in a stream of upward-flowmg gas. Products introduced Into the bed become engulfed in the particles, which then rad~ate uniform heat Temperature and atmosphere can be altered quickly, and high heat-transfer rates, hlgh thermal efficiency, and low fuel consumption have been observed. S~nce atmosphere changes can be performed in minutes, a single furnace can b e used for nitriding, stress relievmg, carburizing, carbonitrldlng, anneahng, and hardenmg. When a liquld heatmg medium 1s preferred,salt bath furnaces are apopular cholce. Electrically conductive salt can be heated by passtng a current between two electrodes suspended in the bath.The electr~cal currents also cause the bath to circulate and thereby malntam uniform temperature. Nonconductive salts can be heated by some form of ~mmersion heater, or the containment vessels can be externally fired. In these furnaces, themolten salt not only serves as a un~form source of heat but also can be selected to prevent scaling or decarbunzati0n.A lead pot IS a s~milar device,where molten lead replaces salt as the heat-transfer medlunl. Electrical znductzon heating is another popular means of heating conductive materials, such as metal. Small parts can be through-heated and hardened. Longproducts can be heated and quenched in a continuous manner by passmg them through a statlonary heatmg coil or by having a moving coil traverse a stationary part. Locahzed or selective heahng can also be performed at rapld product~on rates Flex~bility is another attiactive feature, slnce a standard induction unlt can be adapted to a wide variety of products simply by changlng the induction coil and adjusting the equipment settings.

FURNACE CONTROL All heat-treatment operatlons should be conducted w ~ t h r ~ g control ~d if the desired results are to be obta~ned in a consistent fashion. Most furnaces are equipped with one or more temperature sensors, which can be coupled to a controller or computer to regulate the temperature and the rate of heating or cooling. It should be remembered, however, that it is the temperature of the workpiece, not the temperature of the furnace, that controls the result, andit is this temperature that should be monitored.

114

C H A ~ 5 R Heat Treatment

I 5.8 HEAT TREATMENT AND ENERGY

Because of the elevated temperatures and the time required at those temperatures, heat treatments can consume considerable amounts of energy. However,if one considers the broader picture, heat treatment may actually prove to be an energy conservation measure.The manufacture of higher-quallty, more durable products can often eliminate the need tor frequent replacements. Higher strengths may also permit the use of less matenal in the manufacture of a product, thereby saving additional energy. Further savlngs can often be obtained by integrating the manufacturing operations. For example, a d ~ r e cquench t and temper from hot forging may be used to replace the conventional sequence of forge, conventional air cool, reheat, quench, and temper. One should note, however, that the mtegrated procedure quenches from the cond~tions of forgtng, which generally have greater variabihty in temperature, un~formity of that temperature, and austemte gram slze. If these variations are too great, the additional energy for the reheat and soak may be well justified. Heat treatment, a business worth $15-20 billion a year in the United States,Impacts nearly every industrial market sector. It is both capital intensive (specialized and dedicated equrpment) and energy intensive Industry goals currently Include reducmg energy consumption, reducing processing times, reducing emissions, increasing furnace life, improving heat transfer during heating and c o o h g , reducing distortion, and improving uniformity of structure and properties,both withim a given part and throughout an entire productton quantity.

I Key Words
A, A, A,, age hardenrng aging annealmg art~f~cial aging artificral gas atmospheres ausform~ng austemperlng autoquenchmg ha~n~te batch furnaces C-C-T diagram carburlzing coherency contrnuous furnaces crit~eal coollng rate
drspers~on hardenmg electron-beam hardening equ~l~bnum phase dragram flame hardenrng flu~d~zed-bed furnaces full anneal gram slze refinement hardenahility hardness heat treatment homogenrzatlon ~nducuon hardening ~nterstit~al solut~on ionitrid~ng rsothermal anneal rsothermal-transformation diagram Jomlny test laser-beam hardening martempering martens~te natural aging nitrrding normahe overaged pearlrte phase transformat~on strengthenrng polymer quench preap~tation harden~ng process anneal processing heat treatments quench and temper quench crackmg quenchant quench~ng recrystalhzat~on res~dual stresses reta~ned austenrte solrd-solutron strengthening solut~on treatment sphero~drzatlon stram hardening stress rel~ef anneal subst~tutional solut~on surface hardening syntheuc quenchant T-T-T diagram tempered martensrte temper~ng processing thermomechan~cal vapor jacket

I Review Questions
1 What is heat treatment? 2 What types of properties can be altered through heat treatment7 3 Why should people performing hot formlug or weld~ng he 4. 5.
9. Why are the hypereutecto~dsteels not furnace cooled fro

6.

7
8.

the all-austenrteregion? 10. What is the major process d~fference between full anneah aware of the effectsof heat treatment? and normalnmg? What is the broad goal of the processing heat treatments? Crte 11. m l e uormalrz~ng is less expensive than a full anneal, so some of the specific object~ves that may be sought manufacturers clte cost saving through the use of a full Why m~ght equrl~hrrum phase diagrams he useful aids rn neal. How is t h ~achreved? s des~gnmg and understanding the processing heat treatments? 12. What are some of the process heat treatments that can hepe What are the A , , Az, and A, lines? formed w~thout reausten~tizrng the material (heating abov What are some possible objectives of annealrng operations? the A, temperature)? Whrle full anneals often produce the softest and most duct~le 13. What types of steel would be cand~dates for a process anneal structures what may he some of the oh~eCUons orundes~rable Spheroidizat~ou? features of these treatments7

4 How m~ght steel composition influence the selectton of a 36. What conditions are used to standardize the quench in the
:at he on xte ess raL c e er. Jns hat nal
lCts

dirgY ife, ing dre

Jominy test? 37. What is the assumption that allows the data from a Jominy nonferrous metals often heat treated? test to be used to predict the properties of various locations on 6 What are the SIX malor mechanrsms that can be used to a manufactured product? increase the strength of a metal? 38. What is hardenability? What capabilities are provided by highy. What is the most effective strengthenmg mechanrsm for the hardenability materials? nonferrous metals? 39. What are the three stages of liquid quenching? d What are the three steps m an age-hardenmgtreatment? 110. What aresome of the major advantages and disadvantages of & What is the drfference between a coherent precrpitate and a a water quench? dlstrnct second-phase part~cle? Why does coherency offer stg- 41. Why is an oil quench less likely to produce quench cracks than n~ficant strengthening7 water or brine? !O. What IS overaging? 42. What are some of the attractive qualities of a polymer or syn!l. What is the dfference between natural and arhGc~al aging? thetic quench? Whrch offers more flenbrltty? Over which does the englneer 43. What are some undesirable design features that may be prehave more control? sent in parts that are to be heat treated? ! 2 Why IS it more d~fhcult to understand the nonequrlibrium 44. Why would the residual stresses in steel he different from strengthening treatments? the residual stresses in an identically processed aluminum part? 3 What types of heatrng and cooling conditions are nnposed m an I-T or T-T-T diagram? Are they realistrc for the processmg 45. What are some of the potentially undesirable effects of residof commeroial items? ual stresses? !4. What are the stable equilibrium phases for steels at tempera- 46. What causes quench cracking to occur when steel is rapidly tures below the A, temperature7 cooled? 5. What are some nonequilibnum structures that appear in the 47. Describe several techniques that utilize simultaneous transT-T-T diagram for a eutectoid composrtion steel? formation to reduce residual stresses in steel products. 16. Whrch steel structure is produced by a diffusionless phase 48. What is thermomechanical processing? 49. What are some of the methods that can he used to selectively change? alter the surface properties of metal parts? !7 What n the malor factor that Influences the strength and hard ness of martenslte? 50. What are some of the attractive features of surface hardening 28 Most structure changes proceed to completron over tnne.The with a laser beam? martenslte transformatron IS merent. What must be done to 51. How can a laser beam be manipulated and focused? How are produce more martensite a a partially transformed structure? these operations performed with an electron beam? 29. Why 1s retarned austenlte an undes~rahle structure in heat- 52. What is carburizing? Weated steels? 53. Why does a carburized parthave to be further heat treated after 10 What types of steels are more prone to retarned austenite? the carbon is diffused into the surface? What are the various 31 Why are miatens~tic structures usually tempered before bemg options? put Into use? What propertres increase during tempering? 54. In what ways might ionitriding he more attractive than conWh~ch ones decrease? ventional nitriding or carburizing? 32. In what ways is the quench-and-temper heat treatment siml- 55. For what type of products or product mixes might a batch furlar to age harden~ng? How are the property changes d~fferent nace be preferred to a continuous furnace? in the two processes? 56. What aresome possible functions of artificial atmospheres in 33 What is a C-C-T diagram? Why a it more useful than aT-T-T a heat-treating furnace? diagram? 57. How are parts heated in a fluidized-bed furnace? What are 34 What two features combine to determme the structure and some of the attractive features? properties of a heat-treated steel? 58. In what ways might the heat treatment of metals actually be 35 How do the via~ous locatrons of a Jomrny test specimen coran energy conservation measure? relate w ~ t h cooling rate?

processing heat

treatment?

d Other than increasrng strength, for what three purposes are

Pro
rom
I. A number of heat treatments have been devised to harden the surfaces of steel and other enpneering metals Conslder the followrng processes: a Flame harden~ng h Induction hardening c. Laser-beam hardenmg d. Carburizmg e Nitridmg f Ion~triding or each of these processes, provide information relat~ng to: (1) A baslc description of how the process works (2) Typtcal materials on whch the process is performed
(3) Type of equrpmeut requlred (4) ?h,~caltimes, temperatures, and atmospheres requred (5) Qpical depth of hardenrng and reasonable 1im1t.s (6) Hardness achievable (7) Subsequent treatments or processes that might be requlred (8) Informat~on relating to drstort~on andlor stresses (9) Ah~hty to use the process to harden selectwe areas 2. Investigate the nme areas IU Problem 1for one of the lesser known surface-mohfication treatments, such as Ion rmplantation, bondmng, chromiung, or other srnnlar techurques. 3. This chapter presented four processing-type heat treatments whose primary oblecttve a to soften, weaken, enhance

per>eve

1 16

CHAPTER 5 Heat Treatment

4. Anumber of different quenchants were discussed in the chap

ductility, or promotemachinability. Consider each of the following processes as they are applied to steels: a. Full annealing b. Normalizing c. Process annealing d. Spheroidizing Provide information relating to: 1. A basicdescription of how the process works and what its primary objectives are 2. Typical materials on which Lhe process is performed 3. Type o i equipment used 4. Typical times temperatures, and atmospheres required 5 . Recommended rates of heating and cooling 6. Typical properties achieved

ter, including brine, water, oil, synthetic polymer mixes, and even high-pressure gas flow. Select two of these and investigate the environmental concerns that may accompany their use. 5. Traditional manufacturing generally separates mechanical processing (such as forging, extrusion, presswork, or machining) and thermalprocessing (heat treatment),and applies them as sequential operalians. Ausforming was presented as an example of a thennomechanical process where themechanical and thermal processes are performed concurrently. When this is done, the resulting structures and properties are often quite different from the traditional. Identify another thermomechanical process and discuss its use and attributes. 6. It has been noted that hot oilis often amore effective quench than cold oil. Can you explain this apparent contradiction?

arpenter claw hammers a r e actually a rather sophisticated metallurgical product, since the loadings differ for the various locations. The claw sees static bending, while the eye sustains impacts, and the striking face sees impact contact with potentially hard surfaces.

As one might expect, the optimum properties and microstructures vary with location. While hammer handles have been made from a variety of materials, including heat-treated tubular 4140 steel, our problem will focus on the head.

The following information was obtained from the American National Standards Institute (ANSI) Standard 8173.1, "American National Standard Safety Requirements for Nail Hammers.!This i s a voluntary specification (recommenda-

According to the ANSI specification: "Hammerheads shall be forged in one piece from spehot rolled carbon steel bars;while the speccial ification for steels ranging from 1045 to ,088, two major manufacturers of high-quality tools have

EPm d
ate
3.

r * n k ! )
I as

mmd

s is lite
ne-

och
?

The hammer striking face "shall be hardened and tempered to a Rockwell hardness of not less than C 40 or more than C 60, and the steel directly behind the striking face shall be a toughened supporting core gradually decreasing in hardness." "Hammer claws shall be hardened to a Rockwell hardness of not less than C 40 or more than C 55 for a minimum length of i-inch from the tip end; the remaining length to the base of theV-slot shall be of the same through hardness, or shall contain a toughened core gradually decreasing in hardness to the While there is no specification for the eye region, many manufacturers prefer forthis area to have the greatest toughness (i.e., even softer still-as low as R,25!). ' In essence, we are looking at a single piece of heattreated steel that preferably exhibits different properties at different locations. For example, one top-quality hammer has astriking face of R , 55 to 58, coupled with a claw of R , 46 to 48. Another top-quality hammer has a striking face hardness of Rc 50 to 58, claw tip hardness of R , 47

i
1

(R, 41 to 48) to prevent chipping-a characteristic feature of this particular manufacturer. Fixing our material as the above-used 1078 hot-rolled steel bar, and using forging as our shaping process: 1. What problems might be expected if the material on the striking face were too hard?Too soft? Consider each with respect to possible liability. 2. Describe some heat-treatment processes or sequences that could be used to produce a quality product like those des~ribed above. 3. Discuss the methods of heating, cooling or quenching, target temperatures, and so on that you are proposing to accomplish this task. 4. Finally, how might you duplicate the rim softening being achieved by the cited manufacturer? 5. Inexpensive hammers frequently use a single material and single heat treatment, rendering the properties similar for all locations. What are the major compromises? If these hammers were to be used by a professional carpenter, how might they be deficient7

FERROUS METALS AND ALLOYS

6.1 INTRODUCTION TO HISTORYDEPENDENT MATERIALS 6.2 FERROUS IVIETALS 6.3 IRON 6.4 STEEL Solidification Concerns Deoxidation and DegassifiCation Plain-Carbon Steel Alloy Steels AISI-SAE Classification System

11 6.1 INTRODUCTION TO HISTORY-DEPENDENT MATERIALS

Engineering materials are available with a wide range of useful properties and charac. teristics. Some of these are inherent to the particular material, hut many others can be varied by controlling the manner of production and the details of processing. Metals are classic examples of such "history-dependent'' materials. Their final properties are clearly allected by their past processing history.The particular dctails of the smelting and refining process control thc resulting purity and the type and nature of any influential contaminants. The solidification process imparts structural features that may be transmitted to thc final product. Preliminary operations such as the rolling of sheet or plate often impart directional variations to properties. and their impact should be considered during subsequent processing and use. Thus, whilc it is easy to take the attitude that "metals come Crom warehouses," it is important to recognize that aspects of prior processing can significantly influencc further operations as well as the final properties o l the product.The breadth of this book does not permit full coverage of the processes and methods involved in the production of engineering metals, but certain aspects will be presented because of their role in affecting subsequent performance.

6.2 FERROUS METALS

In this chapter we will introduce the majorferrous (iron-based) metals and alloys, summarized in Figure 6-1. These materials made possible the Industrial Revolution and have been the backbone of modern civilization. We see them everywhere in our livesin the cars we drive, the homes in which wc live, the cans we open, and the appliances that enhance our standard of living. Numerous varieties have been developed over the years to meet the specific needs of various industries. The developments and improvements have continued, with recent decades seeing the introduction of a number of new varieties and even classes of Cerrous metals. According to the American Iron and Steel Institute, over50% of t l ~ e steels made today did not exist 10 years ago and over 70% of the steel used in automotive production meets this criteria.The newer steels are stronger than ever, easier to shape. and more corrosion resistant. In addition, all steel is recyclable, and this rccycling does not involve any loss in material quality. In fact, more steel is recycled each year than all other materials combined. including aluminum, glass, and paper. Because steel is magnetic, it is easily separated and recovered from demolished buildings, junked automobiles, and discarded

118

r- .--TastlroG .- -- Malleable lron

Classification of rous metals and

-,
Compacted Graphite lron Austempered Dunrle lron

SECTION 6.3 lron

I19

-_ - -_

Plein-CaroonSteels

_ .- .
-

-.

Alloy Steels

-.

-.

7 - - - - - -

Low-Aloy Steels

Microalloyed Steels

White lron

Tool Steels I_-L__I

appliances. In 2003,nearly 70 million tons of steel were recycled m the United States, for an overall recycling rate of nearly 71%.The recycling rate for steel cans was 60 2%, 89.7% for large apphances, and 102.9% for automob11es.That"scorrect-more stcel was recovered from scrap cars than was used in the production of new vehicles! Each ton of recycled steel saves over 4000 pounds of raw materials and 74% of the energy required to make a ton of new steel.

For centuries,iron has been the most important of the engineering metals. While iron is the. fourth most plentiful element in the earth's crust, it is rarely found in the metallic state. Instead, it occurs in a variety of mineral compounds, known as ores, the most attractive of which are iron oxides coupled with companion impurities. To produce metallic iron, the ores are processed in a manner that breaks the iron-oxygen bonds (chemical reducing reactions). Ore, limestone, coke (carbon), and air are continuously introduced into specifically designed furnaces and molten metal is periodically withdrawn. Within the furnace,other oxides (which were impurities in the original ore) will also be reduced. All of the phosphorus and most of the manganese will enter the molten iron. Oxides of silicon and sulfur compounds are partially reduced, and these elements also become part of the resulting metal. Other contaminant elements, such as calcium, magnesium, and aluminum, are collected in the limestone-based slag and are largely removed from the system. The resultingpig iron tends to have roughly the following composition: Carbon Manganese Phosphorus Silicon Sulfur
3.04.596 0.15-2.5% 0.1-2.0% 1.0-3.0% 0.054.1%

A small portion of this iron is cast directly into final shape and is classifiedas cast iron. Most commercial cast irons, however, are produced by recycling scrap iron and steel, with the possible addition of some newly produced pig iron. The metallurgical properties of cast iron have been presented in Chapter 4, and its melting and utilization in the casting process will be developed in Chapters 11 through 13. Most pig iron, however, is further processed into steel.

F A 6.4 STEEL
Steel is an extremely useful engineering material It offers strength, rigidity, and durabihty. From a manufacturmg perspective,its fonnabihty,joinabihty, and paintabihty, a& well as repairabihty,are all attractwe For the past 20 years, steel has accounted for about 55% of the we~ght of a typical passenger car and is expected to continue at t h s level. Whlle the autoinotive and construction industr~es are major consumers of steel, the mater~al is also used extensively in containers, appliances, and machinery as weU as thc infrastructure of such industr~es as oil and gas. The manufacture of steel is essentially an oxidat~on process that decreases the amount of carbon, sihcon, manganese, phosphorus, and sulfur in a molten rmxture of pig lron and/or steel scrap. In 1856, the Kelly-Bessemer process opened up the industry by enabling the manufacture of commercial quantities of steel. The open-hearth process surpassed the Bessemer process in tonnage produccd m 1908 and was producing over 90% of all steel in 1960. Currently, most of our commercial steels are produced by a variety of oxygen and electr~c-arc furnaces In many of the current processes, air or oxygen passes over or through the molten metal to drive a varlety of exothenn~c refining reactions. C ~bon I oxidlzes to fomgaseous CO or CO,, wh~ch then exits themelt. Other elements, such as s111conand phosphorus, are similarly ox~dized and, being llghter than the mctal.rise to be collected in a removable sIag.At the same time, oxygen and other elements from the reaction gases dicrolve In the molten metal and may later become a cause for concern.

SOLIDIFICATION CONCERNS
Regardless of themethod by which the steel is madesitmust undergo a change from hqu~d to solid before it can become a usable product.The liqu~d can be converted directly into finish-shape steel castings or solidified Into a form suitable for further processing. In most cases,some forin ~Econtinuous casting produces the feedstock material for subsequent forging or rolhng operations Prior to sohd~fication, we want to remove as much contaminarlon as possible The molten metalis poured from the steelmaking furnaceslnto containment vessels, known as iadle~. Historically, the ladles simply served as transfer and pouring containers,but they have recently emerged the slte for additional processmg. LudIe meruliurgy refers to a variety of processes deslgned to provide final purlfrcat~onand to fine-tune both the chemistry and temperature of the melt.Alloy a d d ~ t ~ ocan n s be made; carbon can be further reduced, d~ssolved gases can be reduced or removed;ilnd steps can be taken to control subsequent grain size, lun~t ~nclusion conte~it, reduce sulfur, and control the shape of any included sulfides. Stirring, degasslilg, reheating, and the 1nJechon of powdered alloys or cored wire can all be performed to Increase the cleanliness of the steel and provide for tighter control of the chem~stry and properties. The processed hquid 1s thcn poured Cram these ladlm into molds or some form of connnuous L U S I ~ T usually , through a bottom-pourlng process such as the one sllown By extracting the metal from the bottom of the ladle, slag schematically in Figure 6-2. and floating matter are not transferred, and a cleaner product results. Figure 6-3a ~llus-

Lever for pourtng

I

FIGURE 6-2 D~agram of a bottomGraphlte pourlng hole

pourlng ladle

Ferrous Metals and Alloys

Steel 1s an extremely useful engineering material. ~t offers strength, ligidlty, and durabillty From a manufacturing perspective, its formablhty, joinabillty, and paintabihty, as well as repairability,are all attractive.For the past 20 yeats, steel h a accounted for =bout 55% of the weight of a typical passenger car and is expected to continue at this level WhlIe the automotive and construction lnduslries are major consumers of steel, (he material IS also used extensively in conlainers, apphances, and machmery as well as the infrastructure of such mdustries as 011and gas. l essentially an ox~dation process that decreases the The manufacture of ~ t e e is amount of carbon, silicon, manganese, phosphorus, and suKur In a molten mmxture of pig Iron and/or steel scrap In 1856,the Kelly-Besscmer process opened up the lndus. try by enabling the manufacture of Commercial quant~tles of steel.The open-hearth process suwassed the Bessemer process in tonnage produced in 1908 and was producing over 90% of all steel m 1960. Currently, mast of our commercial steels are prodnced by a variety af oxygcn and electric-arc fusnaces In many of the current.processes,air OToxygen pases over or through (he malten metal to drtve a variety of exothermic refining reactions Carbon oxidizes to formgaseous CO or CO,, which then exits the melt. Other elements, such as siltcon and phosphorus, ate similarly oxidzed and, being llghter than the metal, nse to be collected m a removable slag. At the same time,oxygen and other elements from the reaction gases dtssolve in the molten metal and may later become a cause for concern

SOLIDIFICATION CONCERNS
Regardless of the method by which the steel 1s made,~t must undergo a change from Ilquid to solld before it can became a u.;able product,The liquid can he coltvested directly Into fin~sh-shape steel castings or rolldd~ed into a form suaable for further procesmg. In most cases, some form of continuous casting produces the feedstock materlal for subsequent forgtng or rolling operations Prlor to solidification, we want to remove as much contarninatlon as posslble.The molten metal is poured trom the steelmaking furnaces ~ n t o contalnmenl vessels, known as ladlcs.Historically,the ladles simply served as transfer and pounng containers, but they have recently emerged as the site for addltlonal procewmg Ladle metallurgy refers to a variety of processes des~gned to provide final purification and to fine-tune both the chemistry and temperature of the melt. Alloy additions can be made; carbon can be further leduced; dissolved gases can be reduced or removed; and %teps can be taken to control subsequent gram size, llmit inclusion content, reduce sulfur, and control the shape of any included sulfides. Stlrrmg, degassing, reheating and the injection of powdered alloys or cored wlre can all be performed to Increase the cleanlmess of the steel and provide for hghter control of the chemistry and properties. Tne processed liquld u then poured from these ladles into molds or some form o f continuous caster, usually through a bottom-pounng process such as the one shown schemat~cally in F~gure 6-2. By extracting Lhe metal from the bottom of the ladle, slag and floating matter are not transferred, and a cleaner product results. Figure 6-3a illus-

Graphlte stopper
raphtte pouring hole

of a bottomFIGURE 6-2 D~agrarn

pourlng ladle

SECTION 6.4 Steel

121

Mold d~scharge rack

Vertlcal guide roll rack

siring mill

Reheat furnace

. Torch

cutoff

FIGURE 6-3 (a) Schematic representation of the continuous casting process for producing billets, slabs, and bars. (b) Simultaneous continuous casting of multiple strands. (a) (Courtesy of Materials Engineering, Penton Pubhshing, New York, NK)

trates a typical continuous caster, m which molten metal flows from a ladle, through a tund~sh, and into a bottomless, water-cooled mold, usually made of copper Cooling is controlled so that the outside has solidified bcfore the metal exm the mold Dlrect water sprays further cool the emerging metal to completc the solid~ficahon.The solid metal can then be cut to dcsired length, or, slnce the cast solid is st111hot, it can be bent and fed horidontally through a short reheat furnace or d~rectly to a rolling operatlon.If the size and shape of the mold are varied,products can be cast with a variety of cross sechons w~th names such as slab, bloom, billet, and strand. Figure 6-3b dep~cts the s~multaneous castmg of multiplc strands Compared to the casting of discrete ingots, continuous casting offers significant reduction m cost, energy, and scrap. In addition, the products have improved surfaces, more uniform chemical composition, and fewer oxlde mclusions.

DEOXlDATlON AND DECASSlFlCATlON

As a result of the steelmaking process,large amounts of oxygen can be dissolved in the molten metal. During the subsequent cooling and solidification, the solubility levels decrease s~gn~ftcantly, as shown in Figure 6-4, and the oxygen and other gases are rejec~ed.The rejected oxygen frequently hnks w ~ t h carbon to produce carbon monoxide

122

C H ~ E6 R Ferrous Metals and Alloys

Pony ladle

Temperature

FIGURE 6-4 Solub~l~ty of gas tn a

metal as a function of temperature showtng s~gnif~cant decrease upon sol~dlf~cation.

Method of degass~ng steel by pourtng through a vacuum.

FIGURE 6 5

gas, which may escape through the liquid or become trapped to produce a porous solid. The bubble-induced porosity may take various fonns ranging from small, d~spersed voids to large blowholes. While these pores can often be welded shut during subsequent hot forming, some may not be fully closed, and others may not weld upon closure. Cracks and internal volds can persist into a finished product. Porosity problems can often be avoided by either removing the oxygen prior to solidification or by making sure it does not reemerge as a gas Aluminum, ferromanganese, or ferrosllicon can be added to molten steel to provlde a matenal whose affintty for oxygen is higher than that of carbon The rejected oxygen then reacts with these deoxidizer additions to produce solid metal oxides that are either removed from the molten metal or become dispersed throughout the structure. While deoxidizer addiQons can effectively tie up dissolved oxygen, small amounts of other gases, such as hydrogen and nitrogen, can also have deleterious effects on tha performance of steels.This is particularly important for alloy steels because the solubihty of these gases tends to be increased by alloy additions, such as vanadium, niobium, and chrommm. Alternative degassing processes have been devised that reduce the amounts of all dissolved gases. Figure 6-5 illustrates one form of vacuum degasswag, in whch an ingot mold is placed m an evacuated chamber, and a stream of molten metal passes through a vacuum during pounng. Because a large amount of exposed surface is created during the pouring operation, the vacuum is able to extract most of the dissolved gas. An alternative to vacuum degassingis the consumable-electroderemeltwg process, where an already-solidified metal electrode replaces the ladle of molten metal. As the electrode is progressively remelted,molten droplets pass through a vacuum, and the extremely high surface area again provides an effective means of gas removal. If the melting is done by an electric arc, the process is known as vacuum arc remelting (VAR).If induction heating is used, the process becomes vacuum rnduction melting ( V I M ) .Both are highly effective in removing dissolved gases, but they fall to remove any nonmetallic impurities that may be present in the metal. The electroslag remelting process (ESR),shown in Flgure 6-6, can be used to produce extremely clean, gas-free meta1.A solid electrode is again melted and recast using an electric current, but the entire remelting is conducted under a blanket of molten flux Nonmetallic impur~ties float and are collected in the flux, leaving a newly solidified metal structure with greatly improved quality. No vacuum is required, since the molten material is confined beneath the flux and the progressive freezing permits easy escape for the rejected gas.This process is simply a large-scale version of the electroslag welding process that will be discussed in Chapter 34.

1,
I

Electrode clamp Stinger Electrode

c -

Cooling water in Movable mold Slag pool Molten metal pool

Ingot

Cooling water out

Starting plate Water cooling

ial
ib1

I
i
I

FIGURE 6-6 (a) Production o f an ingot by the electroslag remelting process; (b) schematic

representation of this process showing the starting electrode, melting arc, and resolidified ingot. (Courtesy of Carpenter Technology Corporation, Reoding, PA.)

PLAIN-CARBON STEEL
While theoretically an alloy of only iron and carbon, commercial steel actually contains manganese,phosphorus, sulfur, and silicon in significant and detectable amounts.When these four additional elements are present in their normal percentages and no minimum amount is specified lor any other constituent, thc product is refcrred to as plaincarbon steel. Strength is primarily a function of carbon content, increasing with increasing carbon, as shown in Table 6-1. Unfortunately, the ductility, toughness,and weldability of plain-carbon steels decrease as the carbon content is increased, and hardenability is quite low. In addition, the properties of ordinary carbon steels are impaired by both high and low temperatures (loss a f strength and embrittlemenl, respectively), and they are subject to corrosion in most environments. Plain-carbon steels are generally classed into three subgroups based on their carbon content. Low-carbon .?reelshave less than 0.20% carbon and possess good formability (can be strengthened by cold work) and weldability.Their structures are usually ferrite and pearlite, and the material is generally used as it comes from the hot-forming or cold-forming processes, or in the as-welded condition. Medium-carbora steels have between 0.20 and 0.50% carbon, and they can be quenched to form martensite or bainite if the section size is smaIl and a severe water or brine quench is used.The best balance of properties is obtained at these carbon levels, where the high toughness and ductility of the low-carbon material is in good compromise with the strength and hardness that come with higher carbon contents.7llese steels are extremely popular and find numerous mechanical applications. High-carbon steels have more than 0.50% carbon.Toughness and formability are quite low, but hardness and wear resistance are high. Severe quenches can form martensite, but hard448 65 enability is still poor. Quench cracking is often a problem when the material is pushed to its limit. Figure 6-7 depicts the characteristic properties of low-, medium-, and high-carbon steels using a

0.30%

124

CHAIIEH 6 Ferrous Metals and Alloys

A comparison of low-carbon, medrum-carbon, and h~gh-carbon steels in terms o f the~r properbes. relat~ve (a) balance Low-carbon of steel
FIGURE 6-7

E P m Z I
5

has excellent ductll~ty and fracture resistance, but lower strength; (b) medium-carbon steel has balanced properbes, (c) highcarbon steel has high strength and hardness at the expense of ductility and fracture resistance

2 2 x

H! 1 !,q
0 .

rn I c

a
0

~j

rn c

n 2 01
C

(D

0)

r:
5

2
5 3

5
r)

" ,
(a1

iD m

G1
tj
(b)

'?!
(cl

Low carbon steel

Medlum-carbon steel

H~gh-carbon steel

balance of properties that shows the offsetting characteristics of "strength and hardness" and "ductility and toughness." Compared to other engineering materials, the carbon steels offer h~gh strength and hlgh sbffnesa, coupled wlth reasonable toughness. Unfortunately, they atso rust easily and generally requue some form of surface protection, such as palnt, galvanizin other coating The platn-carbon steels are generally the lowest-cost steel mater~al should be glven frat consideration for many applications Then limitat~ons, however, may become restrictive.When improved performance 1s required, thesc steels can be upgraded by the addltlon of one or more alloying elements.

ALLOY STEELS
The differentiation between plan-carbon and alloy steel is often somewhat arbitrary Both contain carbon, manganese, and usually sihcon. Copper and boron are possible additions to both classes. Steels containing more than 1.65% manganese, 0.60% s~licon, or 0.60% copper are usually designated as alloy steels. Also, a steel IS cons~dered to be an alloy steel ~f a defii~te or mlnkmum amount of other alloy~ng element 1s specified.The most common alloy elements are chrom~um, nickel,molybdenum,vanadium, tungaten, cobalt, boron, and copper, as well as manganese,sil~con, phosphorus, and sulfur in amounts greater than are normally present. If the steel contains less than 8% of total alloy addit~on,~t IS considered elements are high-alloy steels to be a low-alloysteel. Steels with more than 8% alloy~ng In general, alloying elements are added to steels in small percentages (usually less than 5 % ) to Improve strength or hardenability, or in much larger amounts (often up ta 20%) to produce specla1 properties such as corrosion reslstance or stability low temperatures Add~tions of manganese, silicon, or alummum may he made the steelmaking process to remove dissolved oxygen from the melt. Manganese, nickel, and copper add strength by forming solid solutions m fernte. Chromlu um, molybdenum, tungsten, and other elements Increase strength hy form~ng di second-phase carbides N~ckel and copper can be added in small amounts to impro corrosion res~stance. N~ckel has been shown to Impart increased toughness and impa reslstance, and molybdenum helps resist embrittlement. Zircoll~um, cerium, and calcium can also promote increased toughness by controlling the shape of inclus~ons. Machinability can be enhanced through the formation of manganese sulfides or by adhtions of lead, bismuth, selenium, or tellurium. Still other additions can be used to provide ferrite or austenite grain size control. Select~on of an alloy steel still be identifymg the proper carbon content Table shows the effect of carbon on the strength o quenched-and-tempered alloy steels.The streh values are s~gnificantlyhigher than those Carbon Content Mpa ksi Type of Steel Table 6-1, reflecting the difference bet 0 30% 4130 1030 150 annealed and quenched-and-tempered 4330 0 30% 1030 150 crostructures The 4130 steel has abou 8630 0 30% 1030 150 alloying elements, 4330 has 3.0%, an 4140 0 40% 1241 180 about 1.3%, yet all have the same quenched-and-i 4340 0 40% 1241 180 tempered tensile strength. Strength and hard depend primarily on carbon wntent.The primar uata from ASTM Speclt~catlon A732

6.4 Steel SECTION

125

Element
Alum~num Btsmuth Bmn
Chrumkum

Percentage
0.95-1 30

Copper
Lead Manganese

Primary FunCtton Alloying elernelit in nltridingsteels Improves rnach~nahtl~ty Pewerful hardennb~hly agent Increasc of hardenability Corrosrnn res~stancc Corros~on reslstaner
lmpmed machinabil~ty Combme wtth sulfur lo prevent bnttlenei? lncreasrs hardendh~ltty bq lowcnng transfonnarton polnli and causing transformatmnsto be sluggish Slahle carhldcamh~b~ts grarn growth Toughener Corros~on resistance increases strength

th
19-

Molyhdenurn Nickel

0.2-5 2-5

or ld tr, :n

L 2 2 0
0.24 7
2 Higher percenlages
008-0 -

15

Spnng sreek lmprovrs magnetic properti6 Free-machinmg properties F~xeccarbnn i n inort particles
Rsduwsmartens~tichardncis In chromiuin %feels Hardness at high temperatures Stilhleearb~dts, lncrcsses slrength wluleretammg ductility, Promotes frne gram qtructure

role of an alloy addition is usually to increase har&nabrliry, but other effects are also possible, such as modified toughness or machinability The most common hardenab~l~ty-enhancing elements (ld order of decreasing effectiveness) are manganese, molybdenum, chromium,sil~con,andnickel. Boron 1s an extremely powerful hardenability agent. Only a few thousandths of a percent are sufficient to produce a significant efEect in low-carbon steels,but the results dlm~nish rapidly wlth increasing carbon content.Sinceno carbide formation or ferrite strengthening accompanies the addition, improved machimabil~ty and cold-forming characteristics may favor the use of boron in place of other hardenab~hty additions.Small amounts of vanadium can also be quite effective, but the response drops off as the quantity is increased. Table 6-3 summarizes the primary effects of the common alloying elemants in steel. A working knowledge of this information may be useful in selecting an alloy steel to meet a given set of requirements. Alloying elements are often used in combinat~on, however, resulting in the immense variety of alloy steels that are comnlercially available. To provide some degree of simplification, a class~fication system has been developed and has achieved general aweptance in a variety ofindustries.

AlSl-SAE CLASSIFICATION SYSTEM The most common classification sche&e for alloy steels is the AISI-SAE zdentij7cafion sysf#m.T h i i system, which classifies alloys by chemistry, was started by the Society of Automotrve Engineers (SAE) to provide some standardization for the steels used in the automotive industry. It was later adopted and expanded by the American Iron and Steel Institute (AISI) and has been incorporated into the Universal Numbering System l l lengineering metals. Both plain-carbon and low-alloy that was developed to include i steels are identified by afour-digit number, where the ftrst number indicates the major anoy~ng elements and the second number designates a suhgrouplng withln the major alloy system.These first two dlgits can be interpreted by looking themup on alist,such as the one presented in Table 6-4. The last Two digits of the number Indicate the approximate amount o f carbon, expressed as "points:' where one pdnt is equal to 0.01 %. Thus, a 1080ateel would he a plain-carbon steel with 0.80% carbon.Sirmlarly,a 4340 steel

:6 Fermus Metals and Alloy6

AlSI Number
lxxa
l k

Type
Catbob steds Rbrn &on Free cuttip (8)

llxx I2xx lkx 13%~ 2xxx

Jxex
~KXX

Fxee m b a g ($2 and iP) Hjghmang~nese IIiehrnanganW

Ni&d steels

NockeI-chrormum MOlybdenum

.fDxx 4Ixx
43xx
44x2

MQ
Mo, Cr Mo, Cr, Ni
NO

0,154z30
0.404.10
'

O.OBa3Z

1.65-2 W

0 . 4 0 0 W

02M.M
035460 0.15-030 01 5 4 40 0.26R30

d6xx
47x1

Mo,Nf (law)

Mo,Cr,Nl
g o , Nz (higb) Ckramim

0 70-2 M) 0 90-1 20
3.25-3.75

0 354 55

48m
Sxxx SQxx 51xx
6 m

0 20-0 60 0.7bl.U;

Chromum-vdum

M x x
8xxs 8lXx 86xx 87xx
B8xx

0 . 5 C k 1 . 1 0
R,G,Mo
0 2M.40 0 4W70 0.40-0.70 0.404 70
Qt h a
0.3M 55 O.W.60 O.W.60 D.4M.W
O.O&QlS

0.15-025 D.2M.30
0.3N.40

9xxa

92xx 93xx
Y4xx

Highdiem NG,Mo
Nl,Q,Ma

3.00-3 50

1 m1.40

0W . 1 5

MdPtrird*&-Uh$gP;8se(fflbk addition of boron. S'lmilEirl~ an L in this position indicates a lead maehinab'e A left& prefix may aIso he employed to designate the process used t produce the steel, such as E for electric furnace.

structural steels.

SECTION 6.4

Steel

127

SELECTING ALLOY STEELS From the previous discussion it is apparent that two or more alloying elements can often produce similar effects. Thus, when properly heat treated, steels with substantially different chemical compositions can possess almost identical mechanical properties. Figure 6-8 clearly demonstrates this fact, which becomes particularly important when one realizes that some alloying elements can be very costly andothers may be inshort supply due to emergencies or political constraints. Overspecification has often been employed to guarantee success despite sloppy manufacturing and heat-treatment practice. The correct steel, however, is usually the least expensive one that can be consistently processed to achieve the desired properties.This usually involves taking advantage of the effects provided by all of the alloy elements. When selecting alloy steels, it is also important to consider both use and fabrication. For one product, it might be pe~missible to increase the carbon content to obtain greater strength. For another application, such as one involving assembly by welding, it might be best to keep the carbon content low and use a balanced amount of alloy elements, obtaining the desired strength while minimizing the risk of weld cracking. Steel selection involves defining the required properties, determining the best microstructure to provide those properties (strength can be achieved through alloying, cold work, and heat treatment, as well as combinations thereof), determining the method of part or product manufacture (casting, machining, metalforming, etc.), and selecting the steel with the best carbon content and hardenability characteristics to achieve those goals. HIGH-STRENGTH LOW-ALLOY STRUCTURAL STEELS Among the general categories of alloy steels are: (1) the constructional alloys, where the desiredproperties are typically developed by a separate thermal treatment and the specific alloy elements tend to be selected for their effect on hardenability, and (2) the high-strength low-alloy (HSLA) or microalloyed types, which rely largely on chemical composition to develop the desired properties in the as-rolled or normalized condition. The constructional alloys are usually purchased by AISI-SAE identification, which

Yield strengthd&

the
the wd I to
-

ud.
t 5)

Water quenched

SAE
A

md

V
0

all, ven her

1330 2330 3130 4130 5130 6130

T !
-

Reduction

A 01

W
.
-

m-v
A
R7

ins

80

I I I I I I 120 160 200 Yield strength. 1000 psi

I I 2400

I I l l 20 40 60 and red. of area % Elongat~on

12d

CHAPTER 6 Ferrous Metal? and Alloys

Strength Properties Grnup

Columb~um or vanad~urn

Chemical Compositions"%) Mn St Ch

ksi

MPa

ksr

Low manganesevanadium

Mangamse-

copper

0 25

L . 2 0

5 30

Manganese-

vanad~um-copper

n14Uha~004%P,0U5%S,~dOZ09~C~

effectively specifies chemistry.The HSLA matefials generally focus on product (size and shape) and desired properties. When steels are specified by mechanical properties, the supplier or producerls free to adjust the chemistry (w~ihin limits), and substantial cost savings may resu1t.T~ assure sucwss, however, it is important that all of the necessary properties be specified. The HSLA materials provide increased strength-to-weight compared t6 conventional carbon steeis for only a modest increase in eost.They are available in a variety of forms, including sheet, strip, plate, structural shapes, and bars. The dominant property requirements generally are high yield strenglh, good weldability, and acceptable corrosion resistance. Duct~lity and hardenability may be somewhat lim~ted, however The ~ncrease in strength, and the resistance to martensite formation in a weld zone, is obtained by cnntrolIing the amounts of carbon,rnanganese, and silicon, with the addition of small amounts of niobium, vanadium, titanium, or other alloys. About 0 2 % copper can be added to improve corrosion resistance. Because of their higher yield strength, weight savings of 20 to 39% can often be achieved with no sacrifice to strength os safety. RoIled and welded HSLA steels are being used in automobiles trains, bridges, and buildings Because o f their low alloy content and high-volume applicabon,their cost 1s often little more than that ofthe ordinary plain-carbon steels.Table 6-5 presents the chemistries and properties of several of the more common types.

MICROALLOYED STEELS I N MANUFACTURED PRODUCTS In terms of both cost and performance, microalloyed steels occupy aposition between carbon steels and the alloy grades, and they are being used increasingly as substitutes for heat-treated steels in the manufacture of small- to medium-sized discrete part$ These low- and medium-carbon steels contain small amounts (0.05 to 0.15%) of alloying elements, such as niobium, vanadium, titanium, molybden rare earth elements, or combinations thereof.% is to provide grain refmnement and/or precipitauo between 500 and 750 MPa (70 w d 110 ksi) can be Weldabllity can be retained or even improved if the carbon c decreased. In essence, these steels offer maximum strength wi simultaneously preserving wekdability, machinabiiity, and f quenched-and-tempered alternative, however, ductility a somewhat mferior. Cold-formed microalloyed steels require less cold w of strengfh, so they tend to have greater residual ductility. forgings, can often be used in the a~r-woled conditio control and controlled-rate coolingdirectlyfrom the erties can he produced that approximate those of quen Mach~nability can be enhanced because of the more unifonn hardness and.the fact th

SECT~ON 6.4 Steel

129

the ferrite-pearlite structure of the microalloyed steel is often more machinable than the ferrite-carbide structure of the quenched-and-tempered variety. Fatigue life and wear resistance can also be superior to those of the heat-treated counterparts. In applications where the properties are adequate,microalloyed steels can often provide attractive cost savings. Energy savings can be suhstantia1,straightening or stress relieving after heat treatment is no longer necessary, and quench cracking is not a problem. Due to the increase in material strength, the size and weight of finished products can often be reduced. As a result, the cost of a finished forging could be reduced by 5 to 25%. If these materials are to attain their optimum properties, certain precautions must be observed. During the elevated-temperature segments of processing, the material must be heated high enough to place all of the alloys into solution. After forming, the products should be rapidly air cooled to 540" to 600C (1000" to 1100F) before dropping into collector boxes. In addition, microalloyed steels tend to through-harden upon air cooling,so products fail to exhibit the lower-strength,higher-toughness interiors that are typical of the quenched-and-tempered materials

BAKE-HARDENABLESTEELSHEET
Bake-hardenable steel has assumed a significant role in automotive sheet applications. These low-carbon steels are processed in such a way that they are resistant to agingduring normal storage but begin to age during sheet metal lorming. A subsequent exposure to heat during the paint-baking operation completes the aging process and adds an additional 35 to 70 MPd (5 to 10 ksi), raising the final yield strength to approximately 275 MPa (40 ksi). Since the increase in strength occurs after the forming operation, the material offers good formability coupled with improved dent resistance in the final product. In addition,it allows weight savings to he achieved without compromising the attractive features of steel sheet, which include spot weldability, good crash energy absorption, low cost, and full recyclability.

ADVANCED HIGH-STRENGTH STEELS (AHSS) Since 2000, there have been significant developments in automotive materials, with large amounts of low-carbon and HSLA steels likely to he replaced by the advanced highstrength steels (AHSS). The AHSS steels are primarily ferrite-phase, soft steels with varying amounts of martensite, bainite, and retained austenite, which offer high strength with enhanced ductility. While previous high-strenglh grades, such as HSLA, suffered from reduced formability,the AHSS materials enable the stamping or hydroforming of more complex parts. Parts can often be integrated into single pieces, eliminating the cost and time associated with assembly, and the higher strength provides improved fatigue and crash performance, along with the possibility of weight reduction. Dual-phase steels form when we quench material from a temperature that is above the A, but below the A,, where the structure consists of ferrite and high-carbon austenite. During the quench, the ferrite remains unaffected, while the high-carbon austenite transforms to high-carbon martensite. A low- or medium-carbon steel now has a mixed microstructure of weak, ductile ferrite combined with high-strength, high-hardness, high-carbon martensite. The dual-phase structure offers strengths that are comparable to HSLA materials, coupled with improved forming characteristics and no loss in weldability.The high work-hardeuiug rates and excellent elongation lead to a high ultimate tensile strength coupled with a low initial yield strength. The high strain rate sensitivity means that the faster the steel is crushed, the more energy it absorbs-a feature that further enhances the crash resistance of automotive structures. While the dual-phase steels have structures of ferrite and martensite,transfurmariuninducedplusticity (TRIP) steels contain ferrite witb a combination of martensite, bainite, and retained austenite. Since the retained austenite transforms progressively to martensite as the steel is deformed, the high work-hardening rates of the TRlP steels persist to high strains,offeringsignificant advantages in operations such as stretch forming and deep drawing. Alternatively,the transformation to martensite and high rate of work hardening can also be used to provide excellent energy absorption during crash deforination.

Tenslle strength (ksl)

50

100

150

200

5 0

Low

2
;

strength

Htgh ! , / strength

Ultra-hlgh strength
k 7 5 0 MPa)

8
.$
m
E

(c270MPa) 40 --

30-

0
0 -

Advanced high-strength

steels

m 20--

P
FIGURE 6-9 Relat~ve ctrengt > d n u lormdoll~ty (eloriyar.on) u1
lo--

+.

conventrona., h~gh-strengrn low-alloy, and advanced high-strength steels. B H. = bake hardenable; DP =dual phase; and Mart = martens~tic.

500

1000

1500

Tens~le strength iMPal Complex-phase (CP) steels and martensitic (Mart) steels offer even higher strengths with useful capaclty for deformation and energy absorpt~on. The CP steels have a microstructure of ferr~te and a higher volume fraction of hard phases (martenslte and bam~te), strengthened further by a f ~ n preclpltate e of niobium, titanium, or vanadium carbides or nilrides.The Mart steels are almost entirely martensite and can have tensile strengths up to 1700 MPa (245 ksi). F~gure 6-9 shows the relative strengths and formabihty (elongation) of the conventional steels (includingmild steels and bake-hardenable steels), HSLA steels,and the newer AHSS materials. The carbon-manganese steels would br~dge the gap between the mild and bake-hardenable and the HSLA steels. Also mcluded ~n thls figure are some useful distinctions between low-strength steels (ultimate tens~le strength below 270 MPa or 40 ks~), h~gh-strengthsteel, and ultra-h~gh-strengthsteels (ultimate tensile strength above 700 MPa or 100 ksi).

FREE-MACHINING STEELS
The increased use of high-speed, automated machinmg has spurred the use and development of several varieties of free-machmmg steels. These steels machme readily and form small chlps when cut.The smaller chips reduce the length of contact between the chip and cutting tool, thereby reducing the assoelated f r ~ c t ~ o and n heat, as well as required power and wear on the cuttlng tool.The formation of small chips also reduces the likelihood of c h p entanglement in the machine and makes chip removal much easier. On the negative side, iree-mach~ning steels often carry a cost premium of I5 to 20% over conventional alloys, but this Increase may be easily recovered through higher machmmg speeds, larger depths of cut, and extended tool life. Free-machining steels are bas~cally carbon steels that have been mod~fied by an addition of sulfur, lead, bismuth, selenium, tellurium, or phosphorus plus sulfur to enhance machinab~lily. Sulfur combines with manganese to form soft manganese sulfide within the structure The inclusions.These, in turn, serve as chip-breaking dlscontiu~ties inclustons also provide a built-in lubricant that prevents formation of a built-up edge on the cutting tool and imparts an improved cutting geometry (see Chapter 21). In leaded materials, the msoluble lead particles work in much the same way. The bismuth free-machining steels are an attractive alternative to the previous vaneties Bismuth is more env~ronmentally acceptable (compared to lead), has a reduced tendency to form stringers, and can be more uniformly dispersed since its density is better match to that of Eon. Mach~nab~lity is improved because the heat generated b cutting is suffic~ent to form a thin f ~ l m of liquid b~smuth that lasts for only fractions of microsecond.Tool life is noticeably extended and the machrned product is still weldable

The use of free-machining steels is not without compromise. however. Ductility and impact properties are somewhat reduced compared to the unmodified steels. Copper-based braze joints tend to ombrittle when used to join bismuth free-machining steels, and the machining additions reduce the strength of shrink-fit assemblies. If these compromises are objectionable, other methods may be used to enhance machinability. For example, the machinability of steels can be improved by cold working the metal. As the strength and hardness of the metal increase, the metal loses ductility, and subsequent machining produces chips that tear away more readily and fracture into smaller segments.

PRECOATED STEEL SHEET Traditional sheet metal fabrication involves the shaping of components from bare steel, followed by the finishing (or coating) of these products on a piece-by-piece basis. In this sequence, it is not uncommon for the finishing processes to be the most expensive and time-consuming stages of manufacture, since it involves handling, manipulation, and possible curing or drying, as well as adherence to the various EPA (environmental) and OSHA (safety and health) requirements. An alternative to this procedure is to purchase mill-coated steel sheet, where the steel supplier applies the coating when the material is still in the form of a long, continuous strip. Cleaning,pretreatment, coating, and curing can all be performed in a continuous manner, producing a coating that is uniform in thickness and offers improved adhesion. Numerous coatings can be specified,including the entire spectrum of dipped and plated metals (including aluminum,zinc, and chromium), vinyls, paints, primers, and other polymers or organics. Many of these coatings are specially formulated to endure the rigors of subsequent forming and bending.Tne continuous sheets can also be printed, striped, or embossed to provide a number of visual effects. Extra caution must be exercised during handling and fabrication to prevent damage to the coating, but the additional effort and expense are often less than the cost of finishing individual pieces. STEELS FOR ELECTRICAL AND MAGNETIC APPLICATIONS Soft magnetic materials cdn be magnetized by relatively low-strength magnetic fields but lose almost all of their magnetism when the applied field is removed.They are widely used in products such as solenoids, transformers, motors, and generators. The most common soft magnetic materials are high-purity iron, low-carbon steels, iron-silicon electrical steels, amorphous ferromagnetic alloys, iron-nickel alloys, and soft ferrites (ceramic material). In recent years, the amorphous me& have shown attractive electrical and magnetic properties. Since the material has no crystal structure, grains, or grain boundaries, (1)the magnetic domains can move freely in response to magnetic fields, (2) the properties are the same in all directions, and (3) corrosion resistance is improved.The high magnetic strength and low hysteresis losses offer the possibility of smaller,lighter-weight magnets. When used to replace silicon steel in power transformer cores, this material has the potential of reducing core losses by as much as 50%. To exhibit permanent magnetism, materials must remain magnetized when removed from the applied field. While most permanent magnets are ceramic materials or complex metal alloys, cobalt alloy steels (containing up to 36% cobalt) may be specified for electrical equipment where high magnetic densities are required. MARAGING STEELS When superhigh strength is required from a steel, the maragi~zg grades become a very attractive option. These alloys contain between 15 and 25% nickel, plus significant amounts of cobalt, molybdenum, and titanium, all added to a very low-carbon steel. They can be hot worked at elevated temperatures, machined, or cold worked in the aircooled condition, and then aged lo yield strengths in excess of 1725 MPa (250 ksi). with good residual elongation. Maraging alloys are very useful in applications where ultra-high strength and good toughness are important.They can be welded, provided the welding is followed by the full solution and aging treatment.Asmight be expectedfrom the large amount of alloy additions

132

CHAIY~ER 6 Ferrous Metals and Alloys (over 30%) and multistep thermal processing,maraging steels are quite expensive and should be specified only when their outstanding properties are absolutely required.

STEELS FOR HIGH-TEMPERATURESERVICE

As a general rule of thumb, plain-carbon steels should not be used at temperatures in excess of about 250C (500F). Conventional alloy steels extend this upper limit to around 350C (650F). Continued developments in areas such as missiles and jet aircraft, however, have increase,d the demand for metals that offer good strength characteristics, corrosion resistance, and creep resistance at operating temperatures in excess of 550C (1000F). The high-temperature ferrous alloys tend to be low-carbon materials with less than 0.1/6 carbon. At their peak operating temperatures, 1000-hour rupture stresses tend to be quite low, often in the neighborhood of 50 MPa (7 ksi).Wile iron is also a major component of other high-temperature alloys, when the amounts fall below 50%, the metal is not generally classified as a ferrous material. High strength at high temperature usually requires the more expensive nonferrous materials that will be discussed in Chapter 7.

11 6.5 STAINLESS STEELS

Low-carbon steel with the addition of 4 to 6% chromium acquires good resistance to many of the corrosive media encountered in the chemical industry.This behavior is attributed to the formation of a strongly adherent iron chromium oxide on the surface.If more improvedcorrosion resistance and outstanding appearance are required, materials should be specified that use a superior oxide that forms when the amount of chromium in solution (excluding chromium carbides and other forms where the chromium is no longer available to react with oxygen) exceeds 12%.When damaged, this tough,adherent, corrosion-resistant oxide (which is onIy 1-2 nanometers thick) actually heals itself, provided oxygen is present, even invery small amounts. Materials that form thissuperior protective oxide are known as the true stainless steels. Several classification schemes have been devised to categorize these alloys. The American Iron and Steel Institute groups the metals by chemistry and assigns a three-digit number that identifies the basic family and the particular alloy within that family.Ln this text, ostructural families,since it is the basic structure that controls the engineering properties o f the metal. Table 6-6 presents the AISI designation scheme for stainless steels and corAlloys relates it with the microstructuralfamilies. Chrornium,nickel, mnnganesc, or nitropa Austenitic Chromium is a ferrite stabilizer; that is, Chromium and nickd the addition of chromium tends lo increase Austcnitic Chromium and possibly carboo Ferrilic Or the temperature range over which ferrite is the Low chromium (112%) end possibly carbon Martensitic stable structure.With sufficient chromium and a low level of carbon, a corrosion-resistant iron alloy can be produced that is ferrite at all temperatures below solidification.These alloys are known as the ferritic stainless sfeels.They possess rather limited ductility and poor toughness but are readily weldable. No martensite can form in the welds because there is no possibility of forming the face-centered-cubic (FCC) austenite structure tha can then transform during cooling.Tnese alloys cannot he heat treated, and poor duo tility limits the amount of strengthening by cold work.The primary source of strength is the body-centered-cubic (BCC) crystal structure combined with the effects of solid solution strengthening. Characteristic of BCCmetals, the ferritic stainless steels exhibit a ductile-to-brittle transition as the temperature is reduced. The ferritic alloys are the cheapest type of stainless steel, however, and, as such, they should be given first consideration when a stainless alloy is required. If increased strength is needed, the mar?ensiticstainless steels should be considere For these alloys, carbon is added and the chromium content is reduced to a level whet the material can be austenite (FCC) at high temperature and ferrite (BCC) at low. The carbon can be dissolved in the face-centered-cubicaustenite, which can then be quenched

Series
200 303
400 500

Ssnro~ 6.5 Stainless Steels

133

to ttIf ri-

m
no %I-

: K
lor

to trap it in a body-centered martensitic structure.The carbon contents can be varied up to 1.2% to provide a range of strengths and hardnesses. Caution should be taken, however, to emure more than 12% chromium remains in solution. Slow cools may allow the carbon and chromium to react and form chromium carbides.When this occurs, thechromium isnotavailable to react with oxygen and form the protective oxida As a result, the martensitic stainless steels may only exhibit good corrosion resistance when in the martensitic condition (when the chronlium i s trapped in atomic solution) and may be susceptible to red rust when annealed or normalized for machining or fabrication.The martensiticstainless steels cost about if times asmuch as the ferritic alloys,with part o C the increase being due to the additional heat treatment, which generally consists of an austenitization.quench, stress relief, and temper.They are Less corrosion resistant than the other varieties and tend to be used when strength and hardness are the dominant requirements. Nickel is an austenite stabilizer;with sufficientamounts of both chromium and nickel (and low carbon),it is possible to produce a stainless steel in which austenite is the stable structure from elevated to cryogenic temperatures. Known as nrrsrenitic stainless rteels, these alloys may cost two to three times as much as the ferritic variety, but here the added expense is attributed to the cost of the nickel and chromium alloys. Manganese and nitrogen are also austenite stabilizers and may be substituted for some of the nickel to produce a lower-cost, somewhat lower-quality austenitic stainlesssteel (the AISI 200-series). Austenitic stainless steels arc casily identified by their nonmagnetic characteristic (the ferritic and martensitic stainlesses are attracted to a magnet).They are highly resistant to corrosion in almost all media (except hydrochloric acid and other halide acids and salts) and may he polished to a mirror finish, thereby combining attractive appearance and corrosion resistance. Formability is outstanding (characteristic of the FCC crystal structure), and these steels strengthen significantly when cold worked.The following table shows the response of the popular 304 alloy (also known as 18-8 because of the composition of 18% chromium and 8% nickel) to a small amount of cold work: Water Quench
Yield strength [MPa (ksi)] Tensile suength [MPa (ksil] Elangation in 2 in. (%)

Cold Rolled 15%

80s (117) 965 (140) II

260 (38) 620 (90)

68

is, tse the @d mt pse md use hat

Fgth

Po-

it a the pn-

The austeniticstainless steels offer the best combination of corrosion resistance and toughness of the stainless varieties.Since they are also some of the most costly, they should not be specified where the less expensive ferritic or martensitic alloys would be adequate or where a true stainless steel is not required. Figure 6-10 lists some of the major structural classifications and schematically denotes some of their key properties. Table 6-7 shows the basic types and the primary mechanism of strengthening. A fourth and special class of stainless steels is the precipitation-harderzing variety. These alloys are basically martensitic or austenitic types, modified by the addition of alloying elements such as aluminum that permit the precipitation of hard intermetallic compounds at the temperatures used to temper martensite. With the addition of age hardening, these materials are capable of attaining high-strength properties such as a 1790-MPa (260-ksi) yield strength, 1825-MPa (265-ksi) tensile strength, and a 2% elongation. Since the additional alloys and extra processing make the precipitation-hardeningalloys some of the most expensive stainless steels, they should be used only when their high-strength feature is absolutely required. While the four structures described above constitute the bulk of stainless steels, there are also some additional variants. Duplex stainless steels contain between 18 and 25% chromium, 4 to 7% nickel, and up to 4% molybdenum; they can be water quenched

AlSl tvpe Martenstic (hardenableby heat treatment)

Ferr~tic

Usage

General purpose Hardenable by heat treatment

405 430 446 z

(morecorrosbon resistant than martensltlc, but not hardenable by heat treatment) Austen~t~c (best corrosion resistance, but hardenable only by cold work~ng)
FIGURE 6-10

Hardenable by cold work~ng

202
301 302 3028 -":l304L 310 316

For elevated-temperature
S~NICB

_ I/

Modltied for welding Superlor corrosion

resistance

Popular alloys and key properties for different types of stainless steels.

321

from a hot-working temperature to produce a microstructure that is approximately ha1 ferrite and half austemte.This mixed structure offers a higher yield strength and greate resistance to stress corrosion cracking and pitting corrosion than e~ther the full-austen~ti or full-ferritlc grades. S~nce stainless steels are dlfficult to machlne because of their work-hardem properhes and their tendency to seize dur~ng cutting, specialfree-machining alloys hav been produced within each family. Additions of sulfur or selenium can raise mach~nah~h to approximately that of a medmm-carbon steel. Cast stamless steels have structures and properties that are s~mllar to the wrough grades but are specified by the deagnations of the Alloy Castlug 1nstitute.The Csene used primarily to impart corrosion resntance, are used in valves, pumps, and fittings The H grades (heat-resistant), des~gned to provide useful properties at elevated tern perature, are used for furnace parts and turblne components. Several potential problems are umque to the family of stainless steels. Since th protective oxide provides the excellent corrosion reslstance, t h ~ s feature can be lo whenever the amount of chrom~um in solution drops below 12%.A localized depleti of chromium can o c m when elevated temperatures allow chrom~um carbides to fo along grain boundaries (sensitrzat~on).To prevent then formation, one can keep the c bon content of stainless steels as low as possible,usually below 0 10% Anrjther met is to tie up existing carbon with small amounts of stabiliz~ng elements,such as t ~ t a n ~ or n~obmm, that have a stronger affimty for carbon than does chromium. Rapidly c ing these metals through the carbide-forming range of 480" to 820C (900" to 150 also works to prevent carbide formation. Another problem w ~ t h high-chromium stamless steels is an embrlttlement can occur after long times at elevated temperatures.Th~s e attributed to the forma of a brittle compound that forms at elevated temperature and coats gram bounda Known as slgmaphase, this material then prov~des a bnttle crack path through the Stainless steels used m h~gh-temperature semce should be checked periodically to and monitor s~gma-phase formation.

Tool steels are htgh-carbon, high-strength, ferrous alloys that have been modified alloy additions to provide a desired balance of shength, toughness and wear resistan when properly heat treated. Several classrfication systems have been developed, so using chemwtry as a basis and others employmg hardening method or major mecham property. The AISI system uses a letter designation to identify basic features such quenching method, pr~mary appl~cat~on, special alloy or charactenst~c, or speclfic mdus

SECT~ 6.6 ~ N Tool Steels

"'

$%&I ,&4 ;?ct&~.owe.s~ondinq .<.,:.. , . . AlSl Grades . . .

:

3. ' 7 " ' .

Type
1 Wate~-H&rdenlng

AISI Grade W
0 A
D S

S~gniftcant Charactenstlo
Oil-hardcnmn~ Alr-hardmmg medrum alloy
High-carbon-I11gt~-~11rurn1m

2 Cold-work

3 Shock-resistmg 4 H~gh-speed
5 Hot work

T M
H

Tungslen alloy Molybdenum alloy HI-H1Y chromlum alloy HZO-H3Ymngsten alloy H4kH59 molybdenum *day
Low alloy

6. Plashc-mdd
7 Special-purpose

L
F

Cnrbon-tungsten

involved. Table 6-8 listr seven basic fam~lies of tool steels, the corresponding AISI letter grades, and the associatedfeature or characreristie. Individual alloys within the l a ter grades are then Iisted numerically to produce a letter-number identification system. W&r-hardening toolafeels (W grade) are essentially hgh-carbon plain-carbon steels. They are the least expensive variety and are used for a wide range of parts that are usually quite small and not subject to severe usage or elevated temperature. Becausestrength and hardness art- funclions of the carbon tontent, a wide range of properties can be achieved through composition variation. Hardenability is low, so these steels must be quenched in water to attain high hardne6s.They can be used only for relatively thin sections if the full depth of hardness i$ desiredThey are also rather brittle, part~cularly at higher hardness. %pica1 uses of the varlous plam-carbon steels are as follows:
0~61M.7$V/o carbon: machine parts, chisels, setscrews, and similar products where medium hardnesi is required, coupled with good roughness and shock resistance 0.7S-O.90% curbon. f o r ~ n dies, g hammers, and sledges 0.9&l.lO% curbonlgeneneral-purposetoohng applications that require a good balance of we= resistance and toughness, such as dr~lls. cutters, shear blades, and other heavy-duty cutting edges 1.1&1.30% carbon: s m a l l drills, lathe tools, razor blades, and other light-duty applications in which extreme hardness is rcquired without great toughness

I n applications where improved toughness is required, small amounts of manganese,siIicon, andmolybdenum are often added.Vanadtum additions of about 0.20% are used to form strong, stahIe carhides that retain fine grain size during heat treatment, One of the main weaknesses of the plain-carbon tool steels is their Ioss of hardness at elevated temperature, which can occur with prolonged exposure to temperatures over 150C 1'30VF1. When larger parts must be hardened or distortion must bt- mintmized, the wldwork tool sfeels are usually recommended.The alloy additions and higher hardenabilitp of the oil- or air-hardening grades (Q and A designations, respectively) enable hardening by less severe quenches. Tighter dimens~onal folerances can be maintained during heat treatment, and the cracking tendency is reduced. The hrgh-chromtm toal contain between 10 and 18% chromium, and are also steels, designated by the letter D, air-hardening and offer outstanding deep-hardemng wear resistance.Blanking, stamping, and cold-formingdies,punches, and other taols for large production runs are all common applications for this class. Because these steels do not have the alloy content necessary to resist softening at elevated temperatures, they should not be used for applications that involve prolonged servlce at temperatures in excess of 250C (500F).

136

CHAPTER 6 Ferrous Metala and Alloys

Shock-resistrag tool steels ( S dwgnation) offer the high toughness needed f Impact applications Low carbon content (approximately 0.5% carbon) is usually s ified to assure the necessary toughness, w ~ t h carbide-forming alloys provldmg the essary abrasion resistance, hardenability, and hot-work characteristics Applicati include parts for pneumatic tooling, chisels, punches, and shear blades. Nrgh-speed tool steels are used for cutting tools and other applications whe strength and hardness must be retained at temperatures up to or exceeding red-he (about 760C or 1400F). One popular member of the tungsten h~gh-speed tool ste (T designatlon) is theT1 alloy, which contains 0.7% carbon, 18%tungsten, 4% ch um, and 1%vanadium. It offers a balanced combination of shock resistance and a sion resistance and is used for a wide variety of cuttlng applications.The molybden high-speed steels (M designation) were developed to reduce the amount of tungs and chromium requued to produce the high-speed propelties Hot-work toolsteels ( H designation) were developed to provide strength and hard ness during prolonged exposure to elevated temperature. All employ substantial add tions of carbide-fomng alloys. HI to H i 9 are chromium-based alloys with about 5.0 chromium; H20 to H39 are tungsten-based types with 9 to 18% tungsten coupled wl 3 to 4% chrommm;and H20 to H59 are molybdenum-based. The chromium types te to be less expensive than the tungsten or molybdenum alloys. Other types of tool steels rnclude (1) thepla~tic mold steels ( P designation), signed to meet the requiremenls of zinc dle casting and plastic injection molding dl (2) the low-alloy special-purpose tool steels ( L designatlon), such as the L6 extre tool steels (Fd toughness var1ety;and (3) the carbon-tungsterz type of spec~al-purpose ignation), which are water hardening but substantially more wear-reustant than t plain-carbon tool steels Most tool steels are wrought materials, but some are designed specifically for fa ricat~on by castlng Powder metallurgy processing has also been used to produce spec compositions that are difficult or impossible to produce by wrought or cast methods provide key structural enhancements By subjectingthe water-atomized powders to h isastatlc presslng (HIP), 100%-dense billets can be produced wlth fine grain size a small, uniformly distnbuted carbide particles. These mater~als offer superior wear sistance compared to conventional tool steels, combined with useful levels of toughn

6.7 ALLOY CAST STEELS ANDIRONS

The effects of alloying elements are the same regardIess of the process used to produ the final shape When the desired shape is to be made by castlng, some alloys can used to enhance process-spec~fic features, such as fluid~ty and as-solidified propertl I f a ferrous casting alloy contains less than about 2.0% carbon, it 1s considered to be cmtsfeel Alloys with more than 2% carbon are cast irons Most cast irons are used in the as-cast condition, w~th the only heat treatment be a stress relief or annealing. For these applications, the alloy elements are selected their abllity to alter properties by (1)affecting the formation of graph~te or cemen (2) modifying the morphology of the carbon-rich phase, (3) strengthening the ma matenal, or (4) enhancing wear resistance through the formation of alloy carbldes Nic for examplq promotes graphite formation and tends to promote finer graphite str tures. Chromium retards graphite formation and stabilizes cementite. These alloys frequently used together in a ratio of two or three parts of nickel to one part of ch um. Between O 5 and 1.0% molybdenum is often added to gray cast Iron to lmpar tionat sttength,form alloy carbides, and help to control the srze of the graphite fla Hzgh-alloy cast aruns have been deslgned to provide enhanced corrosion resistan gr andlor good elevated-temperature servlce. Withii this famlly, the austen~tic irons, which contain about 14% nickel,% copper, and 2.5% chromium, offer go rosion resistance to many acids and alkalis at temperatures up to about 800C (150 Alloy cast rrons and cast steels are usually specified by their ASTM designation n hers, which relate the mater~als t s their mechanical properties and intended se applicatlons.The Society of Automotive Engineers also has specifications for cast ste used in the automotive indushy.

Cast steels are generally used whenever a cast iron is not adequate for the application. Compared to cast irons, the cast steels offer enhanced stiffness, toughness, and ductility over a wide range of operating temperatures and can be readily welded.They are usually heat treated t o produce a final quenched-and-tempered structure, and the alloy additions are selected t o provide the desired hardenability and balance of properties.The enhanced properties come with a price, however, since the cast steels have a higher melting point (more energy t o melt and higher cost refractories are necessary), less fluidity (leading t o increased probability of incomplete die o r mold filling), and increased shrinkage (since graphite is not formed during solidification). The diverse applications take advantage of the material's structural strength and its ability to contain pressure, resist impacts, withstand elevated temperatures, and resist wear.

I IKey Words
vanced high-strength steel deoxidation dual-phase steels duplex stainless steel electroslag remelting ferritic stainless steel ferrous metals free-machiningsteel high-carbon steel high-speed tool steel high-strength low-alloy steel hot-work too1 steel ladle metallurgy ladles low-carbon steel maraging steel martensitic stainless steel medium-carbon steel microalloyed steel oil-hardenable tool steel pig iron plain-carbon steel precipitation-hardenable stainless steel precoated steel sensitization shock-resistingtool steel sigma phase solidification shrinkage stainless steel steel tool steel TRlP steels (transformationinduced plasticity) vacuum arc remelting vaauum degassing vacuum induction melting water-hardenable tool steel

e-hardenable steel

III Review Questions

Why might it he important to know the prior processing history of an engineering material? What is a ferrous material? How does the recycling of steel compare with the recyclingof What properties or characleristics have made steel such an attractive engineering material? When iron ore is reduced to metalliciron, what other elements are generally present in the metal? Haw does steel differ lrom pig iron? What are some of the modification processes that can he performed on a steel during ladle metallurgy operations? What is the advantage of pouring molten metal from the bottom of a ladle? What are some of the attractive economic and processing )advantagesof continuous casting?
19. Whal are some of the alloy elements that tend to form stable carhides within a steel? 20. What alloys are particularly effective in increasing the hardenability of steel? 21. Whal is the significance of the last two digits in a typical four&git AISISAE steel designation? 22. How are letters incorporated into the AISI-SAE designation system for steel, and what do some of the more common ones mean? 23. What isanH-grad* stee1,and when should it be considered in a material specification? 24. Why should the proposed fabrication processes enter into consideration when selecting a steel? 25. How are the final properties usually obtained in the constructional alloy steels? In the HSLA steels? 26. How are HSLA steels specified? 27. What are microalloyed steels? 7.8. What are some of the potential benefits that may be obtained through the use of microalloyed steels? 29. What is the primary attraction of thebake-hardenahle steels? 30. What are advanced high-strength steels (AHSS)? 31. What are the two phases that are present in dual-phase steels? 32. What is the"transformation" that occurs during the deformation of the "transformation-induced plasticity (TRIP) steels? 33. What are some of the various alloy additions that have been used to improve the machinability of steels? 34. What are some of the compromises associated with theuse of free-machiningsteels? 35. What factors might be used to justify the added expense of precoated steel sheet? 36. Why have the amorphous metals attracted attention as potenrial materials for magnetic applications?

1 are some of the attractive features of electroslag

at are some of the common alloy elements added to steel? what different reasons might alloyingelementsbe added

138

CHAFIER 6 Ferrous Metals and Alloys

45. How can an austenitic stainless steel be easily identified? 46. What two structures are present in a duplex stainless steel? 47. What i s sensitization of a stainless steel, and how can i t be prevented? 48. What i s a tool steel? 49. How does the AISI-SAE designation system for tool steels differ from that for plain-carbon and alloy steels?, 50. For whai types of applications might an air-hardenable to01 steel be attractive? 51. What alloying elements are used to produce the hot-worked tool steels? 52. What are some of the reasons that alloy additions aremade to cast irons that will be used i n their as-cast condition? 53. When should a cast steel he used instead of a cast iron?

37. What are maragingsteel$ and for what conditions might they be required? 38. What are the typical elevated temperature limits of p l a i n carbon and alloy steels? 39. What is a stainless steel? 40. What feature is responsiblefor the observed corrosion resistance of stainless steels'? 41. Why should ferritic stainless steels be given first consideration when selecting a stainless steel? 42. Which of the major types of stainless steel is likely to contain significant amounts of carbon?Why? 43. Under what conditions might a martensitic stainless sleel "rust" when exposed to a hostile environment? 44. What are some of the unique properties of austenitic stainless steels?

he interior tub of a washing machine is the container T t h a t holds the clothes during the washing and rinsing cycles, but it also contains the perforations that permit removal of the water by draining and spinning.The CornPonent will see mechanical loadings from the weight of the clothes and water, and also the dynamic action of spinning water-laden fabrics.There will be exposure to a wide range of water quality, as well as the full spectrum of soaps, detergents, bleaches, and other laundry additives.The surfaces should also be resistant to the impact and abrasion of buttons, zippers, and snaps. This part has traditionally been manufactured by thesequentlal deep drawing, perforating,and trimming of metal sheet, followed by some form of surface-coating treatment. For a long time, the standard material was "enameling irons-a steel sheet with less than 0.03% carbon that was then coated with a fired porcelain enamel. Due to the difficulties of producing ultra-low-carbon material in today's steelmaking operations, enameling iron became increasingly scarce, and manufacturers were forced to substitute the lowest-carbon, most readily available material, namely 1008 steei.This substitutionfurther required modification of the enameling process to prevent defects and blistering from GO evolution. Your employer is presently manufacturing these tubs from 1DO8 steel sheet with a subsequent coating of fired

and then determine what might be involved in converting to stainless steel. Consider the following specific questions: 1. What are the obvious pros and cons of the present product and process? Where would you expect most problems to occur in the current manufacturing process?Which aspects of fabrication are likely to be the most costly? 2, What would be the pros and cons of converting to stainless steel? In what ways would the product be superior? Are there any assets or liabilities associated with product fabrication from stainless material? 3. Which stainless steel would you recommend? Begin by considering the basic types (femitic, austenltic, and martensitic) and then refine your selection to a specific alloy if possible. Discuss the rationale for your selection. 4. Since deep drawing is a metal deformation process, we could use cold working (strain hardening) as a strengthening mechanism.Would you find this to be attractive, or would you prefer to use a recrystallizatlon anneal after drawing and prior to use? Why? If you elect to use cold work, might you want to at least perform a stress-relief heat treatment prior to use? could this be done and still preserve the deformation strengthening? In deep drawing, the deformation is not uniform (increasing as we move up the sidewalls of the mntainer), and the bottom of the tub simply retains the properties Of the starting sheet. In order to ensure a minimum amount of strength at ail locations, it may be desirable to begin the

material, the forming pr

'NONFERROUS METALS AND ALLOYS

neral Properties and Characteristics

.Vonferrolrs metals and nlloys have assumed increasingly important roles in modem technology. Because of their number and the fact that their properties vary widely, they provide an almost limitless range of properties for the design enginrer. While they tend to be more costly than iron or steel, these metals often possess certain properties or combinations of properties that are not available in the ferrous metals, such as:

,
, ,

1. 2. 3. 4. 5. 6.

Resistance to corrosion Ease of fabrication High electrical and thermal conductivity Light weight Strength at elevated temperatures Color

requires about 65 pounds of aluminum, 21 pounds of copper, 12 pounds of lead, 11 pounds of zinc, and 25 pounds of various other nonferrous metals. Figure 7-1 classifies some of the nonferrous metals by advantageous engineering properties, and Table 7-1 shows the increasing role of the nonferrous metals in a typical family vehicle. As a whole, the strength of the nonferrous alloys is generally inferior to that of steel. Also, the modulus of elasticity is usually lower, a fact that places them at a distinct disadvantage when stiffnessis a required characteristic.Ease of fabrication is often attractive.Those alloys with low melting points are easy to cast in sand rnolds,permanent molds, or dies. Many alloys have high ductility coupled with low yield points, the ideal combination for cold working. Good machinability is also characteristic of many nonferrous

,
,

,
i
.

139
-

140

CHAPTER 7 Nonferrous Metals and Alloys

FIGURE 7-1

Some common nonferrous metals and alloys, class~f~ed by attractive englneerlng property.

alloys.The savings obtained through ease of fabrlcat~on can often overcome the higher cost of the nonferrous material and justify its use in place of steel. Weldabillty is the one to steel. W~th fabrication area where the nonferrous alloys tend to be somewhat ~nferior modern jointng techmques, however.it is generally posslble to produce satisfactory weldments In all of the nonferrous metals

AND COPPER ALLOYS IE4 7.2 COPPER

GENERAL PROPERTIES AND CHARACTERISTICS Copper has been anlmportant engineering metal for over 6000 years. As a pure metal

(30 ksl), wlth an elongation of nearly 60%.Through cold working, the tensile streng

Copper and copper alloys are used for a variety of electrical appi~catlons, such as these electrtcal
FIGURE 7-2

strength-to-weight ratio for copper alloys is usually less than that for the weak

East, PA )

While offering good resistance to adhesive wear, copper and copper alloys have po abrasive wear characteristics. attractive ductility even under cryogenic coudlhons Conductiv~ty also tends to increa with a drop in temperature.

SECTION 7.2

Copper and Copper Alloys

141

nonmagnetic, nonpyrophoric (slivers or particles do not bum in air -i.e.,nonsparking), and nonbiofouling (inhibits marine organism growth), as well as offering a wide spectrum of colors,including yeUow, red, brown, and silver.

COMMERCIALLY PURE COPPER Refined copper containing between 0.02 and 0.05% oxygen is called electrolytic tough-pitch (ETP) copper. It is often used as a base for copper alloys and may be used for electrical applications, such as wire and cable, when the highest conductivity is not required. For superior conductivity, additional refining can reduce the oxygen content and produce oxygen-,free h i g h - c o n d u c t i v i t y (OFHC) copper. The better grades of conductor copper now have a conductivity rating of about 102% IACS, reflecting metallurgical improvements made since 1913, when the International Annealed Copper Standard (IACS) was established and the conductivity of pure copper was set at 100% IACS. COPPER-BASED ALLOYS As a pure metal, copper is not used extensively in manufactured products, except in electrical applications, and even here alloy additions of silver, arsenic, cadmium, and zirconium are used to enhance various properties without significantly impairing conductivity. More often, copper is the base metal for an alloy, where it imparts its good ductility, corrosion resistance, and electrical and thermal conductivity. A Eull spectrum of mechanical properties is available,ranging from pure copper, which is soft and ductile, through alloys whose properties can rival those of quenched-and-tempered steel. Copper-based alloys are commonly designated using a system of numbers standardized by the Copper Development Association (CDA). Table 7-2 presents a breakdown of this system, which has been further adopted by the American Society for Testing and Materials (ASTM), the Society of Automotive Engineers (SAE), and the U.S. government.Alloys numberedfrom 100 to 199 are mostly copper with less than 2% alloy addition. Numbers 200 to 799 are w r o u g h t 1 alloys, and the 800 and 900 series are casting alloys. When converted to the Unified Numbering System for metals and alloys, the three-digit numbers are converted to five digits by placing two zeros at the end, and the letter Cis used as a prefix to denote the copper base. COPPER-ZINC ALLOYS Zinc is by far the most popular alloying addition, and the resulting alloys are generally known as some form of brass. If the zinc content is less than 36%, the brass is a singlephase solid solution. Since this structure is identified as the alpha phase, these aUoys are

861-868 872-879

Manganeseand Leadcd manganese bronzes Silicon bronzes and silicon brasses

529 42 461

Copper-tin allays (phosphor bronzes) Copper-aluminum alloys (aluminum bronzes) Copper-silicon alloys (silicon bronzes)

W2-917 922-929 932-945 947-949 952-958 962-966

l l n bronzes
Leaded tin bran,.es High-leaded lin bronzes Nickel-tin bronzes Aiurniounr bronzes Copper nickels

'The term wroaghr means ''shaped or fabricated in the solid state." Key properties for wrought material. generally rclate to ducu!ity. Cast alloys are shaped as a liquid, where the attractive features includelow melting point, high fluidity,ilnd good as-solidified strength.

Nonferrous Metals and Alloys often called alpha brmses.They are quiteductile and formable, with both strength and ductility Increasing with the zinc content up to about 36%. The alpha brasses can be strengthened significantly by cold workmg and are commercially ava~lable in various degrees of cold-worked strength and hardness Cartridge brass, the 70% copper-30% zinc alloy, offers the best overall combmation of strength and duchhty. As its name implies, it has become a popular material for sheet-forming operahons like deep drawing. With more than 36% zinc. the copper-zinc alloys enter a two-phase region involvmg a br~ttle, nnc-rich phase, and ductility drops markedly. Whlle cold-worktng properties are rather poor for these high-zinc brasses, deformallon can be performed easdy at elevated temperature. Many applications of these alloys result from the high electrical and thermal conductivity coupled wlth useful engineermg strength. The w ~ d range e of colors (red, orange, yellow, silver, and white), enhanced by further variations that can be produced through the addition of a third alloy element, account for a number of decorative uses. Since the plating characleristics are excellent, the material is also a frequently used base for decoratlve chrome or simlarcoat~ngs.Another attractive property of alpha brass is its abllity to have rubber vulcanized to it without any special treatment except thorough c1eaning.A~ a result, brass is widely used m mechan~cal rubber goods. Most brasses have good corrosion resistance. In the range of 0 to 40% zinc,the addition of a small amount of tin imparts improved resistance to seawater corrosion Cartridge brass with tin becomes admiralty brass, and the 40% zinc Muntz metal with a tin addit~on is called naval brass. Brasses with 20 to 36% zinc, however, are subject to a selective corrosion, known as dezinczficarron.when exposed to acidic or salt solutions Brasses with more than 15% zmc often experience season crack~ng or stress-corroszon crackzng. Both stress and exposure to GOrroslve media are required for this failure to occur (but residual stresses and atmospheric moisture may be sufficient!).As a result, cold-worked brass is usually stress rel~eved (to remove the residual stresses) befor being placed in service. When high machinability is required, as with automatic screw-machine stock,2 t 3% leadcanbe added to the brass to ensure the formation of free-breaking chips. Bras casting alloys are qurte popular for use in plumbmg fixtures and fittmgs, low-pressur valves, and a variety of decorahve hardware. They have good fluidity during pourin and attractive low melting points An alloy containing between 50 and 55% copper a the remainder zinc is often used as a filler metal in brazmg. It is an effective material f joming steel, cast iron, brasses, and copper, producing joints that are nearly as strong those obtained by weldmng. Table 7-3 lists some of the more common copper-zinc alloys and their compositio properties, and typical uses.

COPPER-TIN ALLOYS Since tin 1s more costly than zinc, alloys of copper and tin, commonly called tin b are usually specified when they offer some form of special property or characteris term bronze is often confusmg, however, s~nce it can be used to designate any copp alloy where the major alloy add~tion is not zinc or nickel.To provide clarification, th major alloy addifion is usually included In the designallon name. The tln bronzes usually contaln less than 12%tin. (Strength continues to increas as tin is added up to about 20%, but the hlgh-tin alloys tend to be brittle.) Tin br offer good strength, toughness,wear resistance, and corrosion resistance.They are used for bearings, gear$ and fittings that are subjected to heavy compressive load When the copper-tin alloys are used for beanng applications, up to 10% lead frequently added. The most popular wrought alloy is phosphor bronze, which usually contains fr 1 to 11%tin.Alloy 521 (CDA),with 8% tin,is typical of this class.Hard sheet has a t sile strength of 760 MF'a (110 ksi) and an elongation of 3%. Soft sheet has a tens strength of 380 MPa (55 ksi) and 65% elongation. The material is often specified f pump parts, gears, springs, and bearings.

5 ,* ~ : , s: r r .l.rs*:"~:3*:\5~ y , ~ ; s . r * + s i l i lr ,
a

sr~:.ltzy.;%%?\r,:"l.
Elongatran m
2111. (%)
45 4
3 47

a;--',

Number

CommonName

Cu

Zn

Sn

Pb

Mn

Cond~tion
Soh sheet Hard sheet

ksi
38

MPa
262 441

Typlcal Urns

Screen wre,
jewelry

64
73
47

kerdware,snews.
archctectural work,

Low brass

80

20

Spring

Annealed sheet
Hard
spring Annealed sheet

503
324

Drawing,

ornamental

75

91
53

517 627

7
3

MwMns,

365
524
634

54

Had SP~Q
Yellow brass
65 35

76
92

7
3 64

hesledsheer
Hard

46

317
524

mstruments, hlhmg Cold Cnmg, rad~atar cores, spnngs, screws

hardware,mus~cal

M u n e metal

60 71

40
28
35.3
39

433-445

Adm~ralty metal

Hot-rolled Cold rolled Soft Hard

76 54 80
45 95

372
S5l 310 655

45 5
60 5

Arclutectural cork,
condenser tube Condens~r tube

(salt water), heat exchangers Screw-machine parts

Clutch disks pump rods,valve
atetems, hl&-

Free-cuttmgbrass

61.5
58 5

Soft 1

47

Has4 Soft Bars half hard

62
65
84

324 427 448 579

60
20
33 19

strength propellers Alloy 905 is a bronze castlng alloy containing 10% tm and 2% zinc. In the as-cast condition, the tensile strength is about 310 MPa (45 ksi), with an elongation of 45%. It has very good resistance to seawater corrosion and is used on ships for pipe fittings, gears, pump parts, bushings,and bearings. Bronzes can also be made by mixing powders of copper and tin,folIowed by lowdensiry powder metallurgy processing (described m Chapter 19).The porous product can be used as a filter for high-temperature or corrosive media, or it can be infiltrated with od to produce self-lubricating bearings.

COPPER-NICKEL ALLOYS Copper and nickel exhib~t complete solub~lity (as shown previously in Figure 4-61, and a wide range of useful alloys have been developed. Key features include high thermal conductivity, high-temperature strength,and corrosion resistance to a range of marerials, including seawater. These properties, coupled with a high resistance to stresscorroslou cracking, make the copper-nickel all~ys a good choice for heat exchangers, cookware, desalination apparatus, and a wide variety of coinage. Cuprorrickels contain 2 to 30% nickel. Ncckel silvers conrain no silver, but 10 to 30% nickel and at least 5% zinc"Ihe bright silvery luster makes them attractive for ornamental applications, and they are also used for musical instruments An alloy with 45% nickel is known as consurntan, and the 67%-nickelmaterial is called Monel. Mane1 w~Il be discussed later in the chapter as a nickel alloy OTHER COPPER-BASED ALLOYS The copper alloys d'1scussed previously acquire their strength primarily through 80lidsolution strengthening and cold work.With'm the copper-alloy family, alloys containing aluminum, silicon, or beryllium can be strengthened by precipitation hardening.

Almmum-bronze alloys are best known for their combination of bgh strength and excellent corroslon resistance, and they are often considered to be cost-effective alternatives to stamless steel and nickel-based alloys.The wrought alloys can be strengthened by solid-solut~on strengthening, cold work, and the precip~tationof iron- or' nickel-rich phases. With less than 8% aluminum, the alloys are very ductile. When aluminum exceeds 9%, however, the ductility drops and the hardness approaches that of steel. Still higher aluminum contents result in brittle, but wear-resistant, materials. By varying the aluminum content and heat treatment, the tensile strength can range from about 415 to 1000 MPa (60 to 145 ksi). Typical applications include marine hardware, power shafts, sleeve bearmgs, and pump and valve components for handling seawater,' sour mme water, and various industrial flulds. Cast alloys are available for applications where casting is the preferred means of manufacture. Slnce aluminum bronze exhib~ts' large amounts of sohdidification shrinkage, castings made of this material should be designed with this in mind. S~ltcon-bronzescontain up to 4% slhcon and 1.5% zinc [higher zinc contents may be used when the material is to be cast). Strength, formability, machmability, and corroslon resistance are all qmte good. Tensile strengths range from a soft condition of about 380 MPa (55 k s ~ through ) a maximum that approaches 900 MPa (130 ksi). Uses Include bailer, tank, and stove apphcations, which require a eombinatlon of weldabh. ty, high strength, and corroslon resistance. Copper-beryllcum alloys, which ordmanly contain less than2% berylhum, can b e age hardened to produce the highest strengths of the copper-based metals but are quite expenslve to use When annealed, the material has a yield strength of 170 MPa (25 ksi): tensile strength of 480 MPa (70 ksi), and an elongation of 50% After heat treatment these properties can rue to 1100 MPa (160 ksi), 1250 MPa (180 km), and 5%, respective Cold work coupled with age hatden~ng can produce even stronger material The modul of elasticity is about 125,000 MPa (8 X 10' psl), and the endurance limit is aroun 275 MPa (40 ksi).These properties make the material an excellent choice for electri contact springs, but cost Irrnxts apphcation to small components requiring long life a high reliabilrty Other applications, such asspark-resistant safety tools and spot-we1 electrodes,utihze the unique combmation of properbes. (1) the material has the of heat-treated steel, but IS also (2) nonsparking,nonmagnetrc,and electricallya mally conductive. Concerns over the toxicity of beryllium have created a demand f subsbtute alloys with simllar properties, but no clear alternative has emerged. LEAD-FREE CASTING ALLOYS For many years,lead has been a common alloy additive to cast copper alloys. It to fill and seal the rmcroporosity that forms dur~ng solidification, thereby provid pressure tightness required for use w ~ t h pressurized gases and fluids.The lead also act as a lubricant and chip-breaker, enhancing the machinability and machined surface ish. Many plumbing components have been made from leaded red and serm-red b casting alloys. With Increased concern about lead in dnnk~ng water and the introduct~on environmental regulations, efforts were made to develop lead-free copper-based ca ing alloys.Among the most common are the EnviroBrass alloys, which use bi~muth selenium as substitutes for lead. Blsmuth 1s not known to be toxic for humans and been used in a popular remedy for an upset stomach Selenlum 1s an essential nutr e for humans. While somewhat lower m duct~lity, the new alloys have been shown to ha mechanical properties, machinability and platability that are quite sim~lar to the tra tional leaded materials.

ALUMINUM ALLOYS
GENERAL PROPERTIES AND CHARACTERISTICS Although alunzinum has only been a commercial metal for about 120 year$ ~tnow ran second to steelin both worldwde quantlty and expenditure, andit is clearly the most portant of the nonferrous metals. It has achieved importance in virtually all segmen the economy, with principal uses m transportation, containers and packaging, buil

SECTIO~V 7.3 Alumlnum and Alurn~num Alloys

145

construction, electrical applications, consumer durables, and mechanical equipment. We are all familiar with uses such as aluminum cookware, window frames, aluminum siding, and the ever-present aluminum beverage can. A number of unique and attractive properties account for the engineering significance of aluminum. These include its workability, light weight, corrosion resistance, good electrical and thermal conductivity, optical reflectivity. and a nearly limitless array of available finishes.Aluminum has a specific gravity of 2.7 compared to 7.85 for steel, making aluminum about one-third the weight of steel for an equivalent volume. Cost comparisons are often made on the basis of cost per pound, where aluminum is at a distinct disadvantage, being four to five times more expensive than carbon steel.There are a number of applications, however, where a more appropriate comparison would be based on cost per unit volume. A pound of aluminum produces three times as many same-sue parts as a pound of steel, so the cost difference becomes markedly less. Aluminum can be recycled repeatedly with no loss in quality, and recychg saves 95% of the energy required to produce aluminum from ore. Since the 1980s,the overall reclamation rate for aluminum has been over 50%. The aluminum can is the most recycled beverage container in North America, and ovcr 85% of all aluminum used in cars is recovered at the end of their useful life. A serious weakness of aluminum from an engineering viewpoint is its relatively low modulus of elasticity,which is also about one-third that of steel. Under identical loadings, an aluminum component will deflect three times as much as a steel component of the same design. Since the modulus of elasticity cannot be significantly altered by alloying or heat treatment, it is usually necessary to provide stiffness and buckling resistance through design features such as ribs or corrugations.These can be incorporated with relative ease, however, because aluminum adapts easily to the full spectrum of fabrication processes.

COMMERCIALLY PURE ALUMINUM In its pure state, aluminum is soft, ductile, and not very strong. In the annealed condition, pure aluminum has only about one-fifth the strength of hot-rolled structural steel. Commercially pure aluminum, therefore,is used primarily for its physical rather than its mechanical properties. Electrical-conductor-grade aluminum is uscd in large quantities and has replaced copper in many applications, such as electrical transmission lines. Commonly designated by the letters EC, this grade contains a minimum of 99.45% aluminum and has an electrical conductivity that is 62% that of copper for the same-size wire and 200% that of copper on an equal-weight basis. ALUMINUMS FOR MECHANICAL APPLICATIONS For nonelechical applications, most aluminum is used in the form of alloys These have much greater strength than pure aluminum yet retain the advantages of light weight, good conductivity, and corrosion resistance.While usually weaker than steel, some alloys are now available that have tensile properties (except for ductility) that are comparable to those of the high-strength low-alloy (HSLA) structural grades. Since alloys can be as much as 30 times stronger than pure aluminum, designers can frequently optimize their design and then tailor the material to their specific requirements Some alloys are specifically designed for casting, while others are intended for the manufacture of wrought products. On a strength-to-weight basis, most of the aluminum alloys are superior to steel and other structural metals, but wear, creep, and fatigue properties are generally rather poor. Aluminum alloys have a finite fatigue l i e at all reasonable values of applied stress. In addition, aluminum alloys rapidly lose their strength and dimensions change by creep when temperature is increased. As a result, most aluminum alloys should not be considered for applications involving service temperatures much above 150"C(300QF). At subzero temperatures, however, aluminum is actually stronger than at room temperature with no loss in ductility. Both the adhesive and the abrasive varieties of wear can be extremely damaging to aluminum alloys. The selection of steel or aluminum for any given component is often a matter of cost, but considerations of light weight. corrosion resistance, low maintenance expense,

and h ~ g h thermal or electrical conduchv~ty may be suff~c~ent to lustlfy the addcd cosl alummum. W ~ t h the drive for lighter,more fuel-efficient vehicle$ the fraction of alumin targeted for transportatlon applicahons rose from 19 4% ~n1992 lo 31.8% in 2002.The 1 of aluminum doubled in cars and trlpled in sport-ut111ty vehicles (SUVs) and lrght truc Alum~num is being used in body panels, engine blocks,maniCold&lransmsion housir and whee1s.h alummum space frame, such as the ones ~ h o w in n F~gure 7-3 for the 2( Ford GT and the 2006 Corvette 206, can reduce the overall welght of the structure, enha~ recyclablhty, and reduce the number ot parts requ~red for the prmary body structure.1 all-alummum space frame of the 2006 206 Corvette resulted in a 30% rcductlon m wei, from the all-steel des~gn of the previous model. In 2001, aluminum passed plastics a percentage ofau(omol~ve mater~al content and 1s now second only to steel and iron.1 average North Amencan automobde now contains over 125 kg (280 pounds) of alurnim

FIGURE 7-3 (a) The space frame chassis for the 2005 Ford GT is comprised of 35 aluminum extrusions, 7 complex castings, 2 semisolid castings, and various aluminum panels, some superplastically formed. (b) The aluminum frame of the 2006 Corvette 206 yielded a 30% weight savings compared to the previous steel design. [(o) Courtesy Ford Motor Cornpony, Deorborn, MI; ond HydroAluminurn of North Americo, Linthicurn, MD). (b) (Courtesy of General Motors, Detroit, MI.)]

CORROSION RESISTANCE OF ALUMINUM AND ITS ALLOYS Pure aluminum is very reactive and forms a tight, adherent oxide coating on the surface as soon as it is exposed to air.This oxide is resistant to many corrosive media and serves as a corrosion-resistant barrier to protect the underlying metal. Like stainless steels, the corrosion resistance of aluminum is actually a property of the oxide, not the metal itself. Since the oxide formation is somewhat retarded when alloys are added, aluminum alloys do not have quite the corrosion resistance of pure aluminum. The oxide coating also causes difficulty when welding. To produce consistentquality resistance welds, it is usually necessary to remove the tenacious oxide immediately before welding. For fusion welding, special fluxes or protective inert gas atmospheres must be used to prevent material oxidation. While welding aluminum may be more difficult than steel, suitable techniques have been developed to permit the production of high-quality, cost-effective welds with most of the welding processes. CLASSIFICATION SYSTEM Aluminum alloys can be divided into two major groups based on the method of fabrication. Wrought alloys are those that are shaped as solids and are therefore designed to have attractive forming characteristics, such as low yield strength, high ductility, good fracture resistance, and good strain hardening. Casting alloys achieve their shape as they solidify in molds or dies. Attractive features for the casting alloys include low melting point, high fluidity, and attractive as-solidified structures and properties. Clearly, these properties are distinctly different, and the alloys that have been designed to meet them are also different. As a result, separate classification systems exist for the wrought and cast aluminum alloys. WROUGHT ALUMINUM ALLOYS The wrought aluminum alloys are generally identified using the standard four-digit designation system for aluminums.The first digit indicates the major alloy element or elements as described below:
Element Major Alloy~ng
Alumnlum, 99 00% and greater Coppcr Manganese S,l,con Magnes~um Magnesmrn and a~ilcon Zlnc Other element

lxxx 2xnx 3xxx 4xxx

Sxxx

6xxx
lxxx
Uxxx

The second digit is usually zero. Nonzero numbers are used to indicate some form of modification or improvement to the original alloy.The last two digits simply indicate the particular alloy within the family. For example, 2024 simply means alloy number 24 within the 2xxx, or aluminum-copper, system. For the lxxx series, the last three digits are used to denote the purity of the aluminum. The four digits of a wrought aluminum designation identify the chemistry of the alloy Additional information about the alloy condition is then provided through a temper designation, in the form of a letter or letter-number suffix using the tollowing system:

-F: as fabricated

-H: strain-hardened
-HI: strain-hardened by working to desired dimensions;a second digit, 1througt~ 9, indicates the degree of hardening, 8 being commercially full-hard and 9 extra-hard -HZ: strain-hardened by cold working, followed by partial annealing -H3: strain-hardened and stabilized

us Metals and Alloys

-0:annealed

-T thermally treated (heat treated)

-TI: cooled from hot working and naturally aged -l2cooled from hot working, cold-worked, and naturally aged -T3: soluhon-heat-treated, cold-worked, and naturdlly aged -T4. solution-heat-treated and naturally aged -T5: cooled from hot working and artificially aged -T6 solut~on-heat-treated and art~ficially aged -T7: solut~on-heat-treared and stab~lized -TE: solut~on-heat-treated, cold-worked, and artifidally aged -T9: solution-heat-treated, artificially aged, and cold-worked -TI@cooled Irom hot working, cold-worked, and artificially aged
-W solution-heat-treated only

The various wrought dlloys are often divided into two basic types: those achieve strength by sol~d-solutionstrengthening and cold workmg, and those that be strengthened by heat treatment (age hardenlng).Table 7-4 lists some of the eomm wrought alumnum alloys m each family. 11 can be noted that the work-hardenable loys (thosc that cannot be age hardened) are primarily those in the lxxx (pure a minum), 3xxx (aluminum- manganese), and 5xxx (aluminum-magnesium) series. comparison of the annealed (0 suffix) and cold-worked (H suffuc) conditions reveals t amount olstrengthening achievable through strain hardening. The precipitat~on-hardenablealloys are found pr~marily in the 2xxx, 6xxx, 7xxx series. By comparing the properties m the heat-treated con&tion to those of stran-hardened alloys, we see that heat treatment offers significantly higher stren Alloy 2017, the original duralumtn, is probably the oldest age-hardenable aluminu alloy.The 2024 alloy is stronger and has seen considerable use in aircraft appl~ca ' An attractive feature of the 2xxx series 1s the fact that ductility does not significant1 crease during the strengthening heat treatment. Withln the 7xxx series are some ne alloys with strengths that approach or exceed those of the h~gh-strength structuralst Ductlhty, however, is generally low, and fabrication is more diffreult than for the 2 type alloys. Nevertheless, the 7xxx series alloys have also found wide use in aircr applications. To maintain properties, age-hardened alloys should not be used at te peratures over 175C (350F). Welding should be performed w ~ t h considerable caut' since the exposure to elevated temperature wlll significantly d~minish the strengthenin achieved through either cold working or age hardening. Because of t h e ~two-phase r structure, the heat-treatable alloys tend to have po er corrosion resistance than either pure aluminum or the single-phase work-hardena alloys When both high strength and superior corrosion resistance are desired, wrou aluminum is often produced as Alclad material. A thln layer of corrosion-resist aluminum is bonded tonne or both surfaces of a high-strength alloy dunng rolling, the material is further processed as a composite. Because only moderate temperatures are requ~redto lower the strength aluminum alloys,extrusions and forgings are relatively easy to produce and are m ufactured in large quantities. Deep drawlng and other sheet-metal-forming o tions can also be carried out qulte eas~ly. In general, the high ductihty and low strehgth of the aluminum alloys make them appropnate for almost all forming erations Good dimensional tolerances and fairly 1n1ricate shapes can be produ with relat~ve ease The machinabihty of aluminum-based alloys, however. can vary greatly, and sp f machining are require cia1 tools and techn~ques may be deslrable if large amounts o Free-machining alloys, such as 2011, have been developed for screw-machine wo These special alloys can be machrned at very high speeds and have replaced brass scre machine stack in many applicatrons

Lomposlnon 17aj Alum~num = Balance DestgnatinnVu

11W4 llW-HI4 110-HI8 3W3-W M03-HI4 3003-HI8 Sm2-0 5052-H32 5052-H36
20174 2017-T4 20244 2024434 2014-0
St

Streneth kst

Streneth"

Elaneauon in Bnnell

Mn

Mg

Others FormTes%ed
9Y A1

MPa

ksl

MPa

2 in. ( O X )

Hardness

Uses and Charactenst~c$ Commerc~al Al: good iorm~ng properues Good corroston resrstanoe,low yield skenglh and tub~ng Cuokrng ulcns~ls.sEet tV l l m $lmdd~ Sl~ghlly strunger and 1~5s duucle Cooking ~tensdq,sheet-metalwrtrk Strongest work hardenmg~lloy Highly yteld soength a d fafaUgue Bm~t &ghly stressed sheet-metal products Duralumin, ortgrnal strong alloy aild a p g Hardened by quench~ng Sva~tger than 201 7 Uspd w~dely in aircraft wnstructlon Stmng alloy for extruded shapes Strong forgrng allay H~gher y~eld strength than Alelad 2024 elad with heat-treatable alloyc Alloy of highest strength Lower riuchl~ry than 2024 Strongest Aldlad product Strungest alioy for erlruslcrns

0 12

0 '"

Work-Hardenmg Allays-Not Heat-Treatable 13 9a 5 34 A-m sheet I6 110 14 97 $-in sheet 24 It% 21 145 +-in sheet 16 110 6 dl
&-m \heel
1. Td-m.shcet

22

152

21

2.5

&-~nsheet 0.25 Cr $-in sheel i6-m sheet &-in sheet

29

200

28 33 40

193
228 276

27 13 28 35

145 186 90

lY3 241

48

05

07
06

06
15

Preclprtation-FTardenlng Alloy8-Heat-Treatable A-".sheet 26 179 10 69

m sheet & ~ n sheet &-m sheet h n , extruded
shapes

62

44
44 08

2 7
64 27 65 70 67 33 76
76 80

428 186

0.8

05

441 186
488

40 11 45 14
55
OW

276 76 290 97

2OIeT6 2014-36 Alclad

z O 1 4 T B

Forgings & m. sheet 45 16

10

483

379 413
386

70754 7075-T6 Alclad 7075-T6 707FT6

6061-?K

08 0.2

04 2.5

0 3 Fr 5 6 Zn

&-ID.sheet &-m sheet & in.sbwit

434 228
52d
524

56
I5

67
67 70 40

103 462 462 4@

028

06
0.4

b0&6
6151-T6 2025-T6 201&T6

45

09 08

0.8

~ - T S
2Ull-11
4

0 9 122
55

in sheet $-in extruded \hapei 0 020 Cr 4-m. exttuded shapes 1 1 . 7 :-ln rod extruded 0.6 025Cr Fo-a For@ns 07 ZNi Forslgs II U9Ni Forgsn& ( 0 . 5B i ) 0.5 Pb i m rod

552
290

42
35
48 55 SS

276
214 Good fomung propeities and corrosion resistance Far mtsieare lr>rgmgi Good forgeabdiiy, Lower cosl Slrong aLr1evated temperatures. forged pston\ Forged alrcrait plrlon5 Free cuttmg, s~rew-machme pnldu~ts

241
331

31
43 30 40 4.6 43

297
207

55 55

379 379 379 378

276
317
297

10
9

I5

100 115 95

'0

annea1ed.T. quedched and aged. H Cold roUd tulurd temper

V~ieldstren$$b lakonat 0 2 % permanent sot

"Clnddmgalley,I liMg 0 7 91.0 5 MI

Nonferrous Metals and Alloys Color anodizing offers an inexpensive and attractive means of surface finishing. A thick aluminum ox~de IS produced on the surface. Colored dye is then placed on the porous surface and is sealed by immersron into hot water.The result is the colored metallic finish commonly observed on products such as bicycle frames and softball bats.

ALUMINUM CASTING ALLOYS Although its low melting temperature tends to make it suitable for casting, pure alumlnum is seldom cast. Its high shrinkage upon solidification (about 7%) and susceptibility to hot crackmg cause considerable difficulty, and scrap is high. By adding small amounts of alloying elements, however, very suitable casting characteristics can be obtained and strength can be mcreased. Aluminum alloys are cast in considerable quantity by a variety of proeesses.Many of the most popular alloys contain enough sllicon to produce the eutectlc reaction, whichis characterized by a low meltmg pomt and bigh ascast strength. S111wnalso improves the fluidty of the metal, makmg it easier to produce complex shapes or thin sections, but high sihcon also produces an abrasive, d~fficnlt-to-cut material. Copper, zinc, and magnesium are other popular alloy additions that permit the formation of age-hardening precipitates. Table 7-5 lists some of the commercial aluminum casting alloys and uses three-digit designatron system of the AluminumAssociation to designate alloy chemi The first dlgit Indicates the alloy group as follows
Major Alloying Element
Alummum, 99.00% and greater Copper Elicon w ~ t h Cu mdinr Mg Slllcon
Magnesium
Zmc

1xx.x
2xx x 3xx x

4xx x
5xx x 7xx x

Tm
Other elemcuts

8xx x Yxxx

The second and third dig~ts identify the particular alloy or aluminum purity, and the la digit, separated by a decimal point, indicates the product form (e.g., casting or ingot). letter before the numerical designation ~ndicates a modification of the original alloy, s as a small variation in the amount of an alloying element or impunty. Aluminum casting alloys have been deslgned for both properties and process, the strength requirements are low, as-cast properties are usually adequate. mgh-streng castings usually requre the use of alloys that can subsequently be heat treated. Sa castmg has the fewest process restrictions. The alummum alloys used for permane mold casting are designed to have lower coefficients of thermal expansion (or contra tion) because the molds offer restraint to the dimensional changes that occur upon co ing Die-castingalloys require high degrees of fluidity because they are often cast in thj sections. Most of the die-casting alloys are also designed to produce high "as-cas strength without heat treatment, uslng the rapid cooling conditions of the die-cas process to promote a fine grain size and fine eutectic stwcture.Tensile strength o aluminum permanent-mold and d~e-casting alloys can be in excess of 275 MPa (4

ALUMINUM-LITHIUM ALLOYS Lithium is the lightest of all metallic elements, and in the search for aluminum alloys wi higher strength, greater stiffness, and lighter weight, aluminum-lithium alloys ha emerged. Each percent of hthium reduces the overall weight by 3% and increases s ness by 6%. The initially developed alloys offered 8 to 10% lower density, 15 to 20 greater stiffness,strengths comparable to those of existlng alloys,and good resistance fatigue crack propagation. Unfo~tunately, fracture toughness, ductility, and stres corrosion resistance were poorer than for conventional alloysThe current-generati

Compos~tion (%)

(Major Alloys > 1%)

Zn
1.0

Tensile Strength
Other

Alloy
Designabon"

Fe
12

Temper ksiC
F
iY

MPa 131 276

221

Elongation in Uses and Chracteristics 2 in (?(I> I5 General-purposes and caattngs can be heat treated Withstands elevated temperatures Structural casungs. heat treatable Permanent-mold version Of295 Generai-purposepermanent mold Superlor ca-lmg characttrishcs High-strength.aucraf1 Highmensh and pressura tightness Stronger and more ductile than
355

241 166 214

bcellent castabiltty and impact strength Snanger and more ductile than 356 High strength-to-weight castings Hrgh-strength atrcraft usage Good cornoakon resistance and &en& Slmllar to 360 High strength and hardness Smilar to 380 High stmngth and hardness H~gh strength and hardness General-purpw, g o d castab~lity Sirmlar to413 Generd-purpose. good castability General-purposesast~nganoy H~gh corrosion resistance Gaad corras~on res~.itancc.strength,and toughness HL&strength withgood ductlhty res~stanceandmachinnhil~ty Good conws~on Good propwhes wirhout heat treatment lo 712 S~mdar A~roraft and computer oomponentq Bearmg alloy

152

C m m 7

Nonferrous Metals and Alloys alloys are alummum-copper-lithium,with ahout 4 % copper and no more than2% lithium.The weight benefits are still sufficient to warrant use in a number of aerospace applications, and the fact that they can be fabricated by conventional processes make them attractive alternatives to the advanced composite?. Since aluminum alloys can comprise as much as 80% of the weight of commercd alrcraft, even small percentage reductions can be significant. Improved strength and stiffness can further facilitate werght reduction. Fuel savlngs over the life of the arrplane would more than compensate for any additional manufacturing expense. As an example of potential, the weight of the external liquid-hydrogen tank on the U.S. space shuttle booster rocket was reduced by approximately 3400 kg (7500 lb) by conversion to an aluminum-lrthium alloy.

ALUMINUM FOAM A material known as "stabilized aluminum foam" can be made by mixing ceramic particles wlth molten aluminum and blowing gas into the mixture.The bubbles remain through sohdification, ylelding a structure that resembles metallic Styrofoam. Original. ly developed around 2000 for automollve, aerospace, and military apphcations, the material has found addiaonal uses in architecture and design. Strength-to-weight is outstand~ng, and the matenal offers excellent energy absorption. The fuel cells of race cars have been shrouded with aluminum foam, and foam fill has been inserted between the front of cars and the driver comparhent.Tubular structures can be filed with fo to increase strength, absorb energy, and provide resistance to crushing. Still other a plicalions capitalue on the excellent thermal msulation, vibration damping, and soun absorption that rezults from the numerous trapped air pockets.

AND

MAGNESIUM ALLOYS

GENERAL PROPERTIES AND CHARACTERISTICS Magnesium w the lightest of the commercially important metals,having a specific gravi of about 1.74 (two-thirds that of aluminum, one-fourth that of steel, and only slight higher than hber-reinforced plast~cs). Like aluminum, magnesium n relatively weak 1 the pure state and for engineering purposes is almost always used as an alloy. Even i alloy form, however, the metal 1s characterized by poor wear, creep, and fatigue pro erties. It has the highest thermal expanston of all engineenng metals. Strength d rapidly when the temperature exceeds 100C (200F), so magnesium should not be sldered for elevated-temperature service. Its modulus of elastinty is even less than ih o f aluminum, being between one-fourth and one-fifth that of steel.%ck sections are r quired to provlde adequate stiffness, but the alloy is so hght that it is oftenposs~hle use thicker seclions for the required rigihty and still have a lighter structure than be obta~ned with any other metal. Cost per unit volume IS low, so the use of thlck lions is generally not proh~bitiveMoreover, since a large portion of magnesium c ponents are cast, the thicker sections actually become a desirable feature. Ductihty 1 frequently low, a character~stic of the hexagonal-close-packed (HCP) crystal strnctur but sqme alloys have values exceedmg 10%. On the more positive side, magnesium alloys have a relatively h ~ g h strength4 weight ratlo, with some commercial alloys a t t h g strengths as high as 380 MPa (55 ks High energy absorption means good damping of noise and vibration, as well as imp and dent resistance. While many magnesium alloys require enamel or lacquer finish to impart adequate corrosion resistance, this property has been improved markedly w the development of hlgher-purity alloys. In the absence of unfavorable galvanic u, ples, these materials have excellent corrosion resistance and are findmg applicatio a wide range of markets, including automotive, aerospace, power tools, sporting and electronic products (where they offer a combination of electromagnet~c shie light weight, and durability exceeding that of plastics and alternative metals). While minum alloys are often used for the load-bearing members of mechanical stmctu magnesium alloys are best suited for those applications where Lightness is the prim consideration and strength is a secondary reqmrement.

SECTION 7.4 Magnesium and Magnesium Alloys

153

MAGNESIUM ALLOYS AND THEIR FABRICATION A designation system for magnesium alloys has been developed by the ASTM, identifying both chemical composition and temper, and is presented in specification B93.Two prefix letters designate the two largest alloying metals in order of decreasing amount, using the following format:
A B C D E aluminum bismuth copper cadmium rare earth

F iron H thorium K zirconium L beryllium

M N P Q

manganese nickel lead silver

R S T Z

chromium silicon tin zinc

Aluminum is the most common alloying element and, along with zinc, zirconium, and thorium, promotes precipitation hardening. Manganese improves corrosion resistance, and tin improves castability. Thc two letters are then followed by two or three numbers and a possible suffix letter. The numbers correspond to the rounded-off whole-number percentages of the two main alloy elements and are arranged in the same order as the letters.Thus the AZ91 alloy would contain approximately 9% aluminum and 1% zinc. A suffix lettcr is used to denote variations of the same base alloy, such as AZ91A. The temper-designation sulfix is quite similar to Lhat used with the aluminum alloys.Table 7-6 lists some of the more common magnesium alloys together with their properties and uses. Sand, permanent-mold, die, semisolid, and investment casting are all well developed for magnesium alloys and take advantage o l the low melting points and high fluidity. Die casting is clearly the most popular manufacturing process for magnesium, accounting for 70% of all castings.Although the magnesium alloys typically cost about twice as much as aluminum, the hot-chamber die-casting process used with magnesium is easier,more economical, and 40 to 50% faster than the cold-chamber process generally required for aluminum. Wall thickness, draft angle, and dimensional tolerances are all lower than for both aluminum die castings and thermoplastic moldings. Die life is significantly greater than that observed with aluminum. As a result, magnesium die castings compete well with aluminum2 and often replace plastic injection-molded components when improved stiffness or dimensional stability, or the benefits of electrical or thermal conductivity, are required. Forming behavior is poor at room temperature, hut most conventional processes can be perconned when the material is heated to temperatures between 250' and 500C (480" and 775F).Since these temperatures are easily attained and generally do not require a protective atmosphere, many formed and drawn magnesium products are manufactured. Magnesium extrusions and sheet metal products have properties similar to the more common wrought aluminum alloys. While slightly heavier than plastics, they offer an order of magnitude or greater improvement in stiffness or rigidity. The machinability of magnesium alloys is the best of any commercial metal and, in many applications, the savings in machining costs, achieved through deeper cuts, higher cutting speeds, and longer tool life, more than compensate for the increased cost of the material. It is necessary, however, to keep the tools sharp and provide adequate cooling for the chips. Magnesium alloys can be spot welded almost as easily as aluminum, but scratch brushing or chemical cleaning is necessary before lorming the weld. Fusion weldingis best performed with processes using an inert shielding atmosphere of argon or helium gas. While heal treatments can be used to increase strength, the added increment achieved by age hardening is far less than observed with aluminum. In fact, the strongest magnesium alloy is only about three times stronger than the weakest. Because of this, designs must be made to accommodate the material, rather than the material being tailored to the design. Considerable misinformation exists regarding the fire hazards when processing or using magnesium alloys. It is true that magnesium alloys are highly combustible
'TIE most commonmagneslum d~e-casung alloy AZ91, has the i d m e yield alrengfh and ductility as the moat wmrnon d ~ cast e alummum. alloy 3M

154

C H A P ~7 ~ R Nonferrous Metals and Alloys

Cornpaition (%)
Allay
AM@A AMI~A
AZXLB

Tempffir A1
F

Rare Earths Mn
0.13

Th Zn

Zr

ksi
30 34

lvma
207 234
22'

lrsi
17 10
*-

MPa
117 69
103

Elongation in Uses and 2 in. (%) Charactemtics

6
6
"

T 4
F
P
Tj

6.0 la D
30 65

0 1 s

Sand end permanent-

-36 34
34

mold rastulgs
e x m o u s , forgIn@

AmIA AZ63A

248

16
11

110

Sheet,piale, ext~usi~m Sanrl and permanent-

6 . 0
8 . 5
7.6

234
2%

76
142
2

A
AZ81A =VIA A&ZA

TS

22
3 4 1

extruaons

T4

34
34 34

234
234
3

76
159 76

% M a d I permanvnt-

F T4

9 . 0 9 . 0

23
1 1

3
6

and parmanen(-mntd

EZ33A HK31A HMZIA

T5
?B4

Sand and p m a n e n t Sheet and plater, castings inT6 tmper

HI@-maperatme

T9

HBZA

15

pI~t~~,f~r&ings Sand and permanent3. I

. l &
LSL

ZH6.2A
ZKfl A

T T
TS

Sand and permaeene5

46

R 7 .

34

234

20

138

Sand and permanent-

17 . 5 ZINC-BASED ALLOYS
log.The resultant coatkg provides awIlent corrosion rwistance, even

aU of the sacrificial zinc ba8 baen rl'epleted. Zinc is also used as ihe bsse metal for avariety of diecasting allvys, For this

property of not adversely affecting steel dies when in contact with molten metal. Unfortunately, pure zincis almost as heavy as steel and is also rather weak and brittle. Therefore, when alloys are designed for die casting, the alloy elements are usually selectedfor their ability to increase strength and toughness in the as-cast condition while retaining the low melting point. The composition and properties of common zinc die-casting alloys are presented in Table 7-7.Alloy AG40A (also known as alloy 903 or Zamak 3) is widely used because of its excellent dimensional stability,and alloy AW1A (also known as alloy 925 or Zamak 5) offers higher strength and better corrosion resistance. As a whole, the zinc die-casting alloys offer a reasonably high strength and impact resistance, along with the ability to be cast to close dimensional limits with extremely thin sections. The dimensions are quite stable, and the products can be finish machined at a minimum of cost. Resistance to surface corrosion is adequate for a number of applications, and the material can be surface finished by a variety of means that include polishing,plating, painting, anodizing, or a chromate conversion coating. Energy costs are low (low melting temperature), tool life is excellent, and the zinc alloys can be efficiently recycled. While the rigidity is low compared to that of other metals, it is far superior to engineering plastics, and zinc die castings often compete with plastic injection moldings. The attractiveness of zinc die casting has been further enhanced by the zinc-aluminum casting alloys (ZA-8,ZA-12, and ZA-27, with 8,12, and 27% aluminum, respectively). Initially developed for sand, permanent-mold, and graphite-mold casting, these alloys can also be die cast to achieve higher strength (up to 60 ksi or 415 MPa), hardness (up to 120 BHN), creep resistance and wear resistance, and lighter weight than is possible with any of the conventional alloys. Because of their lower melting and casting costs, these materials are becoming attractive alternatives to the conventional aluminum, brass and bronze casting alloys, as well as cast iron.

Titunium is a strong, lightweight,corrosion-resistantmetal that has been of commercial importance since about 1950.Because its properties are generally between those of steel and aluminum, its importance has been increasing rapidly. The yield strength of commercially pure titanium is about 210 MPa (30 ksi), but this can be raised to 1300 MPa (190 ksi) or higher through alloying and heat treatment. a strength comparable to that

Nonferrous M~tals and Alloys of many heat-treated alloy s%eelS. Density, an the other hand, IS only 56% that of steel (making strength-to-weight quite amactive), and the modulus of elasticity ratio i s also abo# one-half. Good mschanicd pr~petfies are retained up to temperatureso'f53SQC C1OODaFJ, so the n#tal is often cone~dered to be a high-temperature m n e e a n g material. On thenegative side, titanium and its alloy@ suffer from high cmt, fabrication dBcultieg a high strergy cantent (they require about 10 times as much energy t~ prodm a$ amel), and a high reactivity at elevrrted temperatures (above 535C). Tftanium alloys are designated by majm alloy and mourn (see ASThnspecificanon 8-2651, and are generallygrouped into three clac;sm ba@ ofi &eir micro~trucml Y e a tufes.These classes ase h o w a as alpha-, beta-, and alpha-bpta-titanium alloys, the 'te denoting the stable phase or phases at room temperature.Alloying elementscan he to stab~lize the hexagonal-close-p&ked alpha phase or the bady-centered-cubi~ phage, and fieat trements can be applied to manipulatestructure and improve pro tie$ Fabrication can be by casting (generally invesfment or graphite mold),forging,rol extrusion, or welding, provided that special procas modification$ and controls implemenmdd, Advanced prucessiag methods indude powder metallurgy,mecha albying, rapid-solidification processmg (RSP), superplastic fsrming, diffusion boa and hot+isostatic pressing (KIP). While titanium is an &undant metal, it is dfficult to extract from ore, d~ffictilt proces and difficult to fabnmte. ' I % difficulties & make it significantly more exp khan eithar step1or aluminum,so its uses d a t e pr-nar~ly to its light wes@~t,h&~ .str to-weigh*ratio, good stiffness, good fatigue strength and fracture toughness,excellent r&on resistance (the result of a thin, tenacious oxide coating], and the reten ' mechanicalpropertie8 at elevated temperaturesAltnnmum,magnesim,and berylli the only base metals W a r e ligber than titanium, and none of these come close inei m f ~ c I pedormanw r ~ welwakd-temperature p r ~ r t t e s , A e r q a c e appllcati to dom~nate, with titanium comprising up to 40% 0 1 the stmctural weight performance milit- fi&ttas.Tlta~um and titanium alloys are also wed in swh div areas as chemical- and e l e s t r o c h e m i c a - o e g e q u i p m e n t o d equ heat exchangecs, marine implement^, medical implants, high-perfonnanee btcycle tornotive components, and sportiig goo& They are often used in place of weight s a v i n are ~ desitedand to replm aluminurnsMihere lugh-tempe~atuie pezfom is nemsary, Same hdingapplications.utilize the unique property that titanium wets and some ceramia.The t i t a n i d % alu&um-4% vanadium alloy i s the mast pop titanium alloy,accountingfor nearly 50% of all titanium usage worldwide.Egnre 7-4 sho the elevarrsd temperaturB~trengthsetention of several titanium alloys
2W

150
U1 3
L

1 TI-13V-I
-beta)

7CrdAI

B 700 $3
t 5
TI-GAI4V

3 2(alnhal
Pure TI
& % B
180
200 300 *Do 500

at elevated FIGURE 7-4 temperature Strength retenio for

various tltanlurn alloys.

Ternpergture PC)

S~cnoN 7.8 Superalloys and Other Metals Designed for High-Temperature Service jfsteel 6s also 535C matco drffiioduce [cation ial feal terms p used c beta roperroUlng, )Is are tanical mding,
I

157

ALLOYS

cult to ~ensrve ;e"gthm n t eorlron of um are (either os tend t highdiverse pment, m d anwhere rmance its glass topular I shows

Nickel-based alloys are most noted for their outstanding strength and corrosion resistance, particularly at high temperatures, and are available in a wide range of wrought and cast grades. Wrought alloys are generally known by tradename6,such as Monel, Hastelloy, Inconel, Incoloy, and others. Cast alloys are generally ~dentified by Alloy Castrng Institute or ASTM designat~onsGeneral character~stlcs~ncludegood formability (face-centered-cubic crystal structure), good creep resistance, and the retention of strength and duct~lity at cold or even cryogemc temperatures. Monel metal, an alloy containing about 67% nickel and 30% copper, has been used for years in the chemical- and Iood-processing industries because of its outstandrng corroshon charactenstm. In fact, Mollel probably has better corrosion resistance to more media than any other commercial alloy. It is particularly resistant to salt water, sulfuric and, and even high-velocrty, h~gh-temperature bream. For the latter reason, Monel has been used for steam turbine blades. It can be polished to have an excellent appearance, srm~lar to that of stamles? stee1,and is often used in ornamental trim and household ware. In its most common form.Mone1 has a tensile strength ranglng from 500 to 1200 MPa (70 to 170 ksi). w ~ t h a companion elongat~on ranglng between 2 and 50% Ntckel-based alloys have also been used for electrical resistors and heating etements. These materlals are prlrnaniy n~ckel-chrom~um alloys and are known by the trade name N1chrome.They have excellent reslstance to ox~datron while retaining useful blrength at red heals Invar,an alloy o1nlckel and 36% Iron, has a near-zero thermal expansion and is used where dlmenslons cannot change with a change ~ntemperature. Other nickel-based alloys have been designcd to provide good mechan~cal properhes at extremely high temperatures and are generally classified as superalloys.These alloys will be discussed along with olhcr, ~ ~ p i lmaterlals ar in the following sectlon.

1 '1
'

iND

OTHER METALS DESIGNED

'PERATURE SERVICE
Titan~umand titanium alloys have already been cited as being useful in providing strength at elevated temperatures, hut the maxlmum temperature for these materials is approximately 53SnC(10000C) Jet engine, gas-turbine, rocket, and nuclear applications often requlre matenals that possess high rtrength, creep resrstance. oxidation and corrosion reslstance, and fatigue reslstance at temperatures up to and in excess of 1100C (2000C). Other apphcation areas include heat exchangers, chemical reaction vessels, and furnace components One class of materials offeringthese properties 1s the ~i~peralloys, first developed In the 1940s for use in the elevated-temperature areas of turbojet aircraft.These alloys . are based on nickel, zron and nrckel, or cobalt and have the abllity to retaln most of their strength even after long exposures to extremely high temperatures. Strength comes from solid-solution strengthemng, precipltatlon hardcnmg, and dispersed alloy carbides or ox~des.The nickel-based alloys tend to have higher strengths at room temperature, with y~eld strengths up to 1200 MPa (175 k s ~ and ) ultimate tensile strengths a5 hlgh as 1450 MPa (210 ks~).The 1000-hour rupture strengths of the nickel-based alloys at 81SnC (1500F) are also h~gher than those of the cobalt-based mater~al. Unfortunately, the density of all superalloy metals is signtficantly greater than that of iron, so their use 1s often at the expenqe of additional welght. Most ot the superalloys are diflcult to Corm or machme, so methods such as electrodischarge,electrochem~cal. or ultrasonic machining are often used, or the products are made to final shape as investment castrngs. Powder metallurgy techniques are also used . extensively Because of their ingred~ents, all of the alloys are qutte expenswe, and this limlts their use to small or critical parts where the cost 1s not the determ~ning factor. , A number of engineering applications require materials whose temperature lim~ts ex- ' ceed those of the superalloys.F1gure7-5 shows the high-temperature exhaust of a jet engine. nickel (a powder metallurgy nlckel alloy cantainmgZ% dispersed thorium oxide) can operate

158

CHA~W 7 Nonferrous Metals and Alloys

of jet engine exhaust. (Courtesy Northrop Crummon Corporatron, Los Angeles, Cd.)
- "F

-- 2200 Superalloys

-- 1000 Tltanlum - 650 Alloy steels -- 500 Carbon steels -- 350 Aluminum

fradory metals could be used at higher tempeFaiure6 the uppermost temperature is c rently being set by limitations and restncaons imposed by the coatmg. Table 7-8 presents key properties for several refractoqmetals Unfortunately,all heav~er than steel, and several are slgmficantly heavier. In Fact, tungsten, wth a de about 1.7 times that of lead@ often used m counterbalances, compactflywheels,and we with applications as diverse as mihtary projectiles, gyratory wmpasseh and golf clubs. Other materzals and technologies that offer promise for high-temperature service i clude intermetallic cbmpounds, engineered ceramics, and advanced coating systems Tb

ms and they are excellent candidates for high-temperatureapplicationsThey are hard, s

+
to useful mechanical propert~es for various engineer~ng metals. Figure 7-6 compares the upper limit for useful mechanical properties for a varie of engineering metals.

7.9 LEADAND TIN,AND THEIR ALLOYS

The dominant properties of lead and lead alloys are h ~ g h density coupled with stren and stdfness values that are among the lowest of the engineering metals. The princ'

Tantalum

5430 (3000)

absorbing or sound- and vibration-damplng shields. Lead-ac~d batteries are clearly the dominant product, and over 60% ot U.S. lead consumption IS generated tram battery recycling. Other applications utilize the properties of good corrosion resrslance, low melting point, and the ease of casting o r fommg. As a pure mctal, tin IS used primarily as a corrosion-resistant coatrng on steel. In the form of alloys, lead and tin are almost always used together. Bearimg material and solder are the two most Important uses. One of the oldest and best bearlng materrals is an alloy of 84% tin, 8% copper, and 8% antlmony, known as genuine or tin f the high cost of tin, however, lead babbitt,composed oL85% lead,% babbita. Because o tin, 10% antimony, and 0.5% copper, is a more widely used bearlng material The tin and antlmony combine to form hard particles with the softer lead matrm.The shaft rides on the harder particles with low fr~ct~on, while the softer matrix acts as a cushion that can dlstort sufficiently to compensatc for rmsallgnment and assure aproper f ~between t the two surfaces. For slow speeds and moderate loads, the lead-based babbitts have proven to be quite adequate. Soft solders are basically lead-tin alloys with a chemical composition near the eutectlc value of 61.9% tin (see Figure 4-5). Wh~le the eutcctic alloy has the lowest meltIng temperature, the high cost of lin has forced many users to spec~fy solders w ~ t h a lower-than-optimum tin content. A varlety of composltlons are available, each wlth its own characteristic meltlng range. Env~ronmental concern5 and recent legislation have prompted a move toward lead-free solders lor apphcatlons ~nvolv~ng water supply and distribution. Additional information on solders and boldenng is provided m Chapter 34.

Several of thelesser known metals have ach~eved Importance as a result of their somewhat unique phys~cal and mechanical properties Beryllrum comhlnes a densrty Less than aluminum with a stiffness greater than steel and 1s transparent to X-rays Hafnium, thorium, and beryllium are used in nuclear reactors because of their low neutron-absorptlon charactenstlcs. Depleted uranrum,because oi ~ t very s h~gh density (19.1 g/cm3),is useful in spccial applications where maxlmum weight must be put into a hmlted space, such as counterwe~ghts or flywheels. Cobalt, m addltlon to I & use as a base metal for superalloys, is used as a binder in various powder-based components and srntered carbides, where il provldes good high-lemperature strength. Zrrconium IS used for its outstanding corrosion resistance to most aclds, chlorides, and organic acids. It offers high strength, good weldability and fatigue reststance, and attractive neutron-absorption charscterlstles. Rare earth melals have been Incorporated into magnets that offer increased strength compared to the standard ferrite varlety Neodymium-rron-boron and samarlumxobalt are two common varieties. Wh~le the precious metals (gold,silver, and the platlnum group metal-platinutu, palladium, rhodfum,ruthenium, indrum, and osmium) may seem unlikely as engineering materials, they offer outstanlng corros~on resistance and electrical conductivity, often under extreme conditions of temperature and environment

Metalhc glasses, or amorphous metals, have ex~sted in the form of thin ribbons and fine powders slnce the 1960s By cool~ng liquid metnl at a rate that exceeds lo5 to 1 0 V l second,a rig~d sohd is produced that lacks crystalline structure. Since the structure also lacks the crysfalline"defects" of grain boundaries and dislocations, the materials exhibit strength, large elastic stram, good toughness, extraordinary mechanical properties (h~gh and wear resistance), unusual magnetic behavior, and high corrosion resistance. Recent developments have enabled the productlon of amorphous metal with cooling rates of only 1 to 100 "Cisecond. Known as bulk metullrc glads (BMG),complexshaped parts of this material w ~ t h thcknesses up to several c e n h e t e r s can now be produced by conventional casting methods, such as die casting. Because the mater~al goes from liquld to glass, not l~quid to crystalhne sohd, preclslon products can be made with a total shrinkage that is often less than 0.5%. Pellets or powders of hulk metallic

Nonferrous Metals and Alloys glass can also be produced, and since many of the alloys have low melting temperatures, products can be made by reheating to a soft condition and forming by processes that are conventionally used to shape thermoplastic polymers (compression molding, extrusion, blow molding, and injection molding).Applications have just begun to emerge in areas as diverse as load-bearing structures, electronic casings, replacement joints, and sporting goods. In addition, metallic glasses have also been developed that retain their glassy structure at temperatures as high as 870C (1600F).

While technically not a meta1,graphite is an engineering material with considerable potential.It offers properties of both ametal and nonmetal,including good thermal and eleo trical conductivity, inertness, the ability to withstand high temperature, and lubricity. In addition,it possesses the unique property of increasing in strength as the temperature elevated. Polycrystralline graphites can have mechanicalstrengths up to 70 MPa (10 ks at room temperature, which double when the tempkrature reaches 2500C (4500F). Large quantities of graphite are used as electrodes in arc furnaces, but other u are developing rapidly. The addition of small amounts of borides, carbides, nitrides, a silicides greatly lowers the oxidation rate at elevated temperatures and improves th mechanical strength.This makes the material highly suitable for use as rocket-nozzle in serts and as permanent molds for casting various metals, where it costs less than tool steel requires no heat treating, and has, a lower coefficient of thermal expansion. It can b machined quite readily to excellent surface finishes. Graphite fibers have also foundex tensive use in composite materials. This application w i l l be discussed in Chapter 8.

I I Key Words
Alclad aluminum amorphous metal babbitt beryllium bismuth brass bronze cast cobalt copper dezincification galvanizing graphite intermetallic compound lead magnesium metallicglass molybdenum Monel nickel niobium nonferrous refractory metals rhenium selenium solder stress~orrosion cracking superalloys tantalum temper designation tin titanium tungsten wrought zinc

ITI Review Questions

I . What types of properties do nonferrous metals possess that may not be available in the ferrous metals? 2. In what respects are the nonferrous metals generally inferior to steel? 3. For what type of fabrication processes might the low-meltingpoint alloys be attractive? 4. What are the three properties of coppcr andcopper alloys that account for many of their uses and applications'? 5. What properties make copper attractive for cold-working processes? 6. What are some of the limitingpropertiesof copper that might restrict its area of application'? 7. Why does the copper designation system separate wrought and cast alloys?What properties are attractive for each group? 8. What are some of the attractive engineering properties that account for the wide use of the copper-zinc alpha brassesl 9. Why might cold-worked brass require a stress relief prior to being placed in service? 10. Why might the term bronze be potentially confusing when used in reference to a copper-based alloy?

11. What are some attractive engineeringproperties of coppe nickel alloys? 12. Describe the somewhat unique property combination thatel ists in heat-treated copper-beryllium alloys. What has lid@ its use in recent years? 13. What alloys have been used to replace lead in coppercas alloys being targeted to drinking water applications? 14. What are some of the attractive engineering properties of a1 minum and aluminum alloys? 15. How does aluminum compare to steelin terms of weight? cuss the merits of comparing cost per unit weight versus c per unit volume. 16. What is the primary benefit of aluminum recycling cornpar I to making new aluminum from ore? 17. How does aluminum compare tocopper in terms of electri conductivity? 18. What features might limit the mechanical uses and appliq tions of aluminum and aluminum alloys? 19. What features make aluminum attractive for transportatio applications?

case study

161

. How is the corrosion-resistance mechanism observed in aluminum and aluminum alloys similar to that obsewed instain-

1. How are the wrought alloys distinguished from the cast alloys in the aluminum designation system? Why would these two f metals have distinctly different properties? groups o 2. What feature in the wrought aluminum designation scheme is used to denote the condition or structure of a given alloy? What is the primary strengthening mechanism in the highstrength "aircraft-quality" aluminum alloys? 4. What unique combination of properties is offered by the composite Alclad materials? 5. What surface finishing technique is used in the production of numerous metallic colored aluminum products. 6. What specific material properties might make an aluminum casting alloy attractive for permanent mold casting? For die

33. What is the primary application of pure zinc? Of the zincbased engineering alloys'? 34. What are some of the attractive features of the zinc-aluminum casting alloys? 35. What are some of the attractive engineering properties of titanium and titanium alloys? 36. What feature is used to provide the metallurgical clsssification of titanium alloys? 37. What temperatureis generally considered to be the upper limit for which titanium alloys retain their useiul engineering properties? 38. What conditions favor the selection and use of nickel-based alloys? 39. What property of Mouel alloys dominates most of the L applications? 40. What metals or combinations of metals form the bases of the

,es
nd

he
in-

el, be :x-

aluminum-lithium alloys? 8. What are some possible applications of aluminum foam? 9. What are some attractive and restrictive properties of magnesium and magnesium alloys? . Describe the designation system applied to magnesium alloys.

42. Which metals are classified as refractory metals? 43. What are some general characteristics of intermetallic compounds? 44. What is the dominant product for which lead is used?

. Under what conditions should magnesium be considered to

be a flammable or explosive material?

46. What are some of the attractive properties of metallic glasses? 47. What unique property of graphite makes it attractive for elevated-temperature applications?

he ivanwoldKendoric tool-manufacturing company is hand tools to include safety tools capable of being used in areas such as g a s leaks where the potential of explosion or fireexists. Conventional irons and steels are pyrophoric (i.e., small slivers or fragments can burn in air, forming sparks if dropped or impacted on a hard surface). You are asked to evaluate potential materials and processes that might be used to manufacture a nonsparking pipe wrench.This product is to b e produced in the same shape and range of sizes a s conventional pipe wrenches and needs to possess all of the same characteristic properties (strength in the handle, hardness in the teeth, fracture resistance, corrosion resistance, etc.). In addition, the new safety wrench must b e nonsparking (or nonpyrophoric). Your initial review of the nonferrous metals reveals that aluminum is nonpyrophoric but lacks the strength and ic but is heavier than

steel, and this may be unattractive for the larger wrenches. Copper-2% beryllium can b e a g e hardened to provide the strength and hardness properties equivalent to the steel that is currently being used for the jaws of the wrench, but the Wst of this material is also quite high. Titanium i s difficultto fabricate and may not possess the needed hardness and wear resistance. Mixed materials may create an unattractive galvanic corrosion cell. Both forging and casting appear to b e viable means of forming the desired shape.You want to produce a quality product but also wish to make the wrench in the most economical manner possible s o that the new line of safety tools is attractive to potential customers. Suggest some alternative manufacturing systems (materials coupled with companion methods of fabrication) that could be used to produce the desired wrench. What might be the advantages and disadvantages of each? Which of your alternatives would you recommend to your

1
I

T
I
I

NONMETALLIC MATERIALS: PLASTICS, ELASTOMERS, CERAMICS, AND COMPOSITES

8.1 INTRODUCTI~N 8.3 ELASTOMEKS

8 2 PLASTICS Molecular Structure of Plastics Isomers Forming Molecules by Polymer~zation Thermosetting and Thermoplastic Materials Properties and Applications CommonTypes or Fam~lies of Plastics Additive Agents in Plastics Oriented Plastic? Ensneering Plastics Plastics as Adhesives Plastics forToollng Foamed Plast~cs Polymer Coatlngs Plastlcs versus Other Materials Recycling of Plastlcs

Rubber Artific~al Elastomers Selection of an Ehstomer Elastomers for Tooling Applications 8.4 CERAMICS Nature and Structure of Ceramcs Ceramics Are Brlttle but Can Be Tough Clay and Whiteware Products RefraLtory Materials Abrasives Ceramics for Electrical and Magnet~c Applrcatlons Glasses Glass Ceramics Cermet.; Cements

Ceram~c Coatings Ceramics for Mechanical App1ications:The Structural and Advanced Ceramics Advanced Ceramics as Cutting Tools 8.5 COMPOSITE MATERIALS Laminar or Layered Composites Particulate Composites F~ber-ReinforcedComposites Advanced Fiber-Reinforced Composites Hybrid Composites Desrgn and Fabrication Assets and Llmitatlons Areas of Application Case Study: TWO-WHEEL DOLLY HANDLES

lls" 8.1 INTRODUCTION

requirements of products contmue to push the lim~ts of traditional materials, the rol the manufactured nonmetallic materials will no doubt continue to expand.

~f they may be reasonable candidates for specific products and applications. For det

information about specific materials within these fam~lies, more extensive and dedicat texts, handbooks, and compilations should be consulted.

SECTION 8.2 Plastics

163

It is difficult to provide a precise defimtion of the term plastics From a technical viewpoint, the term is appl~ed to engineered materials characterized by large molecules that are built up by the joining of smaller molecules. On a more practical level. these materials are natural or syntheticresins, or their compounds, that can be molded, extruded, cast, or used as thin films or coatings. They offer low density, low tooling costs, good resistance to corrosion and chemicals, cost reduction, and deslgn versatihty. From a chemical vlewpomt, most are organic substances contaming hydrogen, oxygen, vrbon, and nitrogen. In less than a century,we have gone from a world without plastic to a world where its use and applications are 1imitless.The United States currently produces more plastlc than steel, alumnurn, and copper combined. Plastics are used to save lives in applications such as artificial organs, shatter-proof glass, and hullet-proof vests. They reduce the weight of car\ provide thermal insulation to our homes, and encapsulate our medicines. They form the base material in products as diverse as shower curtans, contact lenses, and clothing, and compose some of the primary components in televisions, computers, cell phones, and furniture Even the Statue of L~berty has a plastic coatlng to protect it from
corrosion.

n-b-L

-r7

Ethane
co-I
irrc~~~lnr~!yut

, a n d hydrogen to form _. _le and ethane molecules, r h dash represents a shared ectron pair or covalent bond.
1
H

Acetylene Double and tr~pl onds exlst between 1 atoms in unsaturate nd acetvlene
-2

MOLECULAR STRUCTURE OF PLASTICS To understand the propertles of plasbcs, it is important to first understand their molecular structure. For simplicity, let'^ begin with the paraffin-type hydrocarbons, in whlch the atoms can carbon and hydrogen combine in the relationship C,H2,+2.Theorefically, Link together mdefinitely to form very large molecules, extending the serles depicted in Figure 8-1. The bonds between the various atoms are all pairs of shared electrons (covalent bonds). Bondmg w~thinthe molecule, therefore, is quite strong, but the attractive forces between adjacent molecules are much weaker. Because there 1s no provision for additional atoms to be added to the chain, these molecules are said to be satmtgd. Carbon and hydrogen can also form molecules where the carbon atoms are held together by double or triple covalent bonds. Ethylene and acetylene are common examples (Figure 8-21, Because these molecules do not have the maximum number of hydrogen atoms, they are said to be unsaturated and are important in the polymerzation process, where small molecules link to form large ones with the same constituent atoms. In all of the described molecules, four electron pairs surround each carbon atom and one electron pair is shared with each hydrogen atom. Other atoms or structures can be subst~tuted for carbon and hydrogen. Chlorine, fluorine, or even a benzene ring can take the place of hydrogen. Oxygen, sllicon,sulfur, or nitrogen can take the place of car. bon. Because of these substitutions, a wide range of organic compounds can be created. ISOMERS The samekind and number of atoms can also unite in d~fferent stmctnral arrangements, known as isomers, and these ultimately behave as different compounds with different engineering propertles Figure 8-3 shows an example of this feature, involving propyl and lsopropyl alcohol. Isomers can be considered analogous to allotropism or polymorphism m crystalline materials, where the same material possesses different properties because of ditterent crystal structures.
H H H

H-C-C-C-O-H

H-C-C-C-H

I l

FIGURE 8-3 Linking of eight

H
Propyl Alcohol

lsopropyl Alcohol

hydrogen, one oxygen, and three carbon atoms to form two nrfin,,l 91rnhfi1 lsopropyialiohol. Note the different locations of the -OH attachment.
; C ~ ~ O ~ C .

CHAPIZ~ 8 Nonrnetall~c Materials: Plastics, Elastomers, Cerami~s, and tomposctes

FIGURE 8-4 Addit~on polymefizalion-the linking of H CI H C I monomers; in thls case, ~dent~cal 4- 4ethylene molecules. Monomer Monomer

I I c=c I I

I I c=c I I

Mer

I I I ----c-c-c-c--I I I
H

CI

CI

I I

FORMING MOLECULES BY POLYMERIZATION The polymerization process, or linking of molecules, occurs or cande7wufzonmechanism. E~gure 8-4. illustrates polymer~zati number of basic units (monomers)link together to Corm a large w k h there is a repeated unit (mar).Acirva%ors or caTalysts, such as b &ate andterminate the ohain.Thus,the amount of activator relative monomer determines the average molecular weight (or average length chaii.The average number of mers in the polymer, known as fh tion,ranges from 75 to 750 for most commercial plastics. Chain length controls man the properties of a plastic. Increasing the chain length tends to increase taughness, c resistance, melting temperature, melt viscosity, and difficulty 1n processing. Capolymers are a specla1 category of polymer where two different types of are utmbined into the same addition chain. The format~on of capo analogous to alloys in mezals, greatly expands the possibilit~es of plastics with lmproved phy9ical and mechanical prop~rties. Terpo the pomibilitim by combining three different monomers.

C-c=c-C-C-C-

T'/'/'/TT
I
I I
I

I I I I I c-c=c-c-c-c--I I I

I
I

AdditLon polymerization with two kinds of rner+era, the copolymefization of bntadiene and styrene.
FIGURE 8-5

"

H
<

Butadlene mar

Styrene mer

In cmtrdst t o polymetization b b y addition, where all of the original atoms appear in the product molecule, condemtion poiyt?ter~zation occurs when reacttve molecules combine w ~ t h o n e another to produce a polymer plus small, by-product molecules, such as water. Heat, pressure, and catalysts are often required to drrve the reacbon Figure 8-6 iIlustrates the reaction between phenol and formaldehyde to form Bakel~te, first performed in 1910.The structure of condensation polymers can be either h e a r chains or a three-dimensional hameworkin which all atoms are linked by $trong,ptimary bonds

H
a -

Phenol Phenol Phenol Formaldehyde FIGURE $6 The formstion of phenoj-formaldehyde (Bakellw) bbycandensat~on polymer~zat~on. Note the H1Oor water by-product.

Formaldehyde

H W

THERMOSETTING AND THERMOPLASTIC MATERIALS The terms thermosetting and thermoplastic refer to the material's response to elevated temperature. Addition polymers (or linear condensation polymers) can be viewed as long chains of bondedcarbon atoms with attached pendants of hydrogen,fluorine,chlorine, or benzene rings. All of the bonds within the molecules are strong covalent bonds. The attraction between neighboring molecules is through the much weaker van der Waals forces. For these materials, the intermolecular forces strongly influence the mechanical and physical properties. In general, the linear polymers tend to he flexible and tough. Because the intermolecular bonds are weakened by elevated temperature. plastics of this type soften with increasing temperature and the individual molecules can slide over each other in a molding process. When the material is cooled, it becomes harder and stronger.The softening and hardening of these thermoplastic or heat-softening materials can be repeated as often as desired, and no chemical change is involved. Because thermoplastic materials contain molecules of different lengths, they do not have a definite melting temperature but, instead, soften over a range of temperatures. Above the temperature required for melting, the material can be poured and cast, or formed by injection molding. When cooled to a temperature where it is fully solid, the material can retain its amorphous structure, but with companion properties that are somewhat rubbery. The application of a force produces both elastic and plastic deformation. Large amounts of permanent deformation are available and make this range attractive for molding and extrusion. At still lower temperatures. the bonds become stronger and the polymer is stiffer and somewhat leathery. Many commercial polymers, such as polyethylene. have useful strength in this condition. When further cooled below the glass transition temperature, however, the linear polymer retains its amorphous structure but becomes hard, brittle, and glasslike. Many thermoplastics can partially crysmllize' when cooled below the melting temperature. This should not be confused with the crystal structures discussed previously in this text. When polymers "crystallize," the chains closely align over appreciable distances, with a companion increase in density. In addition, the polymer becomes stiffer, harder, less ductile, and more resistant to solvents and heat.The ability of a polymer to crystallize depends on the complexity of its molecules, the degree of polymerization (length of the chains), the cooling rate, and the amount of deformation during cooling. The mechanical behavior of an amorphous (noncrystallized) thermoplastic polymer can be modeled by a common cotton ball. The individual molecules are bonded within by strong covalent bonds and are analogous to the individual fibers of cotton.The bonding forces between molecules are much weaker and are similar to the Criction forces between the strands of cotton. When pulled or stretched. plastic deformation occurs by slippage between adjacent fibers or molecular chains. Methods to increase the strength of thermoplastics, therefore, focus on restricting intermolecular slippage. Longer chains have less freedom of movement and are therefore stronger. Connecting adjacent chains to one another with primary bond cross-links, as with the sulfur links when vulcanizing rubber, can also impede deformation. Since the strength oI the secondary bonds is inversely related to thc separation distance between the molecules, processes such as deformation or crystallization can be used to produce a tight parallel alignment of adjacent molecules and a concurrent increase in strength, stilfness, and density. Polymers with larger side structures, such as chlorine atoms or benzene rings, may be stronger or weaker than those with just hydrogen,depending on whether the dominant effect is the impediment to slippage or the increased separation distance. Branched polymers, where the chains divide in a Y with primary bonds linking all segments of the chain, are often weaker since branching reduces the density and close packing of the chains. Physical, mechanical, and electrical properties all vary with the above changes in structure.

It should bc nored that the term cryslaiiize, when applied to polymers. has a diilerent lueaning than when applied lo mevals and ceramics. Metals and ceramics are crystalline materials, meaning that the atoms occupy sires in a regular, periodic array, known as a lattice. In polymers, it is not the atoms that become aligncd, but the mulecules. Since van der Waals bonding has a band slrellgth that is inversely related to the separation distance, the parallzl alignmcnr of the crystallized stale is a lower-energy configuration and is llromoled by slow cooling and equilibriun~-type processing conditions.

'

- .

I , '

: ,

1.

~,

.r

Nonmetall~c Materials: Plastics, Elastomers, Ceramics, and Composrtes The four most common thermoplastic polymers are: polyethylene (PE), polyprop lene (PP), polystyrene (PS), and polyvinyl chloride (PVC). In contrast to the thermoplasticpolymers thermoscttingpplastrcsusually have a h~g cross-linked or three-dimens~onal framework structure in whch all atoms are connect by strong, covalent bonds These materials are generally produced by condensationpo merization where elevated temperature promotes an irreversible reactlon, hence term fhermosettang. Once set, subsequent heating will not produce the softening obs \Kith the thermoplastics Instead, tbermosettingmater~als maintain thnr mechanical erties up to the temperature at which they char or burn. Since deformanon requires t breaking of primary bonds, the thermosetting polymers are significantly stronger more rigid than the thermoplastics.They canreslst higher temperatures and have gre drmensioual stabihty,but they also have lower ductihty and poorer Impact properhes. As a helpful analogy, thermoplast~c polymers are a lot like candle wax. They be softened or melted by heat, and then cooled to assume a solid shape.Thermosets more like egg wh~tes or bread dough. Heating changes their structure and properties an irreversible fashlon Although classification of a polymer as thermosetting or thermoplast~c prov' inslght as to properties and performance, it also has a strong effect on fabricat~on. example, thermoplastics can be easily molded. After the hot, soft material has b formed to the desired shape, however, the mold must be cooled so that the plastic harden and be able to retam its shape upon removal.Tl~erepetltlve heating and co ing cycles affect mold lif&,and the time required for the thermal cycles infl productivity.When a part is produced from thermosetting materials,the mold can at a constant temperature throughout the entlre process, hut the sethng or curing o resins now determines the t m e in the mold. Slnce the material hardens as a resul reactlon and has strength and rigidity even when hot, product removal can be perf without coolmg the mold.

PROPERTIES AND APPLICATIONS Because there are so many varieties of plastics and new ones are bang developed alm continuously, it is helpful to have knowledge of both the general properties of plas and the unlque or spec~fic properties of the various famihes. General properties plastics include:

1. L~ght weight. Most plasticshave specific gravities between 1.1and 1.6, compared about 1.75 for magnesium (the lightest engineering metal). 2. Corrosion re~istance. Many plastics perform well in hostlle, corrosive, or chenu environments. Some are notably resistant to a c ~ d corrosion. 3. Electrical resistance. Plastics are widely used as insulatingmaterials. 4. Low thermal conduetlv~ty. Plastlcs are relatively good thermal insulators. 5. Varrety of opt~calproperties. Many plastics have an almost unlimited color range, the color goes throughout, not just on the surface. Both transparent and opa materials are available. 6. Formability or rase of fabrrcatlon. Objects can frequently be produced from plas m a single operation. Raw materlal can be converted to final shape through s processes as castmg, extrusion, and molding. Relatively low temperatures required for the forming of plastics. 7. Surface finish.The same processes that produce the shape also produce excell surface finish. Additional surface finrshing may not be required. 8. Comparatzvely low cost.The low cost of plastics generally applies to both the mat ial itself and the manufacturing process. Plastics frequently offer reduced tool co and high rates of produchon. 9. Low energy content.
While the attractive features of plastics tend to be in the area of physi properties, the inferior features generally relate to mechanical strength. Plastics can flexible or rigid, but none of the plastics possess strength propertres that approach th

of the engineering metals unless they are reinforced in the form of a composite.Their low density allows them to compete effectively on a strength-to-weight (or specific strength) basis, however. Many have low impact strength, although several (such as ABS, high-density polyethylene, and polycarbonate) are exceptions to this rule. Aluminum is nearly 10 times more rigid than a high-rigidity plastic, and steel is 30 times more rigid. The dimensional stability of plastics tends to be greatly inferior to that of metals, and the coefficient of thermal expansion is rather high.Tnermoplastics are quite sensitive to heat, and their strength often drops rapidly as temperatures increase above normal environmental conditions.Thermosettingmaterials offer good strength retention at elevated temperature but have an upper Limit of about 2S0C (SOOT). Low-temperature properties are generally i n f e r i ~ to r those of other materials. While the corrosion resistance of plastics is generally good, they often absorb moisture, and this, in turn, decreases strength. Some thermoplastics can exhibit a 50% drop in tensile strength as the humidity increases from 0to 1M%. Radiation, both ultraviolet and particulate, can markedly alter the properties. Many plastics used in an outdoor environment have ultimately failed due to the cumulative effect of ultraviolet radiation. Plastics are also difficult to repair if broken. Table 8-1 summarizes the properties of a number of common plastics. By considering the information in this table along with the preceding discussion of general properties, it becomes apparent that plastics are best used in applications that require materials with low to moderate strength, light weight, low electrical andlor thermal conductivity,a wide range of available colors, and ease of fabrication into finished products. No other family of materials can offer this combination of properties. Because of their light weight, attractive appearance, and ease of fabrication, plastics have been selected for many packaging and container applications.This classification includes such items as household appliance housings, clock cases, and exteriors of electronic products, where the primary role is to contain the interior mechanisms. Applications such as insulation on electrical wires and handles for hot articles capitalize on the low electrical and thermal conductivities. Soft, pliable, foamed plastics are used extensively as cushioning material. Rigid foams are used inside sheet metal structures to provide compressive strength. Nylon has been used for gears, acrylic for lenses, and polycarbonate for safety helmets and unbreakable windows. There are many applications where only one or two of the properties of plastics are sufficient to justify their use. When special characteristics are desired that are not normally found in the commercial plastics, composite materials can often be designed that use a polymeric matrix. For example, high directional strength may be achieved by incorporating a fabric or fiber reinforcement within a plastic resin.These materials will be discussed in some detail later in this chapter.

C O M M O N TYPES OR FAMILIES OF PLASTICS The following is a brief descriptive summary of the types of plastlcs l~sted inTable 8-1.
THERMOPLASTICS ABS. contains acrylonitnle, butadlene, and styrene: low werght,good strength, and very tough; resists heat, weather, and chemicals quite well; dimensionally stable but flammable Acryl~cs. highest optical clarity, transmitting over 90% of light, common trade names include Luc~te and Plex~glas; h~gh-~mpact, flexural, tensile, and dielectric strengths; available in a wide range of colors:res~stweathering Cellulo~e acetate:wide range of colors; good insulating qualities; easlly molded; high moisture absorption in most grades Cellulose acefare butyrate. higher impact mength and moisture resistance than cellulose acetate; will withstand rougher usage Ethylcellulose: high electrical resistance and impact strength: retains toughness at low temperatures

Special Characteristicsa

Common Forms

Mater~al
Tnermoplastics ABS material
Auctal

Specific Gravity
1.02-1.06 1.4 1.12-1.19 1.25-1.50 1.18-1.24 1.19-1.24 1.4 1.16 2.1-2.3 213.15 1.1-1.2 1.2

Tensile Strength (la00 l b i i 2,

Impact Strength Izod (ft-lbhn. of Notch)

TOP Working Temperature I"Frc)l

0
X
X

x
X

a , . .
0
X X

Acrylics Cellulose acetale Cellulose acetate butyrate Cellnlosz propionate Chloriaated polyether Ethyl cellulose TFE-fluorocarbon CFE-tluorocarban Nylon Palycarbonate Polyethylene Polypropylene Polystyrene Modified palystyreoe Vinyl Thermosetting plastics EPOXY Melamine Phenolic Polyester (othu than molding compounds) Polyester (alkyd, DAP) Silicone Urea

x o . . .
X

...
.

X X
X X

X X

. . ....
. . .

O
X

O
X
X

O
X

.
.

.
.

.
.
.

.
.
.

0.96
0.9-1.27 1.05-1.15 1.0-1.1 1.161.55

0 O
O

X
X

.
. .

X
X
X

O
X

.
.

.
.

.
. .

1.1-1.7 1.76-1.98 1.2-1.45

1.0M.46 1.6-1.75 2.0 1.41-1.80

Flwrocarbum: inert to most chemicals; high temperature resistance; very low coefficients of friction (Teflon);used for nonlubricated bearings and nonstick coatings for cooking uteusils and electrical irons Nylon ~ o ~ m ~ e s s ) : d lo ~w c i e nof t friction;good strength, abrasion resistmce,and baughaass:e d h t dimensional stability; good heat resistance;used for small @am and bearings, zip fasteners, and as monofi1mentsfor textiles, fishing line, and mpes Polycarbonates: hsgh strength and outstanding toughness; good dimensional stability; transparenLor easily colored Polyethylenes: the most common polymer; inexpensive, tough*good chemical resistance to acids, bas%, and salts; high electrical iesisbnce; low strength; emy to shape and join; reasonably clear in thin-film form; subject to weathering vla ultraviolet light;flammable; usedlor gocery bags, milk jug and other food containers, tubes, plpes, sheeting, and electrical wire insulation. Variations include: low-density polyethylene (LDPE-floats in water), high-den~ity polyethylene (HDPE),ulba-high-molecular-weight polyethylene (UHMW) PMMA (pglymethyl methacrylate). had, brittle [&troam temperature),transparent or easdy wlored; used for items hke tool handles and the interior wmdows of airpIanes Polypropylene: inexpensive; stronger, stiffer, and better heat resistance than polyethylefle; transparent; reasonable toughness; used for beverage containers, luggage, pipes, and ropes Polysoren~ high dimensional stabdiiy and stiffnesswith low water absorption;best all-around dielectric; clear, hard, and bnttle at room temperature; often used for rigid packaging;can be foamed to produce expanded poIystyrene (tradename of Styrofoam),burns readily: softens at about @PC Polyvinyl chloride (PVC): general-purpose thermoplastic; good resistance to ultraviolet lrght (good for outside applications); easily molded or extruded;always used with fillers, pldsticizers, and pigments; uses include gas and water pipes as well as window frames Wiflyls.wide range of types, from thin, rubbery films to r~gid forms; tear resistant good aging properties; good dimensional stability and water resistance in rigid forms; used for floor and wall covering, upholstery fabrics,and I'ihYweigbt water hose; common trade names include Saran a n d m o n THERMOSETS Epoxies:good strength,toughness, elasticity, chemical resistance, moistureresistaace, and dimensionalstabi1ity;easily compounded to cure at room temperature;used as adhesives,bonding agents, coating%and m fib& laminates M@lmznes:excellent resistance to heat, water, and many chemicals; full range of translucent and opaque colors: excellent electric arc resistance; tabJeware (but stained by ooffee}; used extensively in treating paper and cloth to impart water-repellent properties Phemlies: ol$est of the-plasria bur still widely used; hard, strong, low cost, and e-asily molded, but rather brittle; resistant to heat and moisture; dimendondly stable; opaque, but pith a wide cohr range; w~de variety of foims: sheet, rod, tube, and laminate; trade names include Bakelite. Polyesters (can be thermoplastic o r : thermoset): strong and good resistance ta environmental influences;uses include boat and carbodieg pipes, . vents and duct$ textiles, adhesives, coatings, and laminates S~lzcones: heat and weather resistant; low maisture absorpion; chemically inert; high diel~ctris properties; excellent sealants Urea-formaldehyde: properties similar l o those of phenolics but available in lighter wlors;usefnl in wntainersand housin~r, but not outdo6rs;used in lighting fixturesbecause of translucencein thin seotiou6;as a foam,may beused as household insu~ation

11

I1

170

~c Plast~cs, Elastomers, Cerarn~cs, and Composites C H A ~ Z R BNonrneti~ l l Materials:

ADDITIVE AGENTS IN PLASTICS For most use* addihonal mater~als are incorporated into plastics to (1)Impart or improv properties, (2) reduce cost, (3) improve moldability, andlor (4) impart color Thes addifrve coustrtuenty are usually classified as fillers and reinforcements,plust~cizers lubricants, colorcng agentr,stabllizers, anfioxrdunts, and pame retardants. Ordinarily,fillers comprlse a large percentage of the total volume of a molded plas hc product. The~r primary roles are to improve strength, stiffness, or toughness; redu shrinkage; reduce weight, or simply serve as an extender, providing cost-saving bu (often at the expense of reduced moldabdity).To a large degree, they determine the ge era1 properties of a molded plastic. Selection tends to favor materials that are much expensive than the plastic resin. Some of the most common fillers and the11 properties

t Woodflour (fine sawdust): a general-purpose filler, low cost with fair strength, goo
2. 3. 4. 5. 6. moldabil~ty Cloth fsbers: improved impact strength, fair moldability Macerated cloth. high impact strength, limited moldability Class fibers: high strength; dlmensioual stability; translucence Mzca: excellent electrical prope&es and low moisture absorption Calcrum carbonate, sskca, talc, and clay serve prmanly as extenders

When fillers are used wlth a plastic resin, the resin acts as a binder, surroundin filler material and holding the mass together.The surface of a molded part, ther will be almost pure resin with no exposed filler. Cutting or scratching through the surface will expose the less attractive filler Coloring agents may be either dye^, which are soluble in the resins, or insolu pzgments, which impart color simply by their presence. In general, dyes are used transparent plastics and pigments for the opaque one5 Optical brighteners c also be used to enhance appearance. Carbon black can provide both a black color electrical conductivity Plarticuers can be added m small amounts to reduce viscosity and improve flow of the plastic during molding or to increase the flexibility of thermoplastic p by reduclng theintermolecular contact and strength of the secondary bonds betw polymer chains. When used for molding purposes, the amount of plast~c~zer is govern by the mtrlcacy of the mold. In general, it should be kept to a minimum because i likely to affect the stability of the finished product through a gradual aging loss Wh used for flexibility,plasticizersshould be selected with minimum volatihty, so as to imp the desired property for as long as possible Lubricants such as waxes, stearates, and soaps can be added to improve t moldability of plastics and to facilitate removal of parts from the mold.They are also to keep thin polymer sheets from sticking to each other when stacked or rolled. On1 minimum amount should be used, however, because the lubricants adversely affeckm engineering properties. Heat, light (especially ultraviolet), and oxidation tend to degrade polym Stab~lizers and antlox~dants can be added to retard these effects. Flame retardants be added when nonflammahility is important. Antistatic agents allow for the migra of electrical charge and may be lncorporated into plastics used for applications su electronics packaging. Antirmcrobial additives can provide long-term protection both fungus (such as mildew) and bacteria. Fibers can be lncorporated to mcre strength and stiffness, and metal flakes, fibers, or powders can mod~fy electrical magnetic properties.Table 8-2 summarizes the purposes of the various additives.

ORIENTED PLASTICS Because the intermolecular bond strength increases with reduced separahon dis any processing that aligns the molecules parallel to the applied load can be used t the long-chain thermoplastics high strength in a given diiect~on.~Thls orientation pro
2Theeffects of orienting can be observed in the common disposable th~n-walled plastic drinkmg cup St the top lip Place a sharp bendin the lip and then tear down Iht s ~ d wall e The matenal tears easlly Movear the l ~ p a b u11slnch t and make dnotlrer side-wall tear-also easy Now try to tear across thestnp that you created.Ths tear IS much more d~ff~cult,s~nce you are tearrng across molecules that have been ortented tically along the cup walls by the cup-forming opentlon.

Type
hlks

Purpose

Enhance mechanical propenlesreducr shrinkage, reduce we~ght, or provrde bulk Plashclzer Increase fl*X?rbihty, unprove llow durrng moldmg, reduce elastlc tdodulus Lubricant Imprwe rnoldah~luy and emacrlon from molds Colonng agents (dyes and pigments) Impart color Stabilizers Retard degradation due to heat or Il&t Antioxidants Retard degtadatron due to oxtdation Flame retardants Reduce flammablllty

can be accomplished by a forming procss, such as stretching,mlling, or extrusion.The material is usually heated prior to the orienting process to aid in overwming theintermolecular forces and is cooled immediately afterward to "freeze"the molecules in the desired orientation. Orienting may increase the tensile strength by more fhan 50-76,bur a 25% lncrease is more typica1.U addition, the elongation may be increased by several hundred percent. If the oriented plastics are reheated, they tend to deform back toward their original shape, a phenomenon known as vwroe6am~c meinory.The various shrink-wrap materials are examples of this effect.

ENGINEERING PLASTICS
The standard polymers tend to be lightweight, corrosion-resistant materials with low strength and low stiffness.They are relatively inexpensive and are readily formed into a wide range of useful shap~s, but they are not suitable for use at elevated temperatures. I n conrrmr. :I group of pl~citics has been dc\,~.I~ped with irnpro\,cd thermal proprrties (up t c ~?51J0(:, or 65U"T;). cnh:rnc<d i~npi~cl :nid Strw resi~rance, Iligll ripidit\,. . . superior electrical characterist~cs,excellent pr&ess~ngproperties, and little dimensional change With varying temperature and humiditpThese true englneenng plastics include the polyamides, poIyacetal$,palyacrylates, polycarbonates. modified polyphenylene oxides, polybutylene terephtalates, polyketanes,polysulfones, polyetherimides,and liquid crystal polymers. Whiie stabilizers,fibrous reinforcements, and particulate fillers can upgrade the convent~onal plastics, there is usually an accompanyingreduchon in other properties.The engineeringplastics offer a more balanced set of propertiesl'hey are mually ptooduwd in small quantities, however, and are often quite expenswe. Materialsproducers have also developed electroconductivepolymers with tailored electrical and electronic properties and high-crystallinepolymers with properties comparable to some metals.

PLASTICS AS ADHESIVES
Polymeric adhesives are used in many industrial applications.They are quite attractive for the bonding of diss~milar materials, such as metals to nonmetals, and have even been used to replace welding or ri7eting.A wide range of mechanical propexties are available through variations ~1composition and additives, and a vanety of curing mechanisms can be used. Exmnples can be found from the thermoplastics (hot-melt glues), thermosets (two-part epoxies), and even elastomers (silicone adhesives). The seven most commonsmcturaladhesives are epoxies,urethanes, cyauoacrylates,acrylics,anaerobics, hot melts, and silicones, Selection usually involves consideration of the manufacturing conditions,thesubstrates to he bonded, the end-me environment, and cost.% yariom features of adhesive bondimg are discussed in greater derail in Chapter 35.

PLASTICS FOR TOOLING

Polymers can also provide inexpensive tooling for applications where pressures, trmperatura, and wear requirements are not extreme. Because of their wide range of properties, their ease of canversion into desired shapes, and their excellent properties

172

C U A ~8R Nonmetallic Mate~alsPlastics, Etastorners, Ceramics, and Cornposrtes

when loaded in ~ p r e e s i o hplasm , have been widely used in applicationssuch as ji fixtures, and a wide vatiety of forming-die components. Both thermoplastic and thermoset polymers (paiticularly the cold-8etting types) have been used. By usingplastics in these application3, costs can be reduced and smaller quantities of products can be economicallyjmtified. In addition, the toolingcan oftea be produced in a much sh~rt er time, enabling quicker production.

FOAMED PLASTICS A nnmbes of polymeric materia& can be produced in the form of foams that incorp rate arrays of gaawus voids in their structure. These materrals are extremely ve*sati withproperties r m e g f r o m soft and flexible to hard and rigid.?he softer foams are erally used for cushioning in upholstery ivld automobiIeseats, and m various applica such as vibration absorbers. Semirigid foams frnd use in floatation devicas,~efrigerata mnsulation, disposable food frays and containets, building insulation panels, and sou attenuation. Rigid foams have been used as construction marerials for boats, airpla wmporrente,electronic encapsulation,and furniture. Foamed materials can be made by a wide variety of p p r m e s , they can elthe made as discrete products or used as a "foamed-in-place" material. In addition to sound-and vibration-attenuation properties mentioned above, foams offer light w and the possibility of improved stiffhessand reduced cost (less material to make the POLYMER COATINGS Polymer coatings are used exken8ively to enhance appearance, but they have assumed a significant role in providing emosion pratecbon. The tough, thick coa must adhere to the oubtrate; not chip or peel; and resfst exposure to heat, mois salt, and chemicals. Polymer coatings have been replacing chrome and cadmium to environmental concerm relating to the heavy metals. In addition, polymers pro better resistance to the effects of acid rain. PLASTICS VERSUS OTHER MATERIALS PolymericmatEtriak have suffiessfuLlycompeted with traditionalmaterials in a numbe areas. Plastics have replaced glass in containers and ather transparent prod1 PVC pipe and fitfigs compstewith copper and brass in many plumbing applimtions.P tics have even replaced ceramic8 in wea8 as diverse as sewer pipe and lavatory faalitr~ While plastics and metals are &en viewed as compebnp materials, their el neeriegprslperties are really quite diffarent. Many of the attractive features of p t ~ c have s already bean discussed. In addition to these, we can add (1)the ability t c fabricated With lower fooling costs; (2) the ability to be molded at the same mti product assemBly, thereby reducing inventory; (3) a possible reduczion in assem operations and easrer assembly through snap fits, frictioli wd&, o r the use of seIf-l ping fasteners; (4) the ability to Feuse manufacturing scrap; and (5) reduced fin ing costs. Metals, on the other hand,are often cheaper a n d offer faster fabricationspeeds greaterimpaa resistance. They me considerably stronger and more rigid and can w stand traditional paint cure temperatures. Zn addition, resistance to flames, acids, various solvents igsignificantly better. Table wmpares the mechanical propemes of selec epolymers to annealed, commercklly pure ; minum and annealed 1040 steel. Note the I of the TS (ksl) E ( 1psi) ~ E I cbic ~a l superiority ~ ~ metals, ~ even tho ~ they are being presented in their we&est c a ozs 90-650 2 dition. Table 8-4 compares the cost per pol 4 fl.lOQ S M o a and elasficmodulus of severalengineering
a375 0 51m
10

Matenal
blyrthylene Pdycthylen& Palyvmylchlonde Pwly$tyrene Bake1118

Conditian
Branched
Crystallized
C1-sde?

ALummum lM0 steel

Benzene-sides Framework Anne;tle*l

240 1 3
1

7 13
75

1U.U
3a a

Ann%&d

15-30 30

t i c s size witbof values his &fixed, e l and aluminum.4 the the part cost per cubic becomes a more valid comparison, and fh ares S ~ Q W plastics to be quite competiuive cause of their low density.

SEC~IQN 8.3 Elastomers

Bjigs,

173

, therModulus [x la6psi) $/pound

0.3M 40

ed"+" -_
- S

~stics s can 1hort-

$/m
0 075-0.10
0.129 0 034

03

I 00

orposatile, e genations pator sound plane

0.030
0 097

The automotive industry is a goodindicat~on o f the expanding use of plastics Polymericmaterials now account for over 250 pounds of a typical vehicle, compared to only 25 pounds in 1960,105 in 1970,195in 4980, and 229 m 1990. In add~tion to the traditional application areas of dashboardslnteriors body panels, and trim, plastics are now being used for bumpers mtake manifolds, vaLve cover$ fuel tanks and fuel lines and fittings. If we include clips and fasteners,there are now over 1000plastic parts m a typical autonrobile.

h a be
to the reight :part). e also ~atings isture, m due rovide

nber of oducts. ts.Plasilities. r eugid plas,y to be rate as sembly eif-tap'finisheds and

Because of the w~de variety of types and compositions all w ~ t h similar phys~calpropertleq the recycling of &cost figurri a+o2006 vnluw nnA are clearly sublea ia d~ange m e d plastics is far more difficult than the recycling of &xed metalsThese mater~als must be sorted not only on the basis of resin type, but also by type of filler and color. If the various typesof resinscan beidentified and keptseparate,many of the thersthe largest moplasticmaterials can ha readily recycled into useful products. Packaging 1 single market fotplastics, and thereik currently a well-established network tomllect and polyethylene recycle PET (the polyester used rn soft-drink bottles) and high-dens~ty (the plasticused in milk, juice,and water jugs).The properties generally deteriorate with recycling, however, so applications must often be downgraded with reuse. PET is being recycled into new bottles, fiber-fill insulation, and caspeting. Recycled polyethylene is used for new containers, plastic bags, and recycling bms. Polystyrene has been recycled into cafeteria trays and videocassette cases. PSasfic"lumber"offc~~ weather and insect resistanceand a reduction in required maintenance (but at hlgher cost than traditional wood). When thermoplastics and thermosets are mixed in varying amounts, the material is often regarded moreas an alternative fuel (competing with coal and oil) than as a resource for recycling into gualrty products. On an equivalent-weightbasis, polystyrene and polyethylene have heat contents greater than fuel oil and far in excess of paper and w ~ o dAs . a recycling alternative,decompo&hon processes can be used to break polymers down into useful building blocks. Hydrolysis (exposure to high-pressure steam) and pyrolysis (heating m the absence of oxygen) methods can be used to convert plastics Into simpte perrochemica1 materials,but even these processes require some control of ?he Input material. As a result, only about one-third of all plast~c now finds a second life.

0.025 0 027 0 036

RECYCLING OF PLASTICS

in withids, and able 8-3 rlected be d u che methough a t mnpound mg plasn. When ~bic jnch I the figitive be-

a contraction of the words elrrsticpulymer, refers to a special ebss The term ela~'&ine$ of linear polymers that display an exceptionally large amount of elastic deformation when a f o ~ c e is applied. Many can be stretched to s e ~ e r atimes l their original length. Upos release of the force, the deformation can be completely recoveted as the material qu~cklyseturns to its orig~nalshape. In addition, the cycle can be repeated numerous times with identicas results, as with the stretching of a rubber band. The elastic properties of most engineering materials are the result of a change m the distance between adjacent atoms (i.e, bondlength) when loads are applied. Hooke's Iaw is commonly obeyed, where twice the force produces twice the stretch. When the applied load is removed, the interatomic forces return all of the atoms to their original posit~on and the elastic deformation is recovered completely. In the elastomesic polymers, the Smear chain-type molecules are twisted or curled, much like a coil spring. When a Fmce 1s applied, the polymer stretches by uncoilir@! When the load is removed, the molecules recoil as the hand angles return to their? original, unloaded value%and the material returns to its original size and shape. relatio~iship between force and stretch, however, does not follow Hooke's law. s a bit mare complex. While the chams inde& In reality, the behavior of elastomers i uncoil when placed under load, they can also slide with resped to one an0 to produce a small degree of viscous deformation. When the load is re

174

CHAPTER 8 Nonmetalllc Materials: Plast~cs, Elastomers, Ceramlcs, and Composites

molecules return to their coiled shape, but the viscous deformation is not recovered and there is some permanent change in shape. By hnking the coiled molecules to one another by strong covalent bonds, a process it is possible to restrict the viscous deformation while retaining known as cross-lznk~ng, the large elastic response. The elasticity os rigidity of the product can be determined by controlling the number of cross-links. Small amounts of cross-linking leave the elastomer soft and flexible, as in a rubber band. Additional cross-linking further restricts the uncoiling, and the material becomes harder, stiffer, and more brittle, l ~ k the e rubber used in bowling balls. Since the cross-linked bonds can only be destroyed by extremely high temperatures, the engineering elastomers can be tailored to possess a wide range of stable properhes and stress-strain characterist~cs If placed under constant stram, however, even highly cross-linked material will nails. While the dimensions remain fixed, the force or stress bemg apphed to the nail will continually decrease. This phenomenon 1s known as stress relaxahon. The rate of this relaxat~on depends on the material, the force, and the temperature.

RUBBER

tioxidants. Accelerators have been found that speed up the vulcanization process.The have enabled a reduction in the amount of sulfur, such that most rubber compoun now contain less than 3% sulfur. Softeners can be added to facilitate processlug, an Fillers can be used to add bulk. Rubber can now be compounded to provide a wide range of characteristi

Vehicle tires and heavy-duty conveyor belts are examples of this technology.

that they not be used at temperatures above 80C (175F). Unless they are specia compounded, they also deteriorate fairly rapldly m direct sunlight.

ARTIFICIAL ELASTOMERS In an attempt to overcome some of these limitations, as well as the uncertainty in

Styrene-butadiene is an oil-derivative, high-volume substitute for natural rubber th

applications, including automotive hoses and belts, footwear, tires, mounting cusho and seals. A number of other artificial elastomers are ava~lable and are idenbfied both chemcal and commercial trade names

Silicone rubbers look and feel like organic rubber but arc based on a linear chain of silicon and oxygen atoms (not carbon). Various mixes and blends offer retention (450F)];flexibilityat low of physicalproperties at ekvated temperatures [as hot as 230UC temperatures [as low as - 100C (150DF)]; resistance to acids. bases,and other aqueous and organic fluids; resistance to flex fatigue; ability to absorh energy and provide damping; good weatherability; ozone resistance; and availability in a variety of different hardnesses. Elastomers can also be classified as thermosets or thermoplastics. The thermoset materials are formed during the irreversible vulcanization (cross-linking) process, which may bc somewhat time consuming. Thermoplastic elastomcrs eliminate thc vulcanization cycle and can be processed into products by a11 of the conventional thermoplastic polymer processes (injection molding, extrusion, blow molding, thermoforming, and others). They soften at elevated temperatures, which the thermosets easily withstand, but offer good low-temperature flexibility,scraprecyclability, availability in a variety of colors, and high gripping friction.Unfortunately,many are more costly than the conventional rubber materials. lm

nY

laed ,le ,

I
,

I I

med In:se ids nd

'

SELECTION OF AN ELASTOMER Elastomeric materials can now be selected and used for a widc range of engineering applications, where they impart properties that include shock absorption, noise and vibration control, sealing, 'orrosion prolection; abrasion protection, friction modification, electrical and thennal insulation, waterproofing, and load bearing. Selection of an elastomer for a specific application requires consideration of many factors, including the mechanical and physical service requirements, the operating environment (including temperature), the desired lifetime, the ability to manufacture the product, and cost. There are a number of families, and within each family there exists a wide range of available properties. Moreover, almost any physical or mechanical property can be altered through additives, which can also be used to enhance processing or reduce cost, and modifications of the processing parameters. Table 8-5 lists some of the more common artificial elastomers, along with natural rubber for comparison, and gives their properties and some typical uses. ELASTOMERS FOR TOOLING APPLICATIONS When an elastomer is confmed, it acts like a fluid, transmitting force un~formly in all directions For this reason, elastomcrs can be sub5tituted for one-half of a die set in sheet-metal-forming operations. Elastomers are also used to perform bulging and to tonn reentranisect~ons that would be impos?~ble to form wlth n g ~ d dies except through the use of costly mult~piece tooling.The engineering eldatomers have become increasingly popular as tool materials because they can be compounded to range from very soft to very hard; hold up well under compressive loadingare imperv~ous to oils,solvents,and other sinular fluids; and can be made into a desired shape qnlckly and econom~cally. I n addition, the elastomer~ctoollng will not mark or damage highly polished or prepainted surfaces.The urethanes are currently the most popular elastomer for tooling applications.

The first matertals used by humans were natural mater~als such as wood and stone The discovery that certain clays could be mixed, shaped, and hardened by firing led to what was probably the first man-made material.Tradit~onal ceramic products, such as bncks and pottery, have continued to be key materials throughout history. More recently, cerumic muterials have assumed important roles in a number of engineering apphcations Most of these ut~liie their outstanding phys~cal properties, including the abilty to w~thstand high temperatures, provide a wide variety ok electrical and magnetic properties, and resist wear i n general, ceramics are hard, brittle, high-melting-point materials wlth low electrical and thermal conductivity,low thermal expansion, good chemical and thermal stability, good creep resistance, high elastic modulus, and h ~ g h compressive

Tensile Strength

cpsi)'
Ela~t~IDel'

Resistance tty3 Pure.

Gum
75

Specific Gravity
0 93 1.20 0 85

Pura
Gum
7300 350

Blaok
40UO

Black
650

Natural tubher
Po!.yaOIy late EDPM (ethylene

Oil P
G

water Swell War

G P
G

Typical Appii~fion

2500

bWr

so0
500

F
G

Ilres,gmI;ets, hpse 6 d h m 0-ria@

Electric ~wolauon, Ewtw~ar, hose,berm Tank hning, o h m x d hose, shoes &e&and heels W ~ r iq$ulab~n, e belts,hweB gaskets, peal$ liamgh TI, sole and hee* ga$ket$ seal$ Same mnatural ~ b b e r

P ~ R P Y ~ ~ ) cllrorwahnated polyethylene Pelychloropr&e (neo$tene) Phlybutadieee

G
G
P

E
G
P

G
G

P
E
P

G 0

d
G

Ss%gasket& dtapfirasm,
valve d & s MoldeQmechanicalgoods, drsposzble pkarmreutrcal
m m q

Silicone

F
G

E
G

P
G

Urethane

E
E

F
E

ElectEtc msulaat~on, r e & s s k m 0c a m Diapba&m,mq scalsmolded goods, tow.kmperatue parts Csster wheels hsel8,foIm paddrng O-ma& sea& ga*kets, toll mering$

strengths that are retained at elevated temperature. A family of "structural ceramics" has also emerged, and these materials now provide enhanced mechanical properties that make them attractive for many load-bearing applications. Glass and glass products now account for about half of the ceramic materials market. Advanced ceramic materials (including the structural ceramics, electrical and magnetic ceramics, and fiber-optic material) compose another 20%. Whiteware and porcelain enameled products (such as household appliances) account for about 10% each, while refractories and structural clay products make up most of the difference.

NATURE A N D STRUCTURE OF CERAMICS Ceramic materials are compounds of metallic and nonmetallic elements (often in the form of oxides, carbides, and nitrides) and exist in a wide variety of compositions and forms. Most have crystalline structures, but unlike metals, the bonding electrons are generally captive in strong ionic or covalent bonds. The absence of free electrons makes the ceramic materials poor electrical conductors and results in many being transparent in thin sections. Because of the strength of the primary bonds, most ceramics have high melting temperatures, high rigidity, and high compressive strength. The crystal structures of ceramic materials can be quite different from those observed in metals. In many ceramics, atoms of significantly different size must he accommodated within the same structure, and the interstitial sites, therefore, become extremely important. Charge neutrality must bemaintained throughout ionic structures. Covalent materials must have structures with a limited number of nearest neighbors, set by the number of shared-electron bonds.These features often dictate a less efficient packing, and hence lower densities, than those observed for metallic materia1s.A.s with metals, the same chemistry material can often exist in more than one structural arrangement (polymorphism). Silica (SiO?), for example, can exist in three forms-quartz, tridymite, and crystobalite-depending on the conditions of temperature and pressure. Ceramic materials can also exist in the form of chains, similar to the linear molecules in plastics. Like the polymeric materials having this structure, the bonds between the chains are not as strong as those within the chains. Consequently, when forces are applied, cleavage or shear can occur between the chains. In other ceramics, the atoms bond in the form of sheets, producing layered structures. Relatively weak bonds exist between the sheets, and these interfacial surraces become the preferred sites for fracture. Mica is a good example of such a material. A noncrystalline structure is also possible in solid ceramics. This amorphous condition is referred to as the glassy state, and the materials are known as glasses. Elevated temperatures can be used to decrease the viscositl of glass, allowing the atoms to move as groups and the material to be shaped and formed.When the temperature is dropped, the material again becomes hard and rigid.The crystalline ceramics do not soften, but they can creep at elevated temperature by means of grain boundary sliding.Therefore, when ceramic materials are produced for elevated-temperature service, large grain size is generally desired. CERAMICS ARE BRITTLE BUT C A N BE T O U C H Both crystalline and noncrystalline ceramics tend to be brittle.The glass materials have a three-dimensional network of strong primary bonds that impart brittleness. The crystalline materials do contain dislocations,but for ceramic materials, brittle fracture tends to occur at stresses lower than those required to induce plastic deformation. There is little that can be done to alter the brittle nature of ceramic materials. However, the energy required to induce brittle fracture (the material toughness) can often be increased. Tempered glass uses rapid cooling of the surfaces to induce residual surface compression.The surfaces cool, contract, and harden. As the center then cools and tries to contract,it compresses or squeezes the surface. Since fractures initiate on the surface, the applied stresses must first cancel the residual compression before they can become tensile. Cermet materials surround particles of brittle ceramic with a continuous matrix of tough, fracture-resistant metal. Ceramic-ceramic composites use weak interfaces that separate or delaminate to become crack arrestors or crack diverters, allowing the structure to continue carrying the load.

Non

:allic Materials: Plastics, Elastomers, Ceramics, and Composites Stabilization involves compounding or alloying to eliminate crystal structur changes and the dimensional expansions or contractions that accompany them. N form heating or cooling can now occur without the stresses that induce fracture. formation toughening stops the progress of a crack by crystal structure changes-th occur when volume expansion is permitted. Fine grain size, high purity, and high densi can be promoted by enhanced processing, and these all act to improve toughness.

CLAY AND WHITEWARE PRODUCTS Many ce~amic products are still based on clay, to which various amounts of quartz a feldspar and other materials are added. Selected proportions are mixed with wa shaped, dried, and fired to produce the structural clay products of brick, roof and str tural tiles, drainage pipe, and sewer pipe, as well as the whiteware products of sanitaty (toilets, sinks, and bathtubs), dinnerware, china, decorative floor and wall tile, potte and other artware. REFRACTORY MATERIALS Refractory materials are .ceramics that have been designed to provide acceptab mechanical or chemical properties at high operating temperatures.They may take form of bricks and shaped products, bulk materials (often used as coatings), and ins lating ceramic fibers. Most are based on stable oxide compounds,where the coarse oxi particles are bonded by finer refractory material.Various carbides, nitrides, and borid can also be used in refractory applications. Refractory ceramics fall into three distinct classes: acidic, basic, and neu Common acidic refractories are based on silica (SiO,) and alumina (A1,0,) and can compounded to provide high-temperature resistance along with high hardness and go mechanical properties. The insulating tiles on the US. space shuttle were made fr machinable silica ceramic. Magnesium oxide (MgO) is the core material for most ba refractories. These are generally more expensive than the acidic materials but are required in metal-processing applications to provide compatibility with the metal. tral refractories, containing chromite (Cr,O,), are often used to separate the acidic basic materials since they tend to attack one another.The combination is often attrac when a basic refractory is necessary on the surface for chemical reasons, and the chea acidic material is used beneath to provide strength and insulation. Figure 8-7 show variety of high-strength alumina components. ABRASIVES Because of their high hardnesh ceramic materials, such as silicon carbide and alumin oxide (alumina), are often used for abrasive applications, such as grinding. Mater such as manufactured diamond and cubic boron nitride have such phenomenal pro ties that they are often termed superabrasives. Materials used for abrasive applica are discussed in greater detail in Chapter 29.

A variety of highstrength alumina (acid refractory) components, including a filter for molten metal. (Courtesy of Wesgo Division, CTE, Hayward, CA.)

FIGURE 8-7

CERAMICS FOR ELECTRICAL AND MAGNETIC APPLICATIONS Ceramic materials also offer a variety of useful electrical and magnetic properties. Some ceramics, such as silicon carbide, are used as resistors and heating elements for electric furnaces Others have semiconducting properties and are used for thermistors and rectifiers. Dielectric, piezoelectric, and ferroelectric behavior can also be utilized in many applications. Barium titanate, for example, is used in capacitors and transducers. Highdensity clay-based ceramics and aluminum oxide make excellent high-voltage insulators. The magnetic ferrites have been used in a number of magnetic applications. Considerable attention has also been directed toward the "high-temperature3'ceramic superconductors. CLASSES When some molten ceramics are cooled at a rate that exceeds a critical value, the material solidifies into a hard, rigid, noncrystalline (i.e.,amorphous) solid: known as a glass. Most commercial glasses are based on silica (SiO,), lime (CaCO,), and sodium carbonate (NaCO,), with additives to alter the structure or reduce the melting point. Various chemistries can be used to optimize optical properties, thermal stability, and resistance to thermal shock. Glass is soft and moldable when hot, making shaping rather straightforward.When cool and solid, glass is strong in compression but brittle and weak in tension. In addition, most glasses exhibit excellent resistance to weathering and attack by most chemicals. Traditional applications include automotive and window glass, bottles and other containers,light bulbs, mirrors, lenses, and fiberglass insulation. There is also a wide variety of specialty applications,including glass fiber for fiber-optic communications,glass fiber to reinforce composites,cookware,TV tubes and monitors, and a variety of medical and biological products. Glass and other ceramic fibers have been used for filtration, where they provide a chemical inertness and the possibility to withstand elevated temperature. CLASS CERAMICS These materials are first shaped as a glass and then heat treated to promote partial devitrification or crystallization of the material, resulting in a structure that contains large amounts of crystalline material within an amorphous base. Since they were initially formed as a glass, glass ceramics do not have the strength-limiting or fracture-inducing porosity that is characteristic of the conventional sintered ceramics. Strength is greater than with the traditional glasses, and the crystalline phase helps to retard creep at high temperatures. Since the thermal expansion coefficientis near zero, the material has good resistance to thermal shock. The white Pyroceram (trade name) material commonly found in Corningware is a common example of a glass ceramic. CERMETS Cermets are combinations of metals and ceramics (usually oxides, carbides, nitrides, or carbonitrides) united into a single product by the procedures of powder metallurgy. This usually involves pressing mixed powders at pressures ranging From 70 to 280 MPa (10 to 40 ksi) followed by sintering in a controlled-atmosphere furnace at about 1650C (3000F). Cermets combine the high hardness and refractory characteristics of ceramics with the toughness and thermal shock resistance of metals.They are used as crucibles, jet engine nozzles, and aircraft brakes, as well as in other applications requiring hardness, stsength, and toughness at elevated temperature. Cemented tungsten carbide (tungsten carbide particles cemented in a cobalt binder) has been used in dies and cutting tools for quite some time. The more advanced cermets now enable higher cutting speeds than those achievable with high-speed tool steel, tungsten carbide, or the coated carbides. See Chapter 22. CEMENTS Various ceramicmaterials can harden by chemical reaction, enabling their use as a binder that does not require firing or sintering. Sodium silicate hardens in the presence of carbon dioxide and is used to produce sand cores in metal casting. Plaster of paris and portland cement both harden by hydration reactions.

ic Materials: Plastics, Elastomers, Ceramics, and Composites

CERAMIC COATINGS A wide spectrum of enamels, glazes, and other ceramic coatings have been developed to decorate, seal, and protect substrate materials. Porcelain enamel can be applied to carbon steel in the perforated tubs of washing machines, where the material must withstand the scratching of zippers, buttons, and snaps along with the full spectrum o laundry products. Chemical reaction vessels are often glass lined. CERAMICS FOR MECHANICAL APPLICATIONS: THE STRUCTURAL AND ADVANCED CERAMICS Because of the strong ionic o r covalent bonding and high shear resistance, ceramic materials tend to have low ductility and high compressive strength. Theoretically, ce ramics could also have high tensile strengths. However, because of their high meltin points and lack of ductility,most ceramics are processed in the solid state, where pro ucts are made from powdered material. After various means of compaction, voids rem between the powder particles, and a portion of these persists through the sinteri process. Contamination can also occur on particle surfaces and then become part oft internal structure of the product. As a result, full theorelical density is extremely di 1 cult to achieve, and small cracks, pores, and impurity inclusions tend to be an integral of most ceramic materials These act as mechanical stress concentrators. As loads applied, the effect of these flaws cannot be reduced through plastic flow, and the resu is generally a brittle fracture. Applying the principles of fracture rnechanie~,~ we fin that ceramics are sensitive to very small flaws.Tensile failures typically occur at stre values between 20 and 210 MPa (3 and 30 ksi),more than an order of magnitude less th the corresponding strength in compression. Since the number,size,shape,and location of the flaws are likely to differ fromp to part, ceramic parts produced from identical material by identical methods ofte fail at very different applied loads. As a result, the mechanical properties of cera products tend to follow a statistical spread that is much less predictable t for metals. This feature tends to limit the use of ceramics in critical high-streng applications. If the various flaws and defects could be eliminated or reduced to very small siz high and consistent tensile strengths could be obtained. Hardness, wear resistance, a strength at elevated temperatures would be attractive properties, along with light we' (specific gravities of 2.3 to 3.85), high stiffness, dimensional stability. low the conductivity, corrosion resistance, and chemical inertness. Reliability might be however, and failure would still occur by brittle fracture. Because of the poor ther conductivity, thermal shock may be a problem. The cost of these "flaw-free' "restricted flaw" materials would be rather high. Joining to other engineering materia and machining would be extremely difficult, so products would have to be fabricate through the use of net-shape processing. Advanced, structural, or engineering ceramic$ is an emerging technology with broad base of current and potential applications. The base materials currently inclu silicon nitride, silicon carbide, partially stabilized zirconia, transformation-toughen zirconia, alumina, sialons, boron carbide, boron nitride, titanium diboride, and cer composites (such as ceramic fibers in a glass, glass-ceramic, or ceramic ma The materials and products are characterized by high strength, high fracture tough fine grain size, and little or no porosity. Applications include a wide variety of we

'According to the principles of iracture mechanics, fracture will occur in a brittlematerial when the fract luughnesb K, is equal to a product involving a dimensionless geometric factor. a, the applied stress, rr. and t square root of the number .ir (3.14) times the sizc of the most critical flawa.

a r (vn)'!"

Whcn the right-hand side is lws than the value of K, the material hears the load without breaking. Fract occurs when the conlhination of applied stress and riaw size equals the critical value K. Sin& K is a mate1 propmy,any attempt to increase theload or stress a matzrial can withstand must be achieved by acompani reduction in flaw size.

FIGURE 8-8 Gas-turb~ne rotors

made of silicon nitnde.

The l~ghtwe~ght materiql (one-

half the weight of stainless steel) offersstrengthat elevated temperature as well as excellent resistance to corrosion and thermal shock. {Courtesy of Wesgo Divis~oo,GTE, Hoyward, CA.)

resistant parts (including bearin@,seals, valves, and dte5), eutting tools,punches, dies, a d englne component$as well as usein heat exchangers,gas turbines, and furnaces Pwous products have been used as substrate mater~al for catalytic converters and as filters for streams of molten metal. Biocompatible ceramiw have been used as substitutes for jo~nts and bones and as dental implants. Alumina (or alum~num oxide) ceramrcs are the most common for industrial appl~calions They are relatively inexpensive and offer high hardness and abrasion resistance, low density, and high ekttrical resistiviry. Alumina is strong in compression and retains useful propert~es at temperatures as high as 1900C(35WF).but it 15 Iimited by low toughnes$ low tenbile strength, and susceptibility to lhermal shock and attack by highly corrosive media. Due to rts high melling point, it is generally processed m a powder form. Silicon carbide and silicon nitride o f k r excellent strenglh and wear resistance with moderate toughness.They work well in high-stress, high-temperature applications, such as turbine blades, and may well repIace nickel- or cobalt-baeed supemIloys Figure 8-8 &ow8 gas-turbine roturs made ffom injeci~on-molded silicon nihide.They are designed t~ operate at 1250T (2300T), where the matcrlal retains over half of its room-temperature strength and does not require external c o o l i . Figure 8-9 shows some addinonal silicon nitride products. Sialon (a silic~n-aluminum-oxygen-nitrogenstructural ceram~c) is really a solid solution of alumma and silicon nitride, and it bridges the gap between them. More aluminum oxlde enhances hardness, wlxile more wlicon nitnde improves toughness. The resultmng matcrial is stronger than steel, extremely hard, and as light as aluminum. It has good resistance to corrosion, wear, and thermal shock; is an electrncal insulator; and retains goad tenslle and compressive strength up to 1400C (255pF). It has ext-ellent dimensional stability, w ~ t h a cocfficlent of thennal expansion that 1s only one-thtrd that of steel and one-tenth that of plastic. When overloaded, however, it exhibits the ceramic property of failure by brittle fracture. Ziconia is inert to most metals and retains strength to temperatures well over 2200C (400UnF). Partially stabihwd zirconia combines the tircoma characteristics bf resistance to thermal shock, wear,and corrosion: low thermal conductivrty; and low friction coeffic~ent w~th the enhanced strength and toughness brought abaut by doping the maternal with oxides of calcium, yttrium, o r magnesium. Transformation-toughened zirwnia has even greater toughness as a result of dispersed second phases throughout the ceramic matrix. When a crack approaches the metastable phase, it transforms to a more stable structure, increasing in volume t o compress and stop the crack. The high coat of the structural ccramics continues to be a harrier to their widespread acceptance. High-grade ceramics are currently several tlmes more expensive than their metal counterparts. Even factor~ng in enhanced lifetime and improved performance, there ts stilI a need to reduce cast. Work continue$ however, toward the deveIopment of a low-cost, high-strength, high-toughness ceramic with a useful temperature range. Parallel efforts are under way to ensure flaw detection in the range of 10 to 50 mm. If these efforts are successfu1,ceramics could compete where tool steels,

8 Nonmetallic Mater~nlr:Plastics, Elastomer$, Ceramics, and Composites

r~?~Z:;;;~i,S:Ili~~:": s&t:?~~~~~qy--r+*&-

frrmrrrarr*I.rtrrr

irau-dii

Mates~al
A1203

g : s {
3.98

Tensile Strength &1)

31)

Compresive Strength

(ksi)
400

Modulus Fractu~e ofEIat~eitg Toughness (lo6psi) (ksi 6,)

56

Sialon SIC ZrO, (partiany stabili&) ZtQ, (ttensforrnati6n toqaendd)

Sr,N, (hot pressed)

3.73 31
5

6Q
73 65 M
XU

5W

560 2 7 0
250
5 W

45 60

4 4
10

39
2 9
45

58
3.2

11 S

powdered metals, coated materials, and tungsten carbide are nowbeing used. Potential applications include engines. turbochargers, gas turbines, bearings, pump and valve seals, and other products that opcrale under high-temperature, high-stress environments. A ceramic automobile engine has been discussed for a number of years. By allowing higher operating temperatures, engine efficiency could be increased. Sliding Criction would be reduced and there would be no need for cooling.The radiator, water pump, coolant, Ian bclt, and water lines could all be eliminated.The net result would be up to a 30% reduction in fuel consumption. Unfortunately, this is still a dream because of the inability to produce large, complex-shaped products with few small-sized flaws. Table 8-6 provides the mechanical propcrtics of some of today's structural ceramics

ADVANCED CERAMICS AS CUTTING TOOLS

Their high hardness, retention of hardness at elevated temperature, and low reactivity with metals make ceramic materials attractive for cutting applications, and cutting tools have improved significantly through advances in ceramic technology. Silicon carbide is a common abrasive in many grinding wheels. Cobalt-bonded tungsten carbide has been a popular alternative to high-spccd tool stccls for many tool and die applications. Many earhidc tools are now enhanced by a variety of vapor-deposited ceramic coatings. Thin layers of titanium carbide, titanium nitride; and aluminum oxide can inhibit reactions between the iiietal being cut and the binder phase of the carbide.This results in a significant reduction in friction and wcar that enables faster rates of cutting. Silicon nitride. boron carbide, cubic boron nitride. and polycrystalline diamond cutting tools now offer even greater tool life. higher cutting speeds, and reduced machine downtime. With advanced tool materials. cutting speeds can be increased from 60 to 1500 mlmin (200 to 5000 ftlmin). The use of these ultra-high-speed materials, however, requires companion developments in the machine tools themselves. High-speed spindles must be perfectly balanced, and workholding devices must withstand high centrifugal forces. Chip-removal mcthods must be able to remove the chips as fast as they are formed. As environmental regulations become more stringent, dry machining may be pursued as a means of reducing or eliminating coolant- and lubricant-disposal problems. Ceramic materials are currently the best materials for dry operations. Ceramic tools have also been used in the direct machining of materials that once required grinding, a process sometimes called hurd muchining. Figure 8-10 shows the combination a1 toughness and hardness for a variety of cutting-tool materials.

WSITE MATERIALS A conapas<iflmaderial is;a n m ,m i @ m

solid can%&ting of two or more different mate+

,als that are mechanically ot( metallurgically banded together. Each of the vakious corn. ponents rerains its ideati5 in the wrnposite and maintains its characteristic structure and properties. There are recognizable interfaes between the materials. The composite material, however, geflerdly possesses characteristic properties (or combinations of

-10 Graphical
~f the cornblned

and hardness for a lutting-tool materials. uperlor hardness of ic rnaterlals.

Toughness

properties), such as stiffness, strength, welght, high-temperature performance, corrosion resistance, hardness, and conduct~vity, which are not possible with the indlv~dual components by thcmse1ves.Analysis of these properties shows that they depend on (I) the properties of the individual components; (2) the relative amounts of the components, (3) the ske, shape, and distribution of the discont~nuous components; (4) the orientat~on of the various components; and (5) the degree of bonding between the components.The matenah Involved can be organics, metals, or ceramics. Hence a wide range of freedom exists, and composite mater~als can often be designed to meet a desired set of engineering properties and characteristics. There are many types of composite materials and several methods of classdying them. One method 1s based on geometry and consists of three dlstinct families:laminar or layered composites, particulate composites, and fiber-reinforced composites.

LAMINAR OR LAYERED COMPOSITES Laminar camposltes have distinct layers of nlatenal bonded together in some manner and include thin coatings,thicker protectwe surfaces,clad&ngs,bimetallics, laminates,sandwiches, and others.They are used to impart properties such as reduced cost, enhanced corro~ion reslstance or wear reslstance, electr~cal ~nsulatlonor conductivity, unique expansion characteristics,lighter weight, improved strength, or altered appearance. Plywood 1s probably the most common engineering material in t h ~ category s and is an example of a lamrnate material. Layers of wood veneer are adhesively bonded with their grain onentatlons at varlous angles to one another. Strength and fracture reslstance are improved, properties are somewhat uniform withln the plane of the sheet, swell~ng and shrinkage tendencies are minim~zed, and large pleces are available at reasonable cost. Safety glass is another laminate in wh~ch a layer of polymeric adhesive IS placed between two pleces of glass and serves to relam the fragments when the glass 1s broken. Aramid-aluminum-laminates (Arall) conslst of thin sheets of aluminum bonded w~th woven adhesive-~mpregnatedaramid fibers.The combination offers light weight coupled w ~ t h high fracture, impact, and fat~gue resistance. Laminated plastics are made from layers of reinforcing material that have been impregnated with thermosett~ng resins, bonded together,and cured under heat and pressure. They can be produced as sheets, or rolled around a mandrel to produce a tube, or rolled tightly to form a rod.Varlous resins have been used with reinforcements of paper, cotton or nylon fabric, asbestos, or glass fiber (usually in woven t o m ) Common applications include a variety of decorative items, such as Formica countertops, ~mitat~on hardwood flooring, and furniture. When combined with a metal layer on one or both surfaces, the material is used for pr~nted circuit boards.

184

C ~ % 8 r n Nonmetal& Mstetials. ~ ~ ~ O Uastcwem, C S ,

Cr~amia a d [email protected].

FIGURE 8-11 Schematic of a

A b~metallic strlp where mater~al has the greater coeff~c~ent of thermal expansion Note the response to cold and hot temperatures

Itshould be noted thatthe. properties,of laminar campositesare alw that is, they are not the same in all directions.Because of the variation in erties will alwaye be diff&ent in the 'rectim pqen&culnr to tha layere

PARTJCULATECOMPOSITES
matrix of anather material. Concrete is a classic example,cmsisting of smd and

ria-dispersed (at TDj nickel, a nickel alloy mntainiig l to 2 w%thoria (ThOZ).

Other types of particulate composites, known as trueparticulate composites, contain large amounts of coarse particles. They are usually designed to produce some desired combination of properties rather than increased strength. Cemented carbides, for example, consist of hard ceramic particles, such as tungsten carbide, tantalum carbide, or titanium carbide, embedded in a metal matrix, which is usually cobalt.Although the hard, stiff carbide could withstand the high temperatures and pressure of cutting, it is extremely brittle. Toughness is imparted by combining the carbide particles with cobalt powder, pressing the material into the desired shape, heating to melt the cobalt, and then resolidifying the compacted material. Varying levels of toughness can be imparted by varying the amount of cobalt in the composite. Grinding and cutting wheels are often formed by bonding abrasives, such as alumina (AI20,). silicon carbide (SiC), cubic boron nitride (CBN), or diamond, in a matrix of glass or polymeric materia1.A~ the hard particles wear, they fracture or pull out of the matrix, exposing Eresh,new cutting edges. By combining tungsten powder and powdered silver or copper, electrical contacts can be produced that offer both high conductivity and resistance to wear and arc erosion. Foundry molds and cores are often made from sand (particles) and an organic or inorganic binder (matrix). Metal-matrix composites of the particulate type have been made by introducing a variety of ceramic or glass particles into aluminum or magnesium matrices. Particulatetoughened ceramics using zirconia and alumina matrices are being used as bearings, bushings, valve seats, die inserts, and cutting-tool inserts. Many plastics could be considered to be particulate composites because the additive fillers and extenders are actually dispersed particles. Designation as a particulate composite, however, is usually reserved for polymers where the particles are added for the primary purpose of property modification. One such example is the combination of granite particles in an epoxy matrix that is currently being used in some machine tool bases. This unique material offers high strength and a vibration-damping capacity that exceeds that of gray cast iron. Because of their unique geometry, the properties of particulate composites are usually isotropic, that is, uniform in all directions.This may be particularly important in engineering applications.

FIBER-REINFORCED COMPOSITES The most popular type of composite material is the fiber-reinforced composite geometry, where continuous or discontinuous thin fibers of one material are embedded in a matrix of another. The objective is usually to enhance strength, stiffness, fatigue resistance, or strength-to-weight ratio by incorporating strong, stifE,but possibly brittle, fibers in a softer, more ductile matrix. The matrix supports and transmits forces to the fibers,protects them from environments and handling. and provides ductility and toughness, while the fibers carry most of the load and impart enhanced stiffness.Wood and bamboo are two naturally occurring fiber composites, consisting of cellulose fibers in a lignin matrix. Bricks of straw and mud may well have been the first human-made material of this variety, dating back to near 800 B.C. Automobile tires now use fibers of nylon, rayon, aramid (Kevlar), or steel in various numbers and orientations to reinforce the rubber and provide added strength and durability. Steel-reinforced concrete is actually a double composite, consisting of a particulate matrix reinforced with steel fibers. Glass-fiber-reinforced resins, the first of the modern fibrous composites, were developed shortly after World War I1 in an attempt to produce lightweight materials with high strength and high stiffness. Glass fibers about 10 pm in diameter are bonded in a variety of polymers, generally epoxy or polyester resins. Between 30 and 60% by volume is made up of fibers of either E-type borosilicate glass (tensile strength of 500 ksi and elastic modulus of 10.5 X 106psi) or the stronger, stiffer, high-performance S-type magnesia-alumina-silicate glass (with tensile strength of 670 ksi and elastic modulus o f 12.4 X l o 6psi).4
'It i s mportant to notc that a flber of material tends to be st~ongc-er than the same rnatenal in bulk form The slze of any flaw 1s hm~ted to the diameter of tho f~ber, and the complete fa~lure of a gwen fiber does not propagatc through the assembly, as would occur in an rdent~cal bulk matcnal

186

C H A ~8 R Nonmetallic

Materials: Plaits, Elastomers, Ceramics, and Cornposies

Glass fibers are still the most widely used reinforcement, primarily because of their lower cost and adequate properties for many applications. Cnrrent uses of glassFiber-reirrforcedplamicsinclude spoi.tinggoads, boar hulls, and bathtubs Limitations o f the glasp-fiber material are generallyrelated tostrength and stiffness.Rltcmat~ve fibprs havb been developed for 5ppEcaIioris requiring enhanced properties. Boron-tungsten fibers (boron deposited on a tungsten core) offer an elastic modulus of 55 X lo6psi wt@ tensile strengths in elceess of 400 ksi. Silicon carbide filaments (Sic on tungsten) have an even higher moduius of elasticity. Graphite (or carbon) and aramid (DuPont tradename uf Kevlar) are other popular reinforcing fibers. Graphite fiberscan be either the PAN type,produced by the. t h e m 1 pyrolys~s of synthetic organic fibers, primartly polyacrylonitrile, or pitch type,

aramid fiber with a tensile Strength up to 650

Ceramic fibers,metal wires, and specia

role. Cotton, hemp, flw,jute, mir (coc mmposites. Themoplastic fibers, s enhaace tha toughnm apd Impacts Table 8-7 lists Some of the key reinforcingf b a s . Smw t h objectiv or high stifhe59 mupled with light weight, properties are oftenreportedasspacific$m and spi@cstifjhess, where the strengthor stiffness d r i e s are divided by density. TIM orientanon of.the fibers within the composite is often bey to properties perfomlance. Sheet-molding mn~pound, bulk-molding compound, and fiberglas erally contain short, randomly orten produce highly directional properties, with the fiber directions bei direction of loading.Woven fabrics or tapes cda be produced and then layered ' orientations to produce a plywood-lie prbduct The layered materials ca stitehed together to add a th'ud dimension to the veveave, and wmplex shapes can be woven fmfibers and later injected with a matrixmateria1. The properties of f i b e ~ r forced ~omposites several characteris ties of the fiber m
Fiber

Matenal
AlxOjwhrsk-~w Boran Ceramic f i t (mullice) E-type glass HI@-strength graphite
High-modulus graplute

Kevbr

5,O 10 I

1410
3.17 608

sic whskers
" i m p d t a e d hy d e n q

26 2

00 % 0952 0.114

6eso
4890

tion of fiber6 generally canno

EIWC m0dn1l~9di~liled by densib, 'Ocmanirnum tomperaturnof use.

Long, thin fibers (higher aspeet provide greater strengh, and a s bond is usually desired betwen fiber and matrk.

SECTION 8.5

Composite Materials

187

The matnx materials should be strong, tough, and ductlle so that they can transmit the loads to the fibers and prevent cracks from propagating through the composite. In addition, the matrlx material is often responsible for provldmg the electrical properties, chemical behavior, and elevated-temperature stability For polymer-matr~x composites, both thermosetting and thermoplastic reslns have been used The thermosets provide high strength and high st~ffness, and the low-v~scosity, uncured resins read~ly Impregnate the fibers. Popular thermosets include epoxies, polyesters, bismalelmides, and polyimides. From a manufacturing viewpomt, it may he easier and faster to heat and cool a thermoplastic than to cure a thermoset. Moreove~, the thermoplastics are tougher and more tolerant to damage. Polyethylene, polystyrene, and nylon are tradit~onal thermoplast~c matrlx mater~als. Improved high-temperature and chemical-resistant propestles can be ach~eved \nth the thermoplastic polyimides, polyphenylene sulfide (PPS), polyether ether ketone (PEEK), and the liquid-crystal polymers When reinforced with high-strength, high-modulus fibers, these materials can show dramatlc improvements m strength, stiffness, toughness, and dimens~onal stability.

ADVANCED FIBER-REINFORCED COMPOSITES Advanced composites are materials that have been developed for applications requiring exceptional combinations of strength, stiffness, and light weight. Fiber content generally exceeds 50% (by weight), and the modulus of elasticity is typically greater 0 ' psi. Superior creep and fatigue. resistance, low thermal expansion, low than 16 X 1 friction and wear, vibration-damping characteristics, and environmental stability are other properties that may also be required in these materials. There are four basic types of advanced composites where the matrix material is matched to the fiber and the conditions of application:

1 . The advanced organrc or resm-mutr lx composites frequently use hlgh-strength,

h~gh-modulus fiber5 of graph~te, aramid (Kevlar), or boron. Properties can be putin desired locations or orlentations at about one-half the weight of aluminum (or one-axth that of steel). Thermal expansion can be des~gned to be low or even negative. Unfortunately, these materials have a maximum service temperature of about 315C (600F) because the polymer ~llatrlx loses strength when heated. Table 8-8 compares the propestles of some of the common resin-matrix compo~ites with those of several oi the lightweight or low-thermal-expansion metals.Typica1 applications include sporting equipment (tenus rackets, sk~s, golf clubs, and fishing poles), hghtweight armor plate, and a myriad of low-temperature aerospace components

I
Material
!

I..

. .. . .. ~ w ' ~ i b i + i n f o ~ c & ~ ~ : o m ~ t e s ( ;he i n Fiber Direction) Compared hi of~w~%mal-~rp"'.nsisn Metals . ' . . . . . . .. . .

Specific Strengthd (lo6 in.)

3.3 0.7 2.1 5.

Density (Iblin?)
0.07 0.065 0.063

Thermal Expansion Coefficient [in./(in.-OF)]

2.2

Thermal Conductiv~ty [Btu/(hr-ft-'F)]

11 01 75
3 05 100 120 h 4

Boron-epoxy
Glass-epoxy (woven cloth) Gmphite-cpoxy: high nlodulus

6
0.5

I'
!
I I

(unidirectinnal) Graphite-epoxy: high strength (unirlirectianal) Kevlar-epoxy (woven cloth)

Aluminum
Beryllium

0.056
0.5 0.10 0.07

1 0.7
1.1

Invar'

0.2

!,
!

Titanium

0.8

0.29 0.16

*Strength divided by density EElaslicmodulus divided by density * A law-expatsion metal conraining3hnA Ni and h4c9i Fe.

Nonmetallic IIlaterials: Plastics, Elastomers, Ceramic$, and Composites 2. M~tal-matrixeomposifes (MMCs) can be used 1250C (2300F), where the conditions require hi trical andior thermal conductivity. exceptional and toughness.The ductile matrix material can be alumi tanium, nickel,superalloy,or even intermetallic compo may be graphite, boron carbide, alumina, or silicon carbide. Fine like single crystals of 1 to 10 pm in diameter) of sapphire, silicon carbide, and silicon nitride have also been used as the reinforcement, as well as w sten, molybdenum, beryllium, and stainless steel. The reinlo ther continuous or discontinuous and typically comprise be con~posite by volume.Compared to the engineering metals, these composi higher stiffness and strength (especially at elevated temperatures); a lower coeff cient of thermal cxpansion; better elevated-temperature properties; and enhance resistance to fatigue, abrasion, and wear. Compared to the organic matrix composite t h ~ offer y higher heat resistance as well as improved electr tivity.They are nonflammable, do not absorb water or gases, and are corros tant to fuels and solvents. Unfortunately, these materials are quite expensive vastly different thermal expansions of the components may lead to debonding, an thc assemblies may be prone to degradation through interdiffusion or galvan corrosion. Graphite-reinforced aluminum can be designed to have near-zero the ma1 expansion in the fiber direction. Aluminum-oxide-reinforcedaluminum used in automotive connecting rods to provide stiffness and fatigue resistance wi lighter weight. Aluminum reinforced with silicon carbide has been fabricated in autornotivc drive shafts, cylinder liners, and brake drums as well as aircraft wingp els, all offering significant weight savings. Fiber-reinforced superalloys may well come a preferred material for applications such as turbine blades.
, Carbon-carbon comflosites (graphite fibers in a graphite or carbon matrix) of

possibility of a heat-resistant material that could operate at temperatures 2000"C(3600"F), along with a strength that is 20 times that of conventional grap a density that is30% lighter (1.38 acmi),and a low co Not only does this material withstand high temperatures when heated. Companion properties include good toughness, good thermal trical conductivity, and resistance to corrosion and abrasion. For temperatures o 540C (lOOOF), however, the composite requires some Form of coating to protect from oxidizing. Various coatings can be used lor different temperature ranges. C rent applications include the nose cone and leading edge of the space shuttle, airw and racing car disc brakes, automotive clutches, aerospace turbines and j components, rocket nozzles, and surgical implants.

. Ceramic-matrix composites (CMCs) offer light weight, high-temperatur

and stiffness,and good dimensional and environmenlalstability.The matri high temperature resistance. Glass matrices can operate a 15QQC (27WF).The mysWliw ceramics, silicon nitride, boron nitride, titanium dibo er temperatures.The fibers add directional strength, increase fracture toughness, prove thermal shock resistance, and can be incorporated in unwoven and braided form. Typical reinforcements include carbon fiber, glass fiber, the various matrix materials, and ceramic whiskers. Composites with disco fibers tend to be used primarily for wear applications, such as cutti dies, and automotive part$ such as valve guides. Other applicati weight armor plate and radomes. Continuous-fiber ceramic composites are us applications involving the combination of hig have been shown to fail in a noncatastrophic man gas-turbine components, high-pressure heat exchanger ters. Unfortunately, the cost of ceramic-ceramic composites rang tremely high, so applications are restricted to those where the benefits are q attractive,

SErrroN 8.5 Composite Mater~als

189

HYBRID COMPOSITES Hybrid composifes involve two or more different types of fihers in a common matrix. Thcparticular combination of fibers is usually selected to balance strength and st~ffness. provide dimensional stabiltty,ieduce cost, reduce weight, or improve fatigue and fracture resatance.Types of hybrid composites include (I) interply (alternat~ng layers of fibers), (2) intraply (muced strands in the same layer), (3) interply-intraply, (4) selectedplacement knittlng (where the more costly material is used only where needcd), and ( 5 ) ~nterply (where phes of one fiber are stitched together with fibers of another type). DESIGN AND FABRICATION The design of composite materials involves the selection of the component materials;the determination of the relative amounts of each component; the delkrmination of size, shape, distribution, and orientation of the components; and the selection of an appropriate fabrication method. Many of the possible fabrication methods have been specifically developed for use with comp~site materials. For example, fibrous composites can be manufactured into useful shapes through compression molding, filament winding, pultrusion (where bundles of coated fibers are drawn through a heated die), cloth lamination, and autoclave curing (where pressure and elevated temperature are applied simultaneously). A variety of fiber-containing thermoset resins premixed with fillers and additives (bulk-molding compounds) can be shaped and cured by compression, transfer, and injection molding to produce three-dimensional fiber-reinforced products or numerous applications. Sheets of glass-fiber-reinforced thermoset resin; again with fillers and additives (sheet-moldir~g compound), can be press formed to provide lightwcight,corrosion-resistant products that are similar to those made from sheet metal.The reinforcing rihers can he short and random, directionally oriented, or fully continuous in a specified direction. With a wide spectrum of materials, geometries, and processes, it is now possible to tailor a composite material product lor a specific application. As one example, consider the cargo beds for pickup trucks, where composite products offer reduced weight coupled with resistance to dents, scratches,and corrosion. A significant portion of Chapter 15 is devotcd to a more complete description of the fabrication methods that have been developed for composite materials. ASSETS AND LIMITATIONS Figure 8-12 graph~cally presents the strength-to-weight ratios of various aerospace mdteriak as a function of temperature. The superiority of the various advanced compositcs over the conventional aerospace metals is clearly evident. The weight of a graphite-epoxy composite I-beam is less than one-fifth thdt of stccl, one-third that of tltanmm, and one-half that of aluminum. Its ultimate tensile strength equals or exceeds that of the other three materials, and it possesses an almost infinite fatigue life. The greatest lim~tations of this and other composites are the11 relative brittleness and the high cost of both materials and fabrication. While there has been considerable advancement in the field,manufacturing with composites can st111 be quite labor intensive, and there is a persistent lack of trained designers, established design guidelines and data, information about fabrication costs, and reliable methods of quality control and inspection. It is often difficult to predict the interfac~al bond strength, the strength of the composite and its response to impacts, and the probable modes of failure Defects can involve delaminations, vo~ds, missing layers, contamination, fiber breakage, and (hard-to-detect) improperly cured resln. There IS often concern about heat resistance. Many composites wlth polymer~c matrices are sensitwe to moisture, acids, chlorides, organic solvents, oils, and ultraviolet radiation. and they tend to cure forever,causlng cont~nually changing properties In addit~on,most composites have limited ability to be repaired if damaged, preventive maintenance procedures are not well estabhshed, and recycl~ng is often extremely difficult.Assembly operations with composites generally require the use of industrial adhesives. On the positrve side, the availability of a corrosion-resistant material w ~ t h strength and st~ffness greater than those of steel at only one-f~lth the weight may be sufficient to justify some engineer~ng compromises Reinforcement fibers can be oriented in

for 9nd

ude fi-

exiitite.

Nonmetallic Materials: Plastics, Elastomers, Ceramics, and Composites High-modulus carbon fiber-reinforced polymer composites

Intermediate-moduluscarbon fiber-reinforced polymer composites

Titanium M M C s
Rapidly solidified

S 0)
C

High-temperature MMCs

Aluminum MMCs Carbonicarbon composites

Aluminum alloys
0

I 500

I
1.000

I 1,500

Temperature. "F
FIGURE 8-12 The strength-to-weight ratio of various aerospace materials as a function of temperature.

Note the superiority of the various fiber-reinforced composites. (Adopted with permission of DuPont Company, Wilmington, DE.)

the direction of maximum stiffness and strength. In addition, products can often be designed to significantly reduce the number of parts, number of fasteners, assembly time, and cost.

AREAS OF APPLICATION
Many composite materials are stronger than steel, lighter than aluminum, and stiffe than titanium. They can also possess low thermal conductivity, good heat resistance, good fatigue life. low corrosion rates, and adequate wear resistance. For these reaso they have become well established in several areas. Aerospace applications frequently require light weight, high strength, stiffnest and fatigue resistance. As a result, composites may well account for a considerable frac. tion of the weight of a current airplane design. Figure 8-13 shows a schematic of th F-22 Raptor fighter airplane.Traditiona1 materials, such as aluminum and steel, makeu only about 20% of the F-22 structure by weight. Its higher speed, longer range, greate agility, and reduced detectability are made possible through the use of 42% titaniu and 24% composite material. Boeing's new 787, a 200-seat intercontinental commerc' airliner, will have a majority of its primary structure, including wings and fuselage, ma of polyme~matrix (carbon-epoxy) composites. A titanium-graphite composite will also be used in the wings of this aircraft, which will use 15 to 20% less fuel than current wid body planes. The Airbus Industries' new A380 wide-body plane will also utilize a hi proportion of composite materials (about 16% by weight) and will mark the intro tion of a new composite material, known as glass-reinforced aluminum, a laminate posite of alternating layers of aluminum and glass prepreg. The new material, w enables a 25% reduction in the weight of fuselage skin, is more fatigue resistant th aluminum and less expensive than a full composite. Sports are highly competitive, and fractions of a second or tenths of a millimetet often decide victories. As a result, both professionals and amateurs are willing to invest in athletic equipment that will improve performance.The materials of choice have evolved from naturally occurring wood, twine, gut, and rubber to a wide variety of high-techno1 ogy metals, polymers, ceramics, and composites. Golf club shafts, baseball bats, Fis

. ~ a n b > s!uual e ~ pue 'tys JaieM ' p o ~ 6u!qs!) e u! ) a~!pe~~e pasn 6u!aq sa~!soduo> (iyb!~) :pJeoqMous Ie!Jaleu al!sodwo3 v ( ~ a i %u!lks 01 osle pue 'ylfiua~ls 46!q pup 'SSaU+j!Js q6!q ' l q 6 ! a ~ 346!1 qfino~q~ a~ueu~opad apjho~d a ~ o ~ d lox u !spoo6 6u!vods u! pasn uavo ale sle!lalecu a l ~ s o d u o j v l - 8 3nn31j

.sl3npold 1er11snp -u! pue ~aurnsuo3 jo 4 a p e e ~103 sZu!snoq [euIa)xa pue ;Z.qsnpu! 1e3!u1aq3 aql 103 slassan pue sad!d'srau!eluo3 pue syum [ernqn3;fie 'slaued 1emnl3al!q3~e 'sr;ey3 'saln13nqs qnz pue JaMoqs uroolyleq 'sIInq leoq se sl3npo1d asIaA!p qans apnpu! suoge3!qdde l a q l o .slaAo3 au!8ua pue 's~aued ~uarunasu! 'spnorqs uej 'sued [ ! o 'saTpoq pue sIIaqs qe3 i o j sal!soduro3 pa3lojuFal-raq!j asu ~ o s1am$~e3nueru u y3n.lJ'0/,~ 01 ~ oz LllelauaS s! sued %ups!xa 01 pamdn103 s a u r ~ e lqS!a~-sradunq s pue 'sZu!~ds 'sveqs a q p apnpu! sIe!mleru al!sodruo3 jo sasn aA!rouroine 's~aued dpoq oi uo!l!ppe u~ ,suo!lea![dde asaqljo Ielanas s ~ o qt.1-8 s a~nZ!d 'sal!sodruo3 snolqg 30 ka!zen ap* e u! a[qeI!eAe ~ o a lue s p ~ e o q ~ opue u s 's!ys 'saurelz apXxq 'slay3el s!uuaj 'smoq h a q m 'spol

('HO ' Y J D ~s/D]aM, '/DUO!]DUJJlU/ NSV f 0 UO!SS!WJ~ l#!M palu!Jda~) '2 q ~ " Z e 6u!X1440 alqede3 s! aueld aq1 'le!JaIeu al!sodwo2 s! %pz pue 'YQzp ( . ~ U ! ~ IJa4sueJl-u!saJ OLU s! Win :a$o~) JOJ slunome lun!uel!l . ~ q f i !Aq a ~%OZ Xluo as!ldwo> 'laair pue unu!lunle se q3ns 's[e!Jalmu leuo!i!peJl-aueld~!e Z Z - j aqj 40 suo!pas sno!JeA aql u! pasn sle!Jaleu aqq f i u ~ ~ oloe~fie!p qs >!leuaq>s 1-83nnl)Y la1468 ~oldea

:sloop Aeq s u o d e a ~ a~!sodwoo :s>ool+ land

wnu!wnle pue al!sodruol WLU awsJ$/spaaqxl

192

CHAITER 8

Nonmetallic Materials: Plastics, Elastomers, Ceramics, and Composites

II Key Words
abrasivc addition polymerization additive agents advanced ceramic advanced composite amorphous anisotropic aspect ratio bulk-molding compound carbon-carbon composite ceramic ceramic-matrix composite cermet clay composite condensation polymerization copolymer cross-linking crystallized polymer degree of polymerization dispersion-strengthened material elastomer fiber-reinforced composite fillers foamed plastic glass hybrid composite isomer isotropic laminar composite nler metal-matrix composite nonmetallic materials oriented plastics particulate composite plastic plasticizer polymer refractory material rubber saturated monomer sheet-molding compound specific stiffness specific strength stress relaxation superabrasive terpolymer thermoplastic thermosetting unsaturated monomer viscoelastic memory whiteware

rlll Review Questions

1. What are syme naturally occurring nonmt.tallic materials that have been used for engineering applications? 2. What are some material families that would be classified under engineering materials? the general term nonme~ullic 3. How might plastics be defined from the viewpoints of chemistry, structure, fabrication. and processing? 4. What is the primary type of alomichonding within polymers? 5. What is the difference between a saturated and an unsaturated molecule? 6. What is an isomer? 7. Describe and differentiate the two means of formiugpolymers: addition polymerization and condensation polymerization. 8. What is degree of polymerization? 9. Describe and differentiate thermoplastic and thermosetting plastics 10. Dcscribe the mechanist~lby which thermoplastic polymers soften under heat and delorm under pressure. 11. What does it mean when a polymer "crystallizes"? 12. What are some of the ways that a themoplasticpolymercan be made stronger? 13. What are the four most common thermaplastic polymers? 14. Why are thermosetting polymers characteristically brittle? 15. How do thermosetting polymers respond to subsequent heating? 16. Describe how thermoplastic or thermosetting characteristics affect productivity during the fabrication of a molded part. 17. What are some attractive engneeringproperties of polymeric materials? 1%.What are some limiting properties of plastics, and in what general area do they fall? 19. What are some environmental conditions that might adversely affect the engineering properties of plastics? 20. What are some reasons that additive agents are incorporated into plastics? 21. What are some functions o f a filler material in a polymer? 22. What are some 01 Lhe more common filler materials used in plastics? 23. What is the difference helween a dye and a pigment? 24. What is the role of a stabilizer o r antioxidant? 25. What is an oriented plastic, and what is the primary engineering benefit? 26. What are some properties and characteristics of the "engineering plastics'? 27. Describe the use of plastic materials as adhesives, In tooling

applications.

28. Describe some of the applications for foamed plastics. 29. What manufacturing features can enhance the attractiven
of plastics as a product material? 30. Which type of plastic is most easily recycled? 31. Why is the recycling of mixed plastics more diEEicult than t recycling of mixed metals? 32. What is the unique meckanical property of elastomericmat rials, and what structural feature is responsible for it? 31. How can cross-linkingbe used to control the engineering pro orties of elastomers? 34. What are some of the materials that can be added to natur rubber, and for what purpose? 35. What are some of the attractive features of the silico rubbers? 36. What is the most common use of an elastomer in a too application? 37. What are some outstanding physical ptoperties of cera materials? 38. Why are the crystal structures of ceramics frequently m complex than those observed for metals? 39. What is the common name given for ceramic material in nancrpstalline, o r amorphous, state? 40. What are someof the ways that toughness can beimparted ceramic mdterials? 41. What is the dominant property of refractory ceramics? 42. What is the dominant property of ceramic abrasives? 43. How are glass products formed or shaped? 44. What are some of the specialty applications of glass? 45. What a r e cermets, and what properties or combination properties do they offer? 46. Why do most ceramic materials fail to possess their theor tally high tensile strength? 47. Why d o the mechanical properties of ceramics gener show a wider statistical spread than the same propertie metals? 48. If all significant flaws or defects could be eliminated fr structural ceramics what properties might be present an features might still limit their possible apphcations? 49. What are some specific materials that are classified as st tural ceramics? 50. What are some attractive and limiting properties of sia (one of the structural ceramics)?

features or properties

composite materials'! feature in a bimetallic strip makes its shape sensitive to ngth that is induced ed particulate comWhich of the three primary composite geometries is most libely to possess isotropic properties? What is the primary role of the matrix in a fiber-reinforced composite? Of the fibers? What are some of the more popular fiber materials used in fiber-reinforced composite materials?

61. What is specific strength? Specificstifmess? 62. What are some possible fiber orientations or arrangements in a fiber-reinforced composite material? 63. What are some features that influence the properties of fiherreinforced composites? 64. What are "advanced composites"? 65. In what ways are metal-matrix composites superior to straight engineering metals? To organic-matrix composites? 66. What featuresmight be imparted by the fibersina ceramic-matlix composite? 67. What are hybrid composites? 68. What is bulk-molding compound and how is it used? Sheetmolding compound? 69. What are some major limitations to the extensive use of composite materials in engineering applications? 70. What are some properties of composite materials that make them attractive for aerospace applications?

Problems
, a.
'

One of Leonardo da V~nci'ssketchbooks contains a crude not attempt to develop or refine this sketch further, possi-

," sketch of an underwater boat (or submarine).Leonardo did

I/ bly because he recognized that the engineeringmaterials of

1
.

his day (wood, stone, and leather) were inadequate for the task. What properties would be required for the body of a ', submersible vehicle? What materials might you consider? / b. Another of Leonardo's sketches bears a crude resemblance

o not exist among

e function of the product or component.

stability, low coefficient ention, and good resisceramic material coat-

ings of titanium carbide (Tic), titanium nitride (TiN),and aluminum oxide (A1,0,), and compare them with respect to the conditions required for deposition and the performance of the resulting coatings 4. Ceramic engines continue to constitute an area of considerable interest and are frequently discussed in the popular literature. If perfected, they would allow higher operating temperatures with a companion increase in engine efficiency. In addition, they would lower sliding friction and permit the elimination of radiators, fan belts, cooling system pumps, coolant lines, and coolant. The net result would be reduced weight and a more compact design. Estimated fuel savings could amount to 30% or more. a. What are the primary limitations to the successful manufacture of such a product? b. What types of ceramic materials would you consider to be appropriate? c. What methods of fabrication could produce a product of the required size and shape? d. What types of special material properties or special processing might he required? 5. Material recyclability has become an important requirement in many manufactured products a. Consider each of the four major materials groups (metals, polymers, ceramics, and composites) and evaluate eachfor recyclability. What properties or characteristics tend to limit or restrict recyclability? h. Which materials within each group are currently heing recycled in large or reasonable quantities? c. Europe has recently legislated extensive recycling of automobiles and electronicproducts. How might this legislation change the material makeup of these products? d. Consider a typical family automobile and discuss how the factors of (1) recyclability, (2) fuel economy, and (3)energy required to producematerials and convert them to products might favor distinctly different engineering materials.

194

CHAPTER 8

Nonmetallic Materials: Plastics, Elastomers, Ceramics, and Composites

Two-Wheel Dolly Handles

he items illustrated in the ftgure are the handle grips for an industrial-quality, pneumabc-tire, two-wheel dolly.They are designed to be bolted onto box-channel tubular sections uslng four bolt-holes, which are sized to accommodate $-mch-diameterbolts The major service requirements are strength, durability, fracture resistance, reasonable appearance, and posslbly light wetght Your employer is currently marketing such a dolly using handles that are made as permanent-mold aluminum castings The firm is in the process of updating its line and is reevaluatrng the design and manufacture of each of its

products The dolly is part of your assignment, and you have been asked spectfically to determine whether the handles should be replaced by an alternative matertal, such as a polymer or low-cost composite Investtgatethe properties and cost* of alternative materials, including means of fabricatingthe desired shape, and make your recommendation Slnce the existing design was for cast metal, you might want to make minor mod~iications. Make sure the alternative mater~als possess adequate properties in the bolt-hole region, and if not, recommend some form of reinforcement

Since the sire of the part will remain relatively unchanged, material costs should be compared on the bass of $/in3or $icm3andnot $Ab or $ 1 ~

ATERIAL

SELECTION
~ ~ ~
,
,

, P ~

,,

, ,

"

'

,
I

.I

Ca%Stndp: M S~mcrrow '

, ,

~,

. ,. .
,,

,
~

"

'[hc object~vc of niar~ufdcturing opcr;~Liuns i s ro 11i;~lic ~ T O ~ U C or I S components that adcquatel! pcriorni their inrcl~dcd task. Mc,cting thk ubi~..ti\.u implid.; thc rnanufacturz

of eompanents kom selected engineering materials, with the requl;ed geometrical shape and precision and with companion material structures and properties that are optimized for the service environment.The idreal product is one that wll just meet all reqnimmenb. Anythlng better will usually incur added cost through higher-grade materials, enhanced prpcmmng, or improved properties that may not be necemary. Anything worse will likely cause product failure, dissatisfied customers. and the possibility of unemployment. It was not that long ago that each of the materials groups had its own well-defined uses and markets. Metals were specified when strength, toughness, and durability were the primaty requirements Ceramics were generally llmited to low-value applications where heat or chern~cal resistance was required and any loadings were compressive. Glass was used for its optical transparency, and plastzcs were relegated to low-value applications where low cost and light weigh? were atttactive features and performance properha were secondary Such clear delinearions no longer exist. Many of the metal alloys in use today did not edsr as little as SO years ago, and the common alloys that have been in use for a century or more have been much improved due to advances in metallurgy and production processes New on the scene are amorphaus metals, dispersion-strmgzhened alloys produced by powdar metallurgy, mechanical alloyed products, and directionally soiidified materials. Ceramic$ polymers, find cornpasites are now avalable wlth specific properties that often transmud the Wsditional limrts and boundaries. Advanced structural materials offer higher streagth and stBness; strength at elevated temperature; light weight; and resbtance to corrosion, creep, and fatigue. Other materials have enhanced thermal, electrical, optical, magnetic, and chemical properties To the inexperienced individual, "wood is wood," hut to The carpenter or craftsman, oak is best for m e application, while maple excels for another, and yellow pine is pstferred fur a third.l"ne mnth edition of "Woldman's Engineering Alloys"' includes over 56,000 metal alloy-s,and that doesn't consider polymers, ceramics, or eompesites. Even if we eitmmate the obsolete and obscure, we are still left with fens of thousands of opttous from which to select the "right" w "best" material for the task at hand. Unforhmately, the avaflability of so many alternatives has often led to poor marerials selection. Money cah be wasted in the unnecessary spec~ficationof an expensive alloy or one that is dmcult to fabricate.At other t i e s , these materials may be absolutely necessary, and selecGon of a cheaper alloy would mean certain failure It is the respunsibility of the design and manufacturing englineer, therefore, to L x
'Woldman1 EngrneeringAliUy~, 91h ehtidn, edited by I. Fnck,ASM Inlernatwnal, Metals Park, OH,2000

195

196

CHAP~ER 9 Material Selection knowledgeable in the area of engineering materials and to b e able to make the be selection among the numerous alternatives. In addition, it is also important that the material selection process be one of c

The automotive industry alone consumes approximately 60 million metric tons engineering materials worldwide every year-primarily steel, cast iron, aluminu copper, glass, lead, polymers, rubber, and zinc. In recent years, the drive toward light more fuel-efficier~l vrhiclcs hasled to an increase in the use of the lightweight metals an high-strength steels, as well as plastics and composites.

rials in airwaft construction has risen froin less than 2% in 1970 lo the point where th

reveals the myriad of components--each with its own characteristic shape, precision,stre

e earliest two-wheeled bicycle frames were constructed of wood, with vari

FIGURE 9-1 (a) Cutaway drawing showing the internal design of a Rolls-Royce jet engine. Notice the number and intricacy of the components. Ambient air enters the front, and hot exhaust exib the back. A wide range of materials and processes will be used in its construction. (b) A full-size Rolls-Royce engine, showing the size and complexity of the product. (Courtesy of Rolls-Royce Corporation, London, England.)

FIGURE 9-2 (a) A tradit~onal two-wheel bicycle frame (1970s v~ntage) made from lolned segments of metal b~cycle w~th one-plece frame, made from hbertublng, (b) a top-of-the-line (Tour de France or tr~alhlon-type) reinforced polymer-rnatr~xcomposlte. (Courtesy of Trek Bicycle Corporabon, Waterloo, WI )

In the 1970s a full circle occurred. Where a pair of bicycle builders (the Wr~ght brothers) pioneered aerospace, the aerospace industry returned to revolutionize bicycles. Lightweight frames were constructed from the aerospace materials oi h~gh-strength alummum, htanium, graphite-reinforced polymer, and even beryllmm. Wall thickness and cross-section profile$ were often modified to provide strength and rig~dity Materials paralleled function as bicycles specialized Into road bkes, high-durabilitymountain bikes, and ultralight racing bikes. Further building on the aerospace expenence, the century-old tubular frame has recently been surpassed by one-piece monocoque frames of either die-cast magnesium or continually wound carbon-fiber epoxy tapes with or wthout selective metal reinforcements One top-of-the-linecarbon-fiber frame now weighs only 2.5 pounds! Figure 9-2 compares a traditional tubular frame wlth one of the newer designs. Window Games were once made almost exclusively from wood. While wood remams a competitive material, a trip to any buildlng supply will reveal a selection that includes anodued aluminum m a range of colors, as well as frames made from colored vinyl and other polymers. Each has its companion advantages and hmitations. Auto bodies were fabr~cated from steel sheet and assembled by resistance spot welding Designers now select from steel, aluminum, and polymeric sheet-molding compounds and may use adhesive bonding to produce the jolnts. The vacuum cleaner assembly shown to Figure 9-3, wh~le not a current model, is typical of many engineering products, where a variety of materials are used for the various components Table 9-1 11ststhe material changes that were recommended in just one past revision of the appliance The materials for 12 components were changed complete% and that for a thirteenth was mod~fied. Eleven different reasons were given for the changes.An increased emphasis on lighter weight has brought about even further changes in both design and materkals. The hst of available engineering materials now includes metals and alloys,ceramics, plastics, elastomers, glasses, concrete, composlte mater~als, and others. It is not surprising, therefore, that a single person might have d~fficulty making the necessary dec~sions concerning the materials in even a s~mple manufactured product. More frequently, the des~gn engineer or design team ~ 1 1work 1 to conjunction with various materials specialists to select the materials that will be needed to convert today's designs into tomorrow's reality.

.2 MATERIAL SELECTION AND MANUFACTURING PROCESSES

The interdependence between materials and then processing must also be recognized New processes frequently accompany new matenals, and their implementation can often cut production costs and improve product quahty.A change in material may well require a change in the manufactur~ng process Conversely, improvements in

198

CHAPTER 9 Material Selection

Inlectlon-molded ABS

Zinc-coated steel s t a m p ~ n g Extruded PVC-

natural rubber

Zinc-coated steel stamplng

FIGURE 9-3 Materials used in

various parts of a vacuum cleaner assembly. (Courtesy

of Advanced Matenols a n d Processes, ASM International, Metals Park, OH.)

\
Black oxide-f~n~shed and lacquered SAE 1113 steel Medlum-dens~ty polyethylene

PVC foot pad, case-hardened and b r ~ g h t

nickel-plated SAE 1010 steel
lever

Part
Bottom plate Wl~eels (carrier and caster) Wheel mount~ng Agitator brush Switch toggle Handle tube Handle ball Motor hood

Former Materral
Assembly of steel stamprngs Molded phmohc Screw-macbne parts H o ~ s e h a bnstles ~r m a die-cast z~nc or slum~num brush back Bakel~te molding AlSI 1010 lock-seam tubing Steel stamplng Molded celluluae acetate (replaced Bakehte) Nickel-plated spring steel, extruded PVC cover Wrapped f ~ b epaper r Molded ABS PVC-coated wire with a s~nalc-nlv - . . PVC extruded wvenng Rubber

New Materral
One-plece alummum dre CaStlnP ~oldeimed~um-densty polyethylene Preasscmbled w ~ t h a cold-headed steel shaft Nylon bristles stapled to a ~olyethvlene brush back Molded ABS Electric ?earn-welded tublng Dre cast alum~num Molded ABS

Benefits More convement servlcmg Reduced n o m

Simplified replacement, more economical Nylon hnstles last seven trmes

lonaer and are now cheaner th& horsehair Breakage e l l m a t e d Less expenswe, better d~mensronal control Better appearance, allowed lower profile for cleanlng under furn~ture Reasonable cost, equal lmpact strength, much improved heat and mglsture resistance el~m~nated warpage problems More economical More flexibll~ty Reduced costs More durahlhty, Lower cost

Extenaon-tube aprlng latch Crev~ce tool Rug oozzle Hose

Molded acetal resin Molded polyethylene figh-rmpact styrene PVC-coated w u e w t h a two-olv W C extruded covering separated by a nylon relnfoiceme~lt PVC

..

Bellows, cleanrngtool nozzles cord insuhtlon, bumper strlps

More e m o m i c a l better agmg and color, lessmarklng

Source MeiaiPmare~r.hypermlsalon

SECTION 9.3 The Des~gn Process

ies

199

processes may enable a reevaluation of the materials being processed. Improper processing of a well-chosen material can definitely result in a defectwe product. If satisfactory products are to be made, considerable care must be exercised in selecting both the engineering rnaterzalr and the rnanufacturmg processes used to produce the product. Most textbooks on materials and manufactunng processes spend considerPerformance able time discussing the mterrelationships between the structure and properties of engineering materials, the processes used to produce a product, and the subsequent performance. As Figure 9-4 attempts to depict, each of these aspects is directly related to all ot the others. An engineering material may possess different properties depending upon its structure. Processing of that material can alter the structure, wh~ch m turn will alter the properties. Altered properties certainly alter perfoimance.7Ae objective of manufacturing, therefore, IS to devise an optimized system of material and processes to produce the desired product.

9.3 THEDESIGN PROCESS

The first step ID the manufacturing process is design-the determiningin rather preose detail what it is that we want to produce and, for each component of the product or assembly, what properties it must possess, what to make it out of, how to make it, how many to make, and what conditions it wiIl see during use Design usually takes place in several dlsknct stages: (1)conceptual, (2) functional, and (3) production. During the conceptual-deagn stage, the designer 1s concerned primarily with the functions that the product is to fulfill. Several concepts are often considered, and a determination is made that the concept is either not practical, or is sound and should be developed further. Here the only concern about materials IS that materials exist that could provide the des~red properties. If such materials are not avaiable, consideration 1s given to whether there is a reasonable prospect that new ones could be developed w~thin the limitations of cost and time At the functional- or engmeenng-der~gnstage, a workable design is developed, including a detailed plan for manufacturing. Geometric features are determ~ned and dimensions are specified,along with allowable tolerances. Specdic materials are selected for each component. Considerahon is given to appearance, cost, reliability, producibiiity, and semlceability,inaddtion to the various functional factors. It is Important to have a complete understand~ng of the functions and performance requirements of each component and to perform a thorough materials analysis, selection,and specification. If these decisions are deferred, they may end up bemg made by individuals who are less knowledgeable about all of the functional aspects of the product Often, a prototype or worhng model is constructed to permit a full evaluation of the product. It is possible that the prototype evaluation will show that some changes have to be made in either the design or materlal before the product can be advanced to production 7A1s should not be taken, however, as an excuse for not doing a thorough job. It is strongly recommended that all prototype? be bullt with the same materials that will be used in production and, where possible, with the same manufacturing techniques. It is of Iittle value to have a perfectly functioning prototype that cannot be manufactured economically in the desired volume or one that is substantially different from what the production units will be like.'

lBecause of the prohibitive cost of a dedicated die or pattern, as might be required for forging o r emtirig, one-of-a-kind ur limited-qllantity prototype pars are often made by machining or one of the newer rapid; prototypetechniques. If the objective is simply to verify dimensional fit and interaction, the prototypr material may be selected For compatibility wilh the prototype process. If performance is to be verified, however, it is best t o use the proper material and process. Machining, for example, simply cuts through the material structure imparted in the manufacture of the starting bar or plate. Casting erases all prior itructure during melting and establishes a new structure during solidiiication. Metalforming processes reorient the starting structure hy @astir flowThe alteredfeaturec caused by these processes may lead to altered performance.

200

CHame~ 9

Material Selection In the production-design stage, we look to full production and determine if t proposed solution is compatible with production speeds and quantities. Can the pa be processed economically,and will they be of the desired quality? As actual manufacturing begins, changes in both the materials and processes be suggested. In most cases, however, changes made after the tooling and machin have been placed in production tend to be quite costly. Good up-front material sele tion and thorough product evaluation can do much to eliminate the need for change. As production continues, the availability of new materials and new processes m well present possibilities for cost reduction or improved performance. Before adop new materials, however, the candidates should be evaluated very carefully to ensu that all of their characteristics related to both processing and performance are welle tablished. Remember that it is indeed rare that as much is known about the properti and reliability of a new material as an established one. Numerous product failures a product liability cases have resulted from new materials being substituted before th long-term properties were fully known.

9.4 PROCEDURES FOR MATERIAL SELECTION

The selection of an appropriate material and its subsequent conversion into a us product with desired shape and properties can be a rather complex process. Nearly e engineered item goes through a sequence of activities that includes: design + mater selection 3 process selection + production + evaluation 7' and possible redesign modification. Numerous engineering decisions must be made along the way. Several methods have been developed for approaching a design and selecti lem. The case-history method is one of the simplest. Begin by evaluating what ha done in the past (engineering material and method of manufacture) or what a co tor is currently doing.This can yield important information that will serve as a start base. Then, either duplicate or modify the details of that solution. The basic assumpti of this approach is that similar requirements can be met with similar solutions. The case-history approach is quite useful, and many manufacturers continu examine and evaluate their competitors' products for just this purpose. The real iss here, however, is "how similar is similar." A minor variation in service requiremen1,su as a different operating temperature or a new corrosive environment, may be suffici to justify a totally different material and manufacturing method. In addition, approach tends to preclude the use of new materials, new technology, and any rn facturing advances that may have occurred since the formulation of the original soluti It is equally unwise, however, to totally ignore the benefits and insights that can gained through past experience. Other design and selection activities occur during the modification of an existi product, generally in an effort to reduce cost, improve quality, or overcome a prob or defect that has been enc0untered.A customer may have requested a product like current one but capable of operating at higher temperatures, or in an acidic enviro ment, or at higher pressure.Efforts here generally begin with an evaluation of the c rent product and its present method of manufacture.The most frequent pitfall, howe is to overlook one of the original design requirements and recommend a change in some way compro~nises the total performance of the product. Examples of s oversights, where materials have been changed to meet a specific objective, are provi in Section 9.8. The safest and most comprehensive approach to part manufacture is to follow t full sequence of design,material selection, and process selection, considering all aspe and all alternatives.This is the approach one would take in the development ofan eati new product.

'

.~ -

S E C ~ 9.4 ~ ~ Procedures N for Mater~al Selection

201

Before any decisions are made, take the time to fully define the needs of the product. What exactly is the "target" that we wish t o hit? We must develop a clear picture of all of the characteristics necessary for this part to adequately perform its intended function and do so with no prior biases about material or method of fabrication.These requirements will fall into three major areas: (1) shape or geometry considerations, (2) property requirements, and (3) manufacturing concerns. By fist formulating these requirements, we will be in a better position t o evaluate candidate materials and companion methods of fabrication.

GEOMETRIC CONSIDERATIONS A dimens~oned sketch can answer many of the questlons about the size, shape, and complex~ty of a part, and these geometrzc or shape consideratzons wlll have a stronglnfluence on dec~slons relatlng to the proposed method or methods of fabrication? While many features of part geometry are somewhat obv~ous, geometric considerahons are often more complex than first imagmed. Typlcal questlons might mclude:

1. What is the relative size of the component? 2. How complex is its shape? Are there any axes or planes of symmetry? Are there any
unifornl cross sections? Could the component be divided into several simpler shapes that might be easier to manufacture? 3. How many d i e u s i o n s must be specified? 4. How precise must these dimensions be? Are all precise? How many are restrictive, and which ones? 5. How does this component interact geometrically with other components? Are there any restrictions imposed by the interaction? 6. What are the surface-finish requirements? Must all surfaces be finished? Which ones do not? 7. How much can each dimension change by wear or corrosion and the part still function adequately? 8. Could a minor change in part geometry increase the ease of manufacture or improve the performance (fracture resistance, fatigue resistance, etc.) of the part? Producing the right shape is only part of the desired objective. If the part is to perform adequately, it must also possess the necessary mechanical andphysicalproperties, as well as the ability to endure anticipated environments for a specified period of time. Environmental considerations should include all aspects of shipping, storage, and use! Some key questions include those listed in the following three sections.

MECHANICAL PROPERTIES
1 . How much static strength is required? 2. If the part is accidentally overloaded,is it permissible to have a sudden brittle fracture,

or is plastic deformation and distortion a desirable precursor to failure?

3. How much can the material bend, stretch, twist, or compress under load and still

4. 5.
6. 7.

function properly? Are any impact loadings anticipated? If so, of what type, magnitude, and velocity? Can you envision vibrations or cyclic loadings? If so, of what type, magnitude, and frequency? Is wear resistance desired? Where? How much? How deep? Will all of the above requirements be needed over the entire range of operating temperature? If not, which properties are needed at the lowest extreme? At the highest extreme?

cision and surface finish become progressively wone as we move from die casting, to permanent mold, to sand. !3xtmriun and rolling can be used to produce long parts with constant cross section. Powder metallurgy parts must be ahle to he cjected from a compacting d i e

MaterialI Selection

PHYSICAL PROPERTIES (ELECTRICAL, MAGNETIC, THERMAL, AND OPTICAL) 1 . Are there any electrical requirements? Conductivity? Res~stivity? 2. Are any magnetic properhes des~red? 3. Are thermal propert~es significant? Thermal conduct~vity? Changes in dimensi with change in temperature? 4. Are there any optical requirements? 5. Is weight a sig~llficant factor? 6. How Important is appearance? Is there a preferred color, texture, or feel? ENVIRONMENTAL CONSIDERATIONS 1 . What are the lowest, highest, and normal temperatures the product will see? temperature changes be cyclic? How fast will temperature changes occur? 2. What is the most severe environment that is ant~c~pated as far as corrosion or de norahon of materlal properties a concerned? 3. What is the deslred service lifetime for the product? 4. What is the anticipated level of inspection and mamtenance dunng use? 5. Should the product be manufactured with d~sassembly,repairab~lity,or recyclab~l in mind? MANUFACTURING CONCERNS A final area of considerat~on is the variety of factors that will directly influence method of manufacture. Some of these manufacturing concerns are:

1 . How many of the components are to be produced? At what rate? (Note: One-of-a-k
parts and small quantities are rarely made by processes that requue dedicated pane molds, or dies, since the expense of the tooling 1s hard to just~fy. High-volume,highproducts may requlre automatable processes.) 2. What is the desired level of quahty compared to similar products on the market? 3. What are the quahty control and inspection requirements? 4. Are there any assembly (or disassembly) concerns? Any key relationships or res tions with respect to mating parts? 5. What are the largest and smallest section thicknesses? 6. Have standard sizes and shapes been specified wherever possible (both as finis shapes and as startmg raw matenal)? What would be the preferred form of sta material (plate, sheet, foil, bar,rod, wre, powder, mgot)? 7. Has the des~gn addressed the requirements that w~ll facilitate ease of manufact (machmabiltty, eastability,iormabllity, weldability,hardenability)? 8. What is the potential liab~lity if the product should fail? 9. Are there any end-of-use disposal concerns? The considerat~ons just mentioned are only a sample of the many questlo must be addressed when precisely definlng what it is that we want to produce there is a natural tendency to want to jump to an answer, in this case a material method of manufacture, time spent determining the varlous requirements w i l l be rewarded. Collectively, the requirements direct and restrict material and process se tions. It is possible that several fam~lies of materials, and numerous members those famd~es, all appear to be adequate. In this case, selections may become a of preference. It is also possible, however, that one or more of the requiremen emerge as a dominant restr~ctor (such as the need for ultra-h~gh strength, supenor reastance, the ab111ty to funct~on at extreme operating temperatures, or the abili w~thstand h~ghly corrosive environments), and selection then becomes focused on t materials offering that specific characterist~c. I t is important that all factors be listed and all service condit~ons and use considered Many fa~lures and product hability cla~ms have resulted from engln oversights or fatlure to consider the entire spectrum of conditions that a product

experience in its lifetime. Consider the failure of several large electric power transformers where fatigue cracks formed at the base of horizontal cooling fins that had been welded to the exterior of the casing.The subsequent loss of cooling oil through the cracks led to overheating and failure of the transformer coils. Since transformers operate under static conditions,fatigue was not considered in the original design and material selection. However, when the horizontal fins were left unsupported during shipping, the resulting vibrations were sufficient to induce the fatal cracks. It is also not uncommon for the most severe corrosion environment to be experienced during shipping or storage as opposed to normal operation. Products can also encounter unusual service conditions. Consider the numerous parts that failed on earthmoving and conslruction equipment when it was used in the construction of the trans-Alaskan oil pipeline. When this equipment was originally designed and the materials were selected, extreme subzero temperatures were not included as possible operating conditions. Once we complete a thorough evaluation of the required properties, it may he helpful to assign a relative importance to the various needs. Some requirements may be absolutes, while others may be relative. Absolute requirements are those for which there can be no compromise.The consequence of not meeting them will be certain failure of the product. Materials that fall short of absolute requirements should be automatically eliminated. For example, if a component must possess good electrical conductivity, most plastics and ceramics would not be appropriate. Relative or compromisable properties are those that frequently differentiate "good,""better,"and "best," where all would be considered as acceptable.

When evaluating candidate materials, an individual is often directed to handbooktype data that has been obtained through standardized materials characterization tests. It is important to note the conditions of these tests in comparison with those of the proposed application. Significant variations in factors such as temperature, rates of loading, or surface finish can lead to major changes in a material's behavior. In addition,one should keep in mind that the handbook values often represent an average or mean and that actual material properties may vary to either side of that value. Where vital information is missing or the data may not be applicable to the proposed use,one is advised to consult with the various materials producers or qualified materials engineers. At this point it is probably appropriate to introduce cosr as an additional factor. Because of competition and marketing pressures, economic considerations are often as important as technical ones. However, we have chosen to adopt the philosophy that cost should not be considered until a material has been shown to meet the necessary reouirementsif aeceotable candidates can be identified. cost will certainlv become an important part of the selection process, and both material cost and the cost of fabrication should be cons~dered.~ Often, the final decision involves some form of compromise among material cost, ease of tabrication, and performance or quality. Numerous questions might be asked, such as: 1. Is the material too expensive to meet the marketing objectives? 2. Is a more expensive material justifiable if it offers improved performance? 3. How much additional expense mght be justified to gain ease of fabrication?

In addition, it is important that the appropriate cost figures be considered. Material costs are most often reported in the form of dollars per pound or some other form of cost per unit weight. If the product has a fixed sue, however, material comparisons should probably be based on cost per unit volume. For example, aluminum has a density about one-third that of steel. For products where the size is fixed, 1pound of aluminum
'A more appropriate w s t consideration might be total lifetime cost, which begins with the starting material, the energy t o produce it, and the environmental impact of its production. To this are added the cost of converting it into the desired product, the cost of operating or using the product through its full lifetime, and finally the cost of disposal orrecycling.

can be used to produce three times as many parts as 1po cost of aluminum were less than three times that of steel, alum the cheaper material. Whenever the densties of materials are quite magnesium and stamless steel, the relative rankings based on cost per pound and cos per cubic ~ n c h can be radically dxferent Material avallabilzty is another important consideration.The mat not be available in the size, quantity, or shape des form at all. The dlverslty and rellab~hty of suppl facilitate competitive prlcing and avoid pruductio may be a problem, one should be prepared to recommen vtded that they, too, are feas~ble candidates for the specific use. Still other factors to be considered when making material select~ons include: 1. Are t h e ~ e possible misuses of the product that should be considered? If the pro IS to be used by the general public, one should definitely anticipate the worst. S drivers are routinely used as chisels and pry bars (different forms of loa the intended torsional twist) Sc~ssors may be used as wire cntt are similarly misused 2. Have there been any falluses of this or stmilar products? If so, what were the ide fled causes and have they been addressed in the current prod results should defin~tely be made available to the designers, who can directly ben from them 3. Has the material (or class of materials) being cons~dered established a favorable unfavorable performance record? Under what conditions was unfavorable perfoman noted? 4. Has an attempt been made to benefit from materlal standardiz ple components are manufactured from the same materlal or by the same tunng process? Although function,reliab~I~ty and appearance sh oneshould not overlook the potential for savings and simphfication that sta has to offer.

OF THE MANUFACTURING

PROCESS

The overall attractiveness of an engineering material depends not only on irs physi and mechanical properties but also on our ability to shape it into useful objects in economical and timely manner. Without the necessary shape, parts cannot wlthout economical production, the material will he limit cations. For this reason, our material selection qhould be further the possible fabrication processes and the su~tability of each "prescreene each of those processes. Famiharity with the various manufacturing alternatives is a cessity, together wlth a knowledge of the assoelated limitations, economics, product q ity, surface finish, prec~sion, and so on. All processes are not compatible with all materi Steel, for example, cannot be fabricated by die castmg. Tltanmm can be forged su fully by isothermal techniques but generally not by conventional drop hammers.W alloys cannot be cast, and casting alloys are not attractive for forming. Certain fabrication processes have d~stmct ranges of product nze, shape, and ness, and these should be compared with the requirements of the has its characteristic precinon and surface finish. Since secondary oper machining, grinding, and polishing, all require the handling, posttionlng, and proce of ind~vidual parts, as well as add~tional toohng, they can add significantly to man turing cost. Usually it is best to h ~the t target wlth as few operations as processes requre prlor heating or subsequent heat treatment. Sttll other mclude production rate, production volume, desired level of automation, of labor required, especially if it is skilled labor.AIl of these concerns will be refle In the cost of fabrication.There may also be additional constra design a product so that it can be produced with existlng equipment or a minimum of lead time, or with a minimal expendltnre for dedicated toolmg.

of a component des~gn, such as the presence of cored features in a castlng, the magnitude of draft allowances, or the recommended surface fin~ahThe designer often spec~fies these features pnor to consultation wlth manufactur~ng experts. It is best, therefore, to consider all possible methods of manufacture and, where dppropnate, work wlth the designer lo Incorporate changes that would enable a more attractwe means of production.

9.7 ULTIMATE OBJECTIVE

The real objective of thrs adrvity is to develop a manufacturing system-a comhinat~on for a given of material and process (or sequence of processes) that IS the best solut~on product. Flgure 9-5 depicts a series of act~vities that move Crom a well-def~nedset ol needs and ohje~tl~e6 through matenal and processselectlon L o the manufacture and evaluat~on of a product. Numerous decisions are required, most of which are judgmental In nature. For example, we may have to select among"good:'"better;' andA'best," where "better" and "best" carry Increments of <zddedcost, or make compromrses when all of the requuements cannot be s~multaneously met. Wh~le Figure 9-5 depicts the various achvities as hav~ng a defmite, sequential pattern, one should be aware that they are often rearranged and are defln~tely interrelated figure 9-6 shows a modified form, where malertal selection and process select~on have been moved to be parallel instead of sequential. It is not uncommon for one of the two selections to be dominant and the other to hecome dependent or secondary. For example, the production of a large quantity of small, intricate parts with thin wall?,preclse dimensions, and smooth surfaces 1s an ideal candidate for d ~ caste ing. Mater~al select~on, therefore, may be limited to d~e-caatable materials-assuming Eeas~blealternatives are ava~lable. In a conveise example, h~ghlyrestrictive mater~al properties, such as the abrhty to endure extremely elebated temperatures or severe corrosive environments, may s~gnificantly limit the mater~al options. Fahricatton oplions will tend to be limited to those proceses that are compatible with the candidate materials. In both models, decls~ons in one area generally impose restrict~ons or limitations in another. As shown 1 1 1 Figure 9-7 selection of a materlal may lim~t processe$ and selection

Matertal Selectran
(based on mechanical and phys~cal properttes
and
. e

Process Select~on
'based
~

On

geometric

environmental
concerns)

and manufactur~ng
requirements)

Manufacture

Manufacturtng System Selectton

____r___
Manufacture

+
Evaluation

RE 9-5 Sequent~al flow

I
FIGURE 9-6 Alternative flow chart showing parallel select~on of material and process.

how~ng activ~t~es leading production of a part or

206

CHAPIER 9

Mater~al Selection

FIGURE 9-7

Compatibility chart of mater~als and processes. Selection of a rnater~alrnav restrict possible rnatenals: Key X = Routinely performed 0 = Performed wlth d~ff~culry, cautlon, or some sacr~f~ce (such as dle life) Blank = Not recommended

properties to the material, and all of these may not be beneficial (considcr anisotrop porosity, or residual stresses). Processes designed to improve certain properties (such as

each with its own strengths and limitations. Compromise, opinion. and judgment enter into the final decision making, where our desire is to achieve the bcst soluti will help ensure that all o l the necessary factors were considered and weighed in m a ing the ultimate decision. If no material-proccss combination meets the requiremen

quirements is acceptable. Materials specialists bring expertise in candidate materi and the ellects of various processing. Manufacturing personnel know the capabilities processes, thc equipment available. and the cost of associated tooling. Quality ande

sentatives or marketing specialists may also be consulted lor their opinions. Clear open communication is vital to the making of sound decisions and compromises. n l e design and manufacture of a successful product is an iterative, evolving,a continual process.The lailure of a component or product may have revealed deficient in design, poor material selection, material defects, manufacturing delects. improp

SECTION 9.9

Effect of Product Liability on Materials Selection

207

assembly, or improper or unexpected product use. The costs of both material and processing continually change, and these changes may prompt a reevaluation.The availability of new materials, technological advances in processing methods, increased restrictions in environment or energy, or the demand for enhanced performance of an existing product all provide a continuing challenge. Materials availability may also have become an issue.A change in material may well require companion changes in the manufacturing process. Improvements in processing may warrant a reevaluation of the material.

As new technology is developed or market pressures arise, it is not uncommon for new materials to be substituted into an existing design or manufacturing system. Quite often, the substitution brings about improved quality, reduced cost, ease of manufacture, simplified assembly,or enhanced performance. When making a material substitution,however, it is also possible to overlook certain requirements and cause more harm than good. Consider the efforts related to the production of lighter-weight, more fuel-efficient, less emission-producing automobiles. The development of high-strength low-alloy steel sheets (HSLA) provided the opportunity to match the strength of traditional body panels with thinner-gage material. Once some of the early forming and fabrication problems were overcome, the substitution appeared to be a natural one. However, it is important to consider the total picture and become aware of any possib1e:compromises. While strength was indeed increased, corrosion resistance and elastic stiffness (rigidity) remained essentially unaltered. The thinner sheets would corrode in a shorter time, and previously unnoticed vibrations could become a significant problem.Measures to retard corrosion and design modifications to reduce vibration would probably be necessary before the new material could be effectively substituted. Aluminum sheet has replaced steel panels, enabling a 50% reduction in weight, but the vibration problems associated with the lower elastic modulus required special design consideration. Aluminum castings might be considered as an alternative to cast iron for engine blocks and transmission housings. Corrosion resistance would be enhanced and weight savings would be substantial. However, the mechanical properties must be ensured to be adequate, and consideration would also have to be given to the area of noise and vibration. Gray cast iron has excellent damping characteristics and effectively eliminates these undesirable features.Aluminum transmits noise and vibration, and its use in transmission housings would probably require the addition of some form of sound isolation material. When aluminum was first used for engine blocks, the transmitted vibrations required a companion redesign of the engine support system. Polymeric materials have been used successfullyfor body panel$ bumper$ fuel tanks, pumps, and housings. Composite-material drive shafts have been used in place of metal. Cast metal, powder metallurgy products, and composite materials have all been used for connecting rods. Ceramic and reinforced plastic components have been used for engine components. Magnesium is being used for instru~nenl panels and steering wheels. F~berreinforced polymer composite has beenused to produce the cargo beds for pickup trucks. When making a material substitution in a successful product, however. it is important to first consider all of the design requirements. Approaching a design or material modification as thoroughly as one approaches a new problem may well avoid costly errors. Table 9-2 summarizes some of the weight-saving material substitutions that have been used on automobiles and calls attention to the Fact that many of these substitutions are accompanied by an increase in cost, where total cost incorporates both cost of the material itself and the cost of converting that material into the desired product.

Product liability actions, court awards, and rising insurance costs have made it imperative that designers and manufacturers employ the very best procedures in selecting and processing materials. Although many individuals feel that the situation has grown to absurd proportions, there have also been many instances where sound procedures were not

laterlal Selection

Magnesmm Msgnes~um Glass fibs

Steel or cast imn

AIumlnum

&35%

Steel

25-35%

100-150% lfl(t150%

reinforced plasoc
Qaaiaiien fxom"Autdmot~w Materisls lq me 21s) CenNry," by WlllamF Yuwen published m Advanced Mulerbls andP+ocers~s. MayZoOO

used in selecting materials and methods of manufacture. In today's business and climate, such negligence cannot be tolerated. An examination of recent product liability claims has revealed that the five m common eansec have been:
1, Failuw to know and use the latest and best information about the materials be

specified 2. Failure to foresee, and account for, all reasonable uses of the product 3. Use of ma@rjdsfot-whichthere %ereinsufficient or uncertain data,parriculaily regard to long-term properties 4. Inadequate and unverified quality cant101 procedures 5. Material selection made by people who were completely unqualified An examination of these faults reveals that there is no good reason for t to exist Consideration of each, however, is good practice. when seeking to ensure production of a quality product and can greatly reduce the number and magnitud product lfabllity clahns,

ATERIAL

SELECTION

From the discussion in this chapter, it is apparent that those who select materials have a broad, basic understanding of rhe natuze and properties of matenals an processing characteristic$,Providing this baek~ound is a primarv purpose of th18 The number of engineering materials is so great, however, and the mass of inform that is both available and usefulis so large, that a srngte book of this type and sue ca be expected to furnrsh all that is required. Anyone who does much work in m selection needs t~ have veeady access to many sources of data. It is ahnost imperative that one have accesq to the infomation contained various volumes ofMe#Is Handbuok,published byA8M 1ntemational.Thismultivo series contains a wealth of information about both engineering metals and assoc m a n u f a d g processes The one-volume Metals Handbook Desk Edition provides highlights o f t h i information in a less voluminous, more concise f0nnst.A par Engineered Iblate12aIs Hmdbook series and one-volume Desk Edinbn prov~des formation for composites, plastic& adhesives, and ceramics. These resowres also available on computer CD-ROMs and directly via the Internet through subscription. ASM also offers a one-volume ASM Metals Reference Book that provides sive data about metals and metalworking in tabuldr or graphic form. Smithells Reference BoDk provides nearly 2000pages af useful information and data. Add handbooks are available for specific classes of materials, such as titanium alloy less steels, tool steels, plastics,and composites.Vahous technical magazines often annual issues that serve as infosmation databooks. Some of these include Plastics, Indwtrrai Ceramics, and ASM's Advmcd Materials and Processes.

handbooks published by various mater~als orgamzatlons, technical soaeties, and trade associations. These may be material related (such as the Alum~num Association's Alumcvlum Standards and Data and the Copper Development Association's Standards Handbook: Copper, Brass, and Bronze), process related (such as the Steel Castings Handbook by the Steel Founder's Society of America and the Heat Treater's Gurde by of ASM International), or profession related (such as the SAE Handbook by the Soc~ety Automotive Engmeers, the ASME Handbook by the American Soclety for Mechanical for ManEngineerqand rhe Tool and Manufacturing Engineers Handbook by the Soc~ety ufacturing Engneers).These may be supplemented further by a variety of supplier-provided information. While the latter is excellent and readily available, the user should recognize that suppher loformation m~ght not provlde a truly oblect~ve v~ewpomnt. It is also important to have accurate informat~on on the cost of various materials. Since these tend to fluctuate,it may be necessary to consult a daily or weekly pubhcatlon such as the American Metal Market newspaper or online servlce Costs associated w ~ t h variou6 processing operations are more difficult to obtain and can vary greatly from one company to another. These wsts may be available from within the firm or may have to be estimated from outside sources. A varlety of texts and software packages are available. Each of the above references provldes focused information about a class of materials or a specific type of process. A number of lexts have attempted to achieve integration with a focus on design and matenal select~on. Possibly the most well known 1s the work of M.F.Ashby, with his Materials Selection in Mechanical Deszgn text and tools, and the Cambndge Materials Selector database that was developed to use them. With the evolution of h~gh-speedcomputers w ~ t h large volumes of searchable memory,materials select~on can now be computerized. Most of the textbook and handbook references are now available on CDs or directly on the Internet, and a 1 1 of the mformation tn an entire handbook series can he accessed almost instantaneously.Programs have been wrltten to ucil~ze informat~on databases and actually perform materials selection. The various property requirements can be specifled and the entlre spectrum of engneermng materials can be searched to ident~fy possible candidates. Search parameters can then be tightened or relaxed so as to produce a deslred number of candidate materials. In a short penod of time, a wlde range of materials can be considered, far greater than could be considered m a manual selection. Process simulation packages can then be used to verify the hkelihood of producing a successful product. While the capabilites of computers and computer software are indeed phenomenal, the knowledge and experience of tralned lndiv~duals should not be overlooked. Experienced personnel should reevaluate the f ~ n a l mater~als and manufacturing sequence to ensure full comphance with the needs of the product. The appendix titled "Selected References For Add~tional Study" prov~des on extensive List of additional resources.

It Key Words
environmental considerations functional design geornetrlc requirements manufacturing concerns matenal ava~lability matenal selection mater~al substitution mechanical propert~e, phys~cal propcmes product liability product~on des~gn prototype recyclab~hty relatlve requlrernent sermce envuonment

El4 Review Questions

. What is the objecnve of a manufacturingoperatlon,and what
are some of the deta~ls m rneetlng this objective?

4. How have d~fferent materials enabled advances and specializations in b~cycle manufacture' What are some poss~ble undesirable features of s~gnificantly 5. Drscuss the ~nterrelatlon between engineerlug matenal and exceedmg the requirements of a product' the fabrlcat~on processes used to produce the desired shape In a manufacturing environment,why should the select~on and properties. and use of englneenng materials be a matter of constant 6. What 1s des~gn?

7. What are the three primary stages of product design, and how does the consideration of materials differ in each? 8. What is the benefit of requiring prototype products to he manufactured from the same materials that will be used in production and by the same manufacturing techniques? 9. What sequence of aclivities is common to nearly every engineered component or product? 10. What are some of the possible pitfalls in the case-history approach to materials select~on? when seekmg to improve an What IS the most frequent p~tfall ex~slmg product? What should hethe first step ~nany m a t e r ~ lselecuon s problem? In what ways d o the concept of shape or geometry go beyond a d~meusioned sketch? How m ~ g h lemperatore t enter into the spec~ficahon of m e chanical properties? 15. What are "physical properties" of materials? 16. Whatare some of the important aspects o i the service envirmment to be considered when selecting an engineering material? 17. What are some of the possible manufacturing concerns that should be considered? 18. Why is it important to resist jumping to the answer and first perform a thorough evaluation of product needs and requirements considering all factors and all service conditions? 19. What is the difference between an absolute and relative requirement'? 20. What arc some possible pitfalls when using handbook data to, assist in materials selection?

'

21. Why might it be appropriate to defer cost considerations after evaluating the performance capabilities of various neering materials? 22. Give an example of aproduct or component where mater cost should be compared on a cost-per-pound basis. Giv contrasting example where cost per unit volume would more appropriate. 23. In what way . mieht - failure analysis data be useful in a mater selection decision? Why should consideration of the various fabrication pro possibilities be included in material selection? What aspec a manufacturing process should be considered? Why might it be better to perform material selection process selectionin a parallel,as opposed to sequential, fashi Give an example of where selection of a material may processes and where selection of aprocess may limit mat 27. Why is it likely that multiple individuals will be iuvolv the material and orocess selection activitv? 28. Why should the design and manufacture of a successfulpr uct be an iterative, evolving, and continual process? 29. Give an example of an unexpected problem that occurred a materials substitution. 30. What are some ofthe most common causes of product liabi losses? 31. How have high-speed, high-capacity computers changedma rials selection? Have they replaced trained individuals?

Ei Problems
I. One hmplc tool that has been developed to assist inmaterials selection is a rating chart, such as the one shown in Figure 9-A.

Absolute properties are identified and must be present for amaterial to be considered. The various relative properties are weyhted as to their significance, and candidate materials are rated on a scale such as 1to 5 or 1 lo 10 wifh regard to their ability to provide that property. A rating number is then computed by multiplying the property rating by its weighted significance andsumming theresults.Potentialmaterials can then be compared in a uniform, unhiased manner, and the best can&dates can often b e identified. In addition, by placing all the requirements on a single sheet of paper, the designer is less likely to overlook a major requirement. Finalist materials should then he reevaluated to assure that no key requirement has been overlooked or excessively compromised. Three materials-X,Y, and Z,-are available for a certain use. Any material sclected must have good weldability Tensile strength, stiffness, stability, and fatigue strength have also been identified as k6y requirements. Fatigue strength is considered the most important of these requirements, and stiffness is least important. The three materials can be rated as follows:

Weldabilrty Tensile strength

Stiffness

Stabd~ty Fatlgue strength

X Excellent Good Good Good

Fax

Y Poor
Excellent

z
Guud Far Good Good Excellent

Good Excellent Good

Develop a ratmg chart cuch as that ln Figure 9-A to determ~ne wh~ch mater~al you would recommend.

The chalk tray on a classroom chalkboard has very few formauce requirements. A s a result, it can be made wide spectrum of materials. Wood,aluminum, and e tic have been used in this application.r)iscuss the per and durability requirements and the pros and cons of listed materials Chalk trays have a continuous cross se but the processes used to produce such a configuratio vary with material. Discuss how a chalk tray might h produced from each of the three materials classifi Might this be a candidate for some form of wood by-pro similar to particle board? Since the product demands are1 might some form of recycled material he considered? Examine the properties of wood, aluminum, and exhuded as they relate to household window frames. Discuss the and cons of each, considering cost, ease of manufacture, aspects of performance, including strength, energy efficie thermal expansion and contraction,response to moisture humidity, durability, rigidity, appearance (the ability to be ished in a variety of colors), ease of maintenance, prop changes with low and high extremes of tempexature, and other factor that you feel is important. Which woul preference for your particular location? Might y o u p change if you were Iocated in the dry Southwest (e.g., New England, Alaska, or Hawaii? Can you imagine means of combining materials to produce windows that he superior to any single material? Which of the features would apply to residential home siding? Automobile body panels have been made from carbonst high-strength steel, aluminum, and various polym@~-b molding compounds (both thermoplastic and thermoset). cuss the key material properties and the relative perform characteristics of each, considering both use and manufac

Range = 1 poorest to 5 best "Code = S =satisfactory U = unsatisfactory Rating chart for comparing materials for a specific application.

FIGURE 9-A

type of vehicle might youptefer thevarious materi-

might the materiak and methods of manufacture be different for these identified groups? Summarize your findings. e have been made concrete, and various metals. Discuss the key material requirements and the pros and cons of the various potential materials highly corrosive operating conditions These demanding conditions severely limit the material possihilities, and most jet engine turbine blades have been manufactured from one of the high-temperature superalloys.The fabrication processes are limited to those that are compatible with both the material and the desired geometry. Through the 1950s and early 1970s the standard method of production was investment casting, and the resultant product was a polycrystalline solid with thousands of polyhedral crystals. In the 1970s production shifted to unidirectional solidification, where elongated crystals ran the entire length of the blade. More recently, advances have enabledthepmduction ofsingle-crystalturbine blades.Investigare this product to determine how the various material and processingconditionsproduce products with differingperformance characteristics.

ility were required? sider the two-wheel bicycle frame and the variety of mate-

inum tubing (strengthened by the age-hardening ), titanium alloy,fiber-rGnforcedcomposite,and e to high temperature will compromisematerials t r e d . others, a one-piece structure (no joints) may be feasiods of manufactnre and concerns you might have. Would r solution be appropriate for high- or low-production bicyWould it be good for pleasure bikes? Rugged mountain ? Racing bikes? What would be its uniqueselling feattrres? to the Local hardware or building supply stme and exam-

able in different grades or classes based on strength or ed use? What are they and how do they differ? How

Material Selection
IS study 1 s destgned to get you to questlon why parts are made from a particular materral and how they could be fabr~cated to the~r final shape. For one or more of the products l~sted below, write a br~ef evaluat~on that addresses the following questions

H. A case for aleweler-quallty wristwatch

1. A jet englne turbine blade to operate in the exhaust reglon of the engrne J. A standard open-end wrench K. A socket-wrench socket to install and remove spark plugs .I The frame of a 10-speedb~cycie

QUESTIONS.

in terms of temperatures, loadmngs, impacts, corrosive med~a, and so on? Are there any unusual extremes? 2 What are the major properhes or oharactenstics that the material must possess In order for the product to
3. What matenal (or materials) would you suggest and

0. The outer Sk~n of an automobile muffler

P . The exterior case for a classroom projector Q. The basket sectton of a grocery store shopplng cart R. The body of a child's toy wagon S. A decorative handle for a k~tchen cablnet T . An automoblle radiator U. The motor housing for a chain saw V. The blade of a household screwdr~ver W . Household dmnerware (bde, fork, and spoon) X. The blades on a hlgh-qual~ty cutlery set Y. A shut-off valve for a %-ln. household water ilne Z. The base plate (W~th heating element) for an electnc Steam Iron AA. The front sprocket of a 10-speed bicycle BB. The load-bearingstructure of a ch~ld's outdoor swing set CC. The perforated spln tub of a washing machine DO A commemorative coin for a corporat~on's 1OGth anniversary EE. The keys for a commerc~al-quality door lock FF. The exter~or canister for an automob~le o~ filter i

4. HOW m~ght you propose to fabr~catethrs product? 5 Would the product requlre heat treatment? For what purpose? What kind of treatment? 6 Would this product requrre any surface treatment or coatlng? For what purpose? What would you
7 Would there be any concerns relatlng to environmenf? Recydlng? Product Ilabllrty?

A. The head of a carpenter's claw hammer B The exterior of an offlcefil~ng cabinet C A resident~al ~nter~or doorkrwb

F A pa~r of serssors G. A moderate to h~gh-qual~ty household cook pot or

EASUREMENT AND

INSPECTION

INTRODUCTION Llnear Measuring Instruments Attnhutes versus Variables Measuring with Lasers TANDARDS OF MEASUREMENT 10 6 VISION SYSTEMS FOIE Length Standards in Industry Standard Measuring Accuracy velsus Prec~sion m
.3 ALLOWANCE AND TOLERANCE Specifying Tolerance and

Geometr~c Toleranees INSPEC~ION METHODS FOR MEMIJREMENT Factors In Selecting Inspect~on MEASURING INSTRUMEN~S

10.16 EDDY-CURRENTTESTING 10.17 Acous~rc EMISSION TEST~NO 10.18 OTHER METHODS OF MEASUREMENT NONDES ~RUCTIVE TESTING 10.7 COORDINATE MEASURING AND INSPECTIOY MACHINES Leak Testing 10.8 ANGLE-MEASURING Thermal Methods INSTRUMENTS Strain Sensing 10.9 CAGES FOR A ~ I B W E S Advanced Opt~cal Methods MEASURING Resistivity Methods Fixed-Type Gages Computed Tomography DeviaQon-TypeGages Chemical Analysis and Surface 10 10 TESTING Topography 10.ll VISUAL INSPECTION 10.19 DORMANTVERSUS 10.12 LIQUID PENETRANT INSPECT~ON C R T ~ C FLAWS ~L 10.13 MAGNETIC PAR~CL INSPECIION E Case Study: MEASURING INSPECTION AN ANGLE 10.14 UI.TRASONIC 10.15 RADIOGRAPHY

10.1 INTRODUCTION
Meairurement, the act of measuring or being measured, is the fundamental activity of testing and inspcction.The intent of iaspectzun is to ensure that what is belngmanufactured will conform to the spwificanons of the product. Tesrragevaluatesproducl quality or performance; trying to ensure there are no defects L o impair performance as is often a f ~ r m of final inspect~on Most products are manufactured to standard slzes and shapes, For example, the base of a 60-W light bulb has been standardized so that when one bulb burns out, the next will also fit the socket in the lamp. The socket in the lamp has also been designed and made to accept the standard bulb size, Christmas tree light bulbs are made to a different standard size. Standardaation is a necessity for interchangeable parts and is also important for economlc reasons A 69-W hght bulb cannot be purchased because that is not a standard wattage. Light bulbs are manufactured only in standard wattages so that they can he mass-produced in large volumes by hi@-speed automated equipment. This resul$s ln a low unit cost. Large-scale manufacluring based onthe principles of standardization of sizes and ~ntmchangeable parts became common practice emly ln the twentieth century.Size control must be built Into machine tools and workholding devices through the precision manufacture of these machines rrnd their tooling.The output of the machines must then becheckedcare[ully (1) to determine thecapability of specificmachines and (2) for the control and maintenance of the quality of the product.A designer who specifies the dimensions and tolerances of a part often doesso to enhance the function of the product, but the deslgner is also determimng the machlnes and proce5sesneeded to make the part. Frequently, the design engineer has to altm the design or the specifications to make the product easier or less costly to manufacture, assemble, or inspect (or all of these). Designers should always be prepared to do this prov~ded that they are not sacrificing functionaltty, product reliability, or performance.

213
- --

214

"

10 "~asurement and Inspection and Testing

ATTRIBUTES VERSUS VARIABLES

1. By attributes, using gages to determine if the product is good or bad, resulting in yeslno, golno-go decision. product for comparison with the size desired.

accepted industrial term for inspection by variables. Gaging (or gauging) is the term determining whether the dimension or characteristic is larger or smaller than established standard or is within some range of acceptability.Variabletypes of inspec generally take more time and are more expensive than attribute inspection, butt yield more information because the magnitude of the characteristic is known in so standard unit of measurement.

1 1 10.2 STANDARDS OF MEASUREMENT

with the ampere and the candela, provide the basis for all other units of measureme as shown in Figure 10-1, along with the original definitions. Most mechanical measu ments involve combinations of units of mass, length, and time.Thus, the newton, a of force, is derived from Newton's second law of motion (f = ma) and is defined as force that gives an acceleration of lmlseclsec to a mass of 1kilogram. Figure 10-2 Table 10-2 provide basic metric-to-English conversions.

LINEAR STANDARDS When people first sought a unit of length, they adopted parts of the human body, the hands, arms, or feet. Such tools were not very satisfactory because they we universally standard in size. Satisfactory measurement and gaging must be bas

Time

Second

Electric curre..,

Ampere

Present: Mass of Prototype Kilogram No. 1 kept at International Bureau of Weights and Measures at S&vre$ France. Originai: 1186,400tof mean solar day. Present: 9,192,631,770cycles of frequency associated with transition between two hyperfine levels of isotope cesium-133. Present: The rate of motion of charge in a circuit is callzd the current.?he unit of current is the ampere. One ampere exists when the charge flaws at a rate of 1 coulomb per second.

Amount of substance Luminous intensity

Candle

of molecules. Present: One lumen per square foot is a footcandle.

I I

SI units

I
I

Base SI units

Derived Si units w ~ t h seec~ai names

indicate
multiplication

Dashed lines indicate

FIGURE 10-1 Relationship of secondary physical quantities to basic 5 1 units. Solid lines signify

multiplication; dashed lines signify division. reliable, preferably universal, standards. These have not always existed. For example, although the musket parts made in Eli Whitney's shop were interchangeable, they were not interchangeable with parts made by another contemporary gunmaker from the same drawings because the two gunsmiths had diffrent foot rulers. Today, the entire industrialized world has adopted the international meter as the standard of linear measurement. The inch, used by both the United States and Great Britain, has been defined ~[licially as 2.54 ccntimeters. Thc U.S. standard inch is 41,929.399 wavelengths of the orange-red light from krypton-86. Although officially the United States is committed to conversion to the metric (SI) system of measurement, which uses millimeters for virtually all linear measurements in

218

CHAPTER 10 Measurement and Inspection and Testing

What's in the bax

9 Blooks 49 Blocks '1a B(ocks d Blocks

0.1007 0.101 C050 1.000

through 0.1009 in. through 0.149 in. through 0.950 in, mrough 4.000 in.

,?

in steps of 0.0001 in, in steps of 0.001 in, in steps o f 0.050 in. i dspsof 1.000 in.

+-Lamed and mirror poibhed to 'a very low mioro-suriaw finkh.

FIGURE 10-3 Standard set of rectangular gage blocks with 0.000050-in. accuracs three individual blocks are shown.

FIGURE 10-4 Seven gage blocks wrung together to build up a desired , dimension. (Courtesy of DoALL ,Compam/.)

Grade 0.5 (grand-master) blocks arc used as a basic refzrencc standard in caiibration laboratories. Grade 1 (laboratory-grade) blocks arc used for checking and cala r ibrating other grades of gagc hlocks. Grade 2 (precision-grade) blocks are used E checking Grade 3 hlocks and master gages. Grade 1 (B or working-grade) blocks are used to calibrate or check routine measuring devices, such as micrometers, or in actual gaging operations. The dimensions of individual blocks are established by light-beam interferometry with which it is possible to calibrate thcse blocks routinely with an uncertainty as low Gageblocks as one part per million. Gage blocks i~sually come in sets containing various numbers of blocks of various sizes, such as those shown in Figure 10-3. By .'wringing the hlocks together" in various combinations. as shown in Figure 10-4, any desired dimension can be obtained. For cxample, if the last two hlocks on the stack arc 0.100 and 0.05 in.. what is the total lcngth o l the wrung-togcther stack of gagc blocks'! Gage blocks are wrung together by sliding one past another using hand pressure. They will adhere to one another with considerable force and must not be lelt in contact for extended periods of time. Gage blocks are available in dilferent shapes (squares Wrung-together angles, rounds, and pins),so standards of high accuracy can be obtained to fill almost any gage blocks in a special holder, need. In addition, various auxiliary clamping, scribing, and base block attach~nents are, used with a dial gage to form a n accurate comparator. (Courtesy of available that make it possible to form very accurate gaging devices, such as the setup DoALL Company) shown in Figure 10-5.
Dial indicator
ge

Because all the commonly used metals are affected dimensionally by temperature, a standard measuring temperature of 68F (20C) has been adopted for precision measuring work. All gage blocks, gages, and other precision-measuringinstruments are calibrated at this temperature. Consequently, when measurements are to be made to accuracies greater than 0.0001 in. (0.0025 mm), the work should be done in a room in which the'temperature is controlled at standard.Although it is true that to some extent both the workpiece and the measuring or gaging device may be affected lo about the same extent by temperature variations, one should not rely on this. Measurements to even 0.0001 in. (0.0025 mm) should not be relied on if the temperature is very far from 68F (20C).

ACCURACY VERSUS PRECISION I N PROCESSES It is vitally important that the difference between accuracy and precision be understood. Accuracy refers to the ability to hit what is aimed at (the bull's-eye of the target). Precision refers to the repeatability of the process. Suppose that five sets of five shots are fired at a target from the same gun. Figure 10-6 shows some of the possible outcomes. In Figure 10-6a,inspection of the target shows that this is a good process-accurate and precise. Figure 10-6b shows precision (repeatability) but poor accuracy. The agreement with a standard is not good. In Figure 10-6c the process is on the average quite accurate, as the X (average) is right in the middle of the bull's-eye, but the process has too much scatter or variability; it does not repeat. Finally, in Figure 10-6d, a failure to repeat accuracy between samples with respect to time is observed; the process is not stable. These four outcomes are typical but not all-inclusive of what may be observed. In Chapter 36, more discussion on accuracy and precision is presented as they relate to process capability. This term is used to describe how well a manufacturing process performs in its part making. In measuring instruments used in the factory, precision (or repeatability) is critical because the devices must be very repeatable as well as accurate. For manually operated instruments, the skill of the operator must also be considered-this is called reproducibility. So accuracy, repeatability- and reproducibility are characteristics of what is called gage cupabilily, whichis also discussed in Chapter 36.

la1 Accurate and

(b)

precise

Prec~se, not accurate

RE 10-6 Accuracy versus iion. Dots in targets sent location of shots. Cross :presents the location of verage position of all shots. :curate and precise; recise, not accurate; :curate, not precise; ~cise within sample, not ie between sampies, not ate overall or within le.

(c! Accurate, not precise

(d!

Prec~se w~th~ sample n Not preclse between samples Not accurate overall or wlthin sample

IND TOLERANCE
If the desired fit between mating parts is to be obtained, the designer,fnust spec [actors, allowance and tolerance. Allowance is the intentional. desiIjkd difference tween the dimensions of two mating parts. It is thedifference between the dirnensio~ the largest interior-fitting part (shaft) and that of the smallest exterior-fitting part (hole Figure 10-7 shows shaft A designed to fit into the hole in block B. This differen (0.5035-0.5025) thus determines the condition of tightest fit between mating pa Allowance may be specified so that either clearance or interference exists between t mating parts. In the case of a shaft and mating hole, it is the difference in diameters the largest shaft and the smallest hole. With clearance fits, the largest shaft is small than the smallest hole, whereas with interference fits, the hole is smaller than t Tolerance is an undesirable but permissible deviation from a desired dimen There is variation in all processes, and no part can be made exactly lo a speci dimension, except by chance. Furthermore, such exactness is neither necessary n economical. Consequently,it is necessary to permit the actual dimension to deviate fr the desired theoretical dimension (called the nominal) and to control the degree deviation so that satisfactor'yfunctioning of the mating parts will still he ensured. Now we can see that the objective of inspection, by means of measurement niques, is to provide feedback infurmation on the actual size of the parts with refer to the size specified by the designer on the part drawing. The manul'acturing processes that make the shah are different from those make the hole, but both the hole and the shalt are subject to deviations in size beca of variability in the processes and the materials.Thus, while the designer wishes ide ly that all the shafts would be exactly 0.500 (Figure 10-8a) and all the holes 0.506,t reality or processing is that there will be deviations in size around these nominal ideal sizes. Most manufacturing processes result in products whose measurements of geometrical features and sizes are distributed normally (Figure 10-8b). That is, m of the (0.0535-0.5025) measurements arc clustered around the average dimension, calculaled as

z=n
-

i il x i

for n items

X will be equal to the nominal dimension only if the process is 100% accurate, that perrcctly centered. More likely, parts will be distributed on either side of the average, the process might be described (modeled) with a normal distribution. In normal dis butions, as shown inFigure 10-8c,99.73% of the measurements (Xi) will fall wit or minus 3 standard deviations ( 1 3 ~ of ) the mean, 95.46% will be within i 2 u , 68.26% will be within i l u where

lT =

F
i= 1

0.506 i 0.0025 0.500 -' 0.0025

Block B

Hole in

FIGURE 10-7 When mathg parts are designed, each shaft must be smaller than each hole for a clearance fit.

Ideal

I
unlikely

Real World

Model of Real World

Spread
+ 0 c
0 ,-

i.

5 a

Pans are normally distributed about

5 n
L

LL

-3c

2 -[

-; : f ;qu
99.73%

+3c

111~ proces, vo.rlo irianr .]I pant Chart ) the came s ze (b) In tne udrL) o f ,Ize, cdn uftcri ue iiiuJclrd real worla of m a n ~ f d c t ~ r ~ n u . n3\e iar anll~f\n w e (r) The n crr~h~tlsn w~th a normal d~stribut~on.

FIGURE 10-8 (a) n rne aea s tdtlori.

In summary, the designer applies nominal values to the mating parts according to the desired fit between the parts.Tolerances are added to those nominal values in recognition of the fact that all processes have some natural amount of variability. Assume that the data for both the hole and the shaft are normally distributed. The i3u added to the mean ( p ) gives the upper and lower natural rvlerunce limits,defined as
p

+ 3u = U N T L
LNTL

p - 3u =

As shown in Figure 10-9a, the average fit of two mating parts is equal to the difference between the mean of the shaft distribution and the mean of the hole distribution. The range o f f i t would be the difference between ihe minimum diameter shaft and the maximum diameter hole.The minimum clearance would be the difference between the

10 Measurement and Inspection and Testin0

Smallest

,500

(at

FIGURE 10-9 (a) The

manner in which the distributions of the two mating parts interact determines the fit. UNTL, (upper natural tolerance limit) = f i + 3u;LNTL, lower natural tolerance limit = f i - 3 0 (b) Shifting the means o f the distributions toward each other results in some interface fits.

-3

-2

-1

the process), the better the fit between the parts.

each other, and the fit will become increasingly tight (the clearance will be

and the fit will become tighter, eventually becoming an interference fit. The designer must speclfy the tolerances accordmg to the function of th

SPECIFYING TOLERANCE AND ALLOWANCES Tolerance can be specified in four ways: bilateral, unilateral, limits, and geometric
Bilateral tolerance is specified as a plus or minus deviabon from the nominal size,s

Secr~o 10.3 ~ Allowance and Tolerance

223

as 2.000 i 0.002 in. More modern practice uses the unilateml3~ptem, where the deviation is in one direction from the basic size, such as +0.1 mm 50.8 mm 0 . 0 mm or 2.000 in. -0.000 in. +0.004 in. in metric

In the first case, that of bilateral tolerance, the dimension of the part could vary between 1.998 and 2.002 in., a total tolerance of 0.004 in.For the example of unilateral tolerance, the dimension could vary between 2.000 and 2.004 in., again a tolerance of 0.004 in. Obviously, to obtain the same maximum and minimum dimensions with the two systems, different basic sues must be used. The maximum and minimum dimensions that result from application of the designated tolerance are called limit dimensions, or limits. (Geometvic tolerances are discussed in the next section.) There can be no rigid rules for the amount of clearance that should he provided between mating parts; the decision must be made by the designer, who considers how the parts are to function. The American National Standards Institute, Inc. (ANSI) has established eight classesof fits that serve as a useful guide in specifying the allowance and tolerance for typical applications and that permit the amount of allowance and tolerance to he determined merely by specifying a particular class of fit.These classes are as follows: : Loose fit: large allowance. Accuracy is not essential. Class 1 Class 2: Freefit: liberal allowance. For running fits where speeds are above 600 rpm and pressures are 4.1 MPa (600 psi) or above. Class 3: Medium fit: medium allowance. For running fits under 600 RPM and pressures less than 4.1 MPa (600 psi) and for sliding fits. Class 4: Snugfit: zero allowance. No movement under load is intended, and no shaking is wanted. This is the tightest fit that can be assembled by hand. Class 5: Wringingfit: zero to negative allowance.Assemblies are selective and not interchangeable. Class 6: Tightfit: slight negative allowance. An interference fit for parts that must not come apart in service and are not to be disassembled or are to be disassembled only seldomly. Light pressure is required Ior assembly. Not to be used to withstand other than very light loads. Class 7: Medium force fit: an interference fit requiring considerable pressure to assemble; ordinarily assembled by heating the external member or cooling the internal member to provide expansion or shrinkage. Used for fastening wheels, crank disks, and the like to shafting.The tightest fit that should be used on cast iron external members. Class 8: Heavy force and shrinkfits: considerable negative allowance. Used for permanent shrink fits on steel members. The allowances and tolerances that are associated with the ANSI classes of fits are determined according to the tl~eoretical relationship shown in Table 10-4. The actual resulting dimensional values for a wide range of basic sizes can be found in tabulations in drafting and machine design books.

Class
of Fit

Allowance
0o o 2 s w 0 0014'@ 0 0009@ 0

Average Interference

Hole

Tolcrance

Shaft Tulcrance

In the ANSI system, the hole size is always considered basic, because the majon ty of holes are produced through the use of standard-dze dr~lls and reamers.The inter nal member, the shaft, can be made to any one dimension as readily as to another Tn allowance and tolerances are applied to the basic hole slze to determ~ne the limit mensions of the mating parts. For example, for a basic hole size of 2 m. and a Class 3 the dimensions would be: Allowance Tolerance Hole Maximum Minimum Shaft Maximum Minimum 0.0014 in. 0.0010 in. 2.0010 in. 2.0000 in. 1.9986 in. 1.9976 in.

It should be noted that for both clearance and interference fits,the permrss~ble toleranc tend to result In a looser fit. The I S 0 System oftrmits and Frts (Figure 10-10a) is wdely used in a number leading metric countries. This system is considerably more complex than the ANS system just discussed In this system each part has a basic size. Each limit of size of part, high and low, is defined by its deviation from the basic size, the magnitude an sign being obtained by subtracting the basic size from the limit m question. The ference between the two l ~ m ~of t ssize of a part is called tolerance, an absolute amo without sign. There are three classes of fits: (1) clearance fits, (2) tramition fits (the assem may have erther clearance or interference), and (3) interference fits. Either a shafthole-basis system may be used (Figure 10-lob). For any given basic slze, a range of to1 erances and deviations may be specified withrespect to the hne of zero deviation, caUe the zero line.The tolerance IS a function of the basic s ~ z and e is designated by a numb symbo1,called thegrade (e.g., the talerancegrade).Theposctron of the tolerance withre spect to the zero h e , also afunctlon of the basic size,ls indicated by a letter symbol two letters)-a capital letter for holes and a lowercase letter for shafts-as illustra in figure 10-10c.'Thus the specification for a hole and a shaft hav~ng a basic size of mm might be 45 H8ig7 E~ghteen standard grades of tolerances are prov~ded, called IT 01, IT 0, and IT through IT 16,providing numencalvalues for each nommal diameter,in arbitrary st up to 500 mm (i e,,0-3,3-6,6-10,. .,400-500 mm) The valve of the tolerance unlt, r, grades 5-16 would be
2 =

0.45fi

+ 0.001 D

where 1 ism micrometers and D 1n millimeters. Standard shaft and hole deviations are provided by s~milar sets of formul However, for pract~cal application, both tolerances and deviat~ons are provlded three sets of rather complex tables. Additional tables give the values for basic siz above 500 mm and for "commonly used shafts and holes" m two categories: "gene purpose" and "fine mechanisms and horology" (horology 1s the art of maki timepieces).

GEOMETRIC TOLERANCES Geometr~c tolera~lces state the maximum allowable devlat~on of a form or a positio from the perfect geometry ~mplied by a drawmg.These tolerances specify the diam ter or the width of a tolerance zone necessary for a part to meet its required accura
'I1 will be recognized that the "posnron" m the IS0 system essenually provides theUallowance"of the AN system

SECTION 10.3
r c 0 .0 .m .a .-

Allowance and Tolerance

225

...
D

= $ L ~ T -

0 .0 .m .m .-

..
K

m Lo en 2

line or line ro deviation

g
Hole

.c 5

5
Shaft

,7 .

La) Basic size, deviation and tolerance in the IS0 system.

Clearance fit

Transition fits interference fit Clearance fit Basic size

Transition fits

Interference fit

Examples taken from the shaft-basis system

Examples taken from the hole-basis system

(b) Shaft-bas~s and hole-bas~s system for specifying fits in the IS0 system.

Negative deviations Positive dev~ations Negative devlations

I Positive deviations

Basic size = -

2 -. -

I-

Basic size

(c) Pos~tion of the varlous tolerance zones for a given d~arneter in the I S 0 system.

FIGURE 10-10 The I S 0 System of Limits and Fits. (Bypermission from Recommendations l . ) R286-7962, System of Limits and Fits, copyright 7962, American Standards Institute, N

Datum surface
(bl

Notes

0
A

0
MMC LMC

0
RFS

Tolerance specification
0.50 tolerance zone

FIGURE 10-11 (a) Geometric tolerancina svmbols; (b) feature controlsymbols for part drawingsf (c) how a

qeometric to.erdlce for flatness :s spec,f;eo; (d) what the cpecification means.

Interpretation
Id1

Figure 10-11shows the various symbols used to specify the required geometric ch acteristics of dimensioned drawings. A modifier is used to specify the limits of size a part when apply~ng geometric tolerances. The maximum material condition (M indicates that apart is made with the largest amount of material allowable (e.g.,ah at its smallest permitted diameter or a shaft at its largest permitted diameter).Thel material condition (LMC) is the converse of the maximum material condit~on. gardlrss offeature size (RFS) indicates that tolerances apply to a geometric feature any size it may be. Many geometric tolerances or feature control symbols are stat with respect to a particular datum or reference surface. Up to three datum su can be given to specify a tolerance. Datum surfaces are generally designated by a symbol. Figure 10 l l b gives examples of the symbols used for datum planes and feat control symbols There are four tolerances that specify the permitted variability of forms:fla straightness, ruundness, and cylindriciry. Form tolerances describe how an actual f may vary from a geometrically ideal feature. The surface of a part is ideally flat if all its elements are coplanar.Theflatness ification describes the tolerance zone formed by two parallel planes that bound a elements on a surface.A 0.5-mm tolerance zone is described by the feature control bol in Figure 10-1lc.The distance between the highest point on the surface to the est point on the surface may not be greater than 0.5 mm, as shown in Figure 10-11 In Section 10.7, on coordinate measuring machines, additional examples geometric tolerances can be found.

The Eield of metrology, even limited to geometrical or dimensional measurements, is far toolarge to cover here.This chap.ter concentrates on basic linear measurements and the measurement and testing devices most commonly found in a company's metrology or quality control facility.At a minimum, such labs would typically contain optical flats: one or two granitemeasuring tables: an assortment of indicators, calipers, micrometers, and height gages; an optical comparator: a set or two of Grade 1gage blocks: a coordinate measuring machine; a laser scanning device; a laser interferometer; a toolmaker's microscope;and pieces of equipmentspecially designed to inspect and test the company's products. Table 10-5 provides a summary of inspection methods, listing five basic kinds of devices: air, light optical andelectron optical, electronic, and mechanical. The variety seems to be endless, but digital electronic readouts connected to any of the measuring devices are becoming the preferred method. The discrete digital readout on-a clear liquid-crystal display (LCD) eliminates reading interpretations associated with analog scales and can be entered directly into dedicated microprocessors or computers for permanent recording and analysis. The added speed and ease of use for this type of equipment have allowed it to be routinely used on the plant floor instead of in the metrology lab. In summary, the trend toward tighter tolerances (greater precision) and accuracy associated with the need for superior quality and reliability has greatly enhanced the need for improved measurement methods.

0.5-10 pin. or 2 to 3% of scale range

0.2-2 pin. or better with laser interferometry 0.5-1 second o f a r c i n autocollimation optical comparators

classifying. Interferometry; checkingflatnws and size of gage bloeks:finding surface flaws; measuring spherical shapes, flatnessof surface plates accuracy of rotary index tables; includes all light microscopes and devices common on plant floor Precision measurement in scanning electron microscopes of microelectronic circuits and other small precision parts. Widely used for machine control, on-line inspectian, sorting, and classification:ODs. IDS,height, surface, and geometrical relationships, profile tracing for roundness, surface roughness, contours, etc; most devices are omparators with movement of stylus or spindle producing an electronic signal that is amplified electronically, commonly connected 14 microprocessors and minicomputers for process adjustment. Large variety ofexternal and internal rnea3urements using dial indicators. micrometers calipers, and the like: commonly used for bench comparators for gage calibration work.

for delicate psrts. Iargest variety of measuring equipment: autocollimators are used for making precision angular measurements; lasers are used to make presisian in-process measurements; laser scanning. Part size restricted by vacuum chamber sue; electron beam can be used for processing and part testing of electronic circuits. Electronic gages conic in many farms but usually have a sensory head or detector comhined with an amplifier; capable of high magnification with resolution limited by size or geometry For sensory head:readouts commonly have multiple magnification steps; solid-state digital electronics make these devices small, portable, stable, and extremely flexible, with extremely fast response time. Moderate cost and ease of use make many of these devices and workhosses of the shop floor; highly dependent on workers'skills and often subject lo problems of linkages.

228

C H A ~ 10 R Measurement and lnspection and Testing

FACTORS IN SELECTING INSPECTION EQUIPMENT Many inspection devices use electronic output to communicate directly to micr processors, Inspection devices are being built into the processes themselves and often computer-aided. In-process inspection generates feedback sensory data from process or its output to the computer control of the machine, which is the first step making the processes responsive to changes (adaptive control). In addition to in-proce inspection, many other quality checks and measurements of parts and assemblies needed. In general, six factors should be considered when selecting equipment for inspection job by measurement techniques.

2. Linearity.This factor refers to the calibration accuracy of the device over its full ing range. Is it linear? What is its degree of nonlinearity? Where does it becom linear, and what, therefore, is its real linear working region? 3. Repeat accuracy, How repeatable is the device in taking the same reading over over on a given standard?

over the actual input dirnension.The more accurate the device, the greater must its magnification factor, so that the required measurement can be read out (or served) and compared with the desired standard. Magnification is often onfus with resolution, but they are not the same thing. or dimensional input that the device can detect or distinguish.The greater the lution of the device, the smaller will be the things it can detect or identify (res

dle versus the size and geometry of the workpieces; the environment; the cost o

Tolerance needed on part r 0.001 on hole diameter

To check and set

FIGURE 10-12 The rule of

10 states that for reliable

Precision needed on gage i 0.0001 in.

a standard of

0,00001 in, is needed

measurements each successive step in the inspection sequence should have 10 times the preci,~ionof the preceding step.

Workpiece

Air gage or
working gage

Master gage standard

SECTION 10.5 Measuring Instruments

229

device; and the cost of installing, training, and using the device.The last factor depends on the speed of measurement, the degree to which the system can be automated,and the functional life of the device in service.

ISTRUMENTS
Because of the great importance of measuring in manufacturing, a great variety of instruments are available that permit measurements to be made routinely, ranging in accuracy from & to 0.00001 in. and from 0.5 to 0.0003 mm. Machine-mounted measuring devices (probes and lasers) for automatically inspecting the workpiece during manufacturing are beginning to compete with post-process gaging and inspection, in which the part is inspected, automatically or manually, after it has come off the machine. In-process inspection for automatic size control has been used for some years in grinding to compensate for the relatively rapid wear of the grinding wheel. Touch trigger probes, with built-in automatic measuring systems, are being used on CNC machine tools to determine cutting tool offsets and compensations for tool wear.These systems are discussed in Chapter 26. For manually operated analog instruments, the ease of use, precision, and accuracy of measurements can be affected by (1)the least count of the subdivisions on the instrument, (2) line matching, and (3) the parallax in reading the instrument. Elastic deformation of the instrument and workpiece and temperature effects must be considered. Some instruments are more subject to these factors than others. In addition, the skill of the person making the measurements is very important. Digital readout devices in measuring instruments lessen or eliminate the effect of most of these factors, simplify many measuring problems, and lessen the chance of making a math error.

LINEAR MEASURING INSTRUMENTS Linear measuring instruments are of two types: direct reading and indirect reading. Direct-readinc instruments contain a line-graduated scale so that the size of the object being measured can be read directly on this scale. Indirect-reading instruments do not contain line graduations and are used to transfer the size of the dimension being measured to a direct-reading scale, thus obtaining the desired size information indirectly. The simplest and most common direct-reading linear measuring instrument is the machinist's rule, shown in Figure 10-13.Metric rules usually have two sets of line graduations on each side, with divisions of and 1mm; English rules have four sets, with divisions of 11 ,, 1 and in. Other combinations can be obtained in each type. The machinist's rule is an end- or line-matching device. For the desired reading to be obtained, an end and a line, or two lines, must be aligned with the extremities of the object or the distance being measured. Thus, the accuracy of the resulting reading is a

,, ,,, &

13 Machinist's tric and (b) inch

10th~ and 100th~ o 32nds and 6 4 t h ~ side. (Courtesy of

Cornpony)

FIGURE 10-14 Combination set. (Courtesy of MTI

Corporation.)

FIGURE 10-15 This vernier caliper can make measurements

using both inside (for holes) and outside (shafts) anvils. function of the alignment and the magnitude of the smallest scale division. Such s are not ordinarily used for accuracies greater than , & in. (0.01 in.), or about f mm. Several attachments can be added to a machinist's rule to extend its usefulness squnre head (Figure 10-14) upright position on a flat surface for making height measurements. It also con small bubble-type level so that it can be used by itself as a level.The bevelprotracto mits the measurement or layout of angles The center head permits the center of drical work to be determined. The vernier caliper (Figure 10-15) is an end-measuring instrument, av various sizes that can be used to make both outside and inside measurement retical accuracies of 0.03 mm or 0.001 in. End-measuring instruments are more ratc and somewhat easier to use than lie-matching types because their jaws are pl against either end of the object being measured, so any difficulty in aligning edges or is avoided. However, the difficulty remains in obtaining uniform contact pressur "feel," between the legs of the instrument and the object being measured. A major feature of the vernier caliper is the auxiliary scale (Figure 10-16 caliper shown has a graduated beam with a metric scale on the top, a metric v plate, and an English scale on the bottom with an English vernier.The manner in readings are made is explained in the figure. Figure 10-17 also shows a vernier depth gage for measuring the depth of holes o length of shoulders on parts and a vernier height gage for rmaking height measurem Figure 10-18 shows calipers that have a dial indicator or a digital readout that rep vernier.The latter two calipers are capable of making inside and outside measu as well as depth measurements. The micrometer caliper, more commoniy called a micrometer, is one of th widely used measuring devices. Until recently, the type shown in Figure 10-19 tually standard. It consists of a fixed anvil and a movable spindle. When the t is rotated on the end of the caliper, the spindle is moved away from the anvil by of an accurate screw thread. On English types, this thread has a lead of 0.025 in. one revolution of the thimble moves the spindle this distance.The barrel, or sb calibrated in 0.025-in. divisions, with each $ of an inch being numbered. The ci ference at the edge of the thimble is graduated into 25 divisions, each represen 0.001 in. A major difficulty with this type of micrometer is making the reading dimension shown on the instrument. To read the instrument, the division on th ble that coincides with the longitudinal line on the barrel is added to the largest exposed on the barrel.

SECTION 10.5

Measur~ng Instruments

231

10-16 Vernier raduated for English ric (direct) reading. ric reading is
1 2 = 27.42 mm.

Refer to the upper bar graduations and metric vernier plate. Each bar graduation is 1.00 mm. Every tenth graduation is numbered in sequence-10 rnm, 20 mm, 30 mm, 40 rnrn, etc.-overthe full range of the bar.This providesfor direct reading in millimeters. Metric vernier

Beam

vernler plate In the picture, the vernier plate zero line is one inch (1.000") plus one-twentieth (0.50") beyond thezera line on the bar, or 1.050". The 29th graduation on the vernier plate coincides with a line on the bar (as indicated by stars). 29 x 0.001 (.029") is therefore added to the 1.050" bar reading, and the total is 1.079".

Vernier height gage

FIGURE 10-17 Varlat~onsin the vernler callper d e s ~ g n result in other basic gages

Outside jaws only

I
Vernler callper wlth Inch or metrlc scales and 0.001-ln. accuracy

0 001-ln accuracy Dial callper w ~ t h

FlGURE 10-18 Three styles of

calipers in common use today: (a) Vernier caliper with inch or metricscales and 0.001-in. accuracy; (b) dial caliper with 0.001-in. accuracy; (c) digital electronic caliper with 0.001-in. (0.03-mm) accuracy and 0.0001 -in. resolution with inch/metrlc conversion.

Digital electronic caliper with O.QO1-in. (0.03-mmlaccuracy and 0.0001-in. resolution with inchimetric conversion.

Measuring faces

Anv~l

Lock nut Sleeve Spindle

Th~mble

Ratchet stop

FIGURE 10-19 Mlcrorneter

caliper graduated In tenthousandths of an ~ n c h w~th Insets A, 0, and C showlng two example readings. (Courtesy Storrett Builetm No 1203 )

Micrometers graduated in ten-thousandths of an inch are the same as those graduated in thousandths, except that an additional vernier scale is placed on the sleeve so that a reading of ten-thousandths is obtained and added to the thousandths reading. The vernier consists of 10 divisions on the sleeve, shown in B, which occupy the same space as 9 divisions on the thimble.Tnerefore, the difference between the width of one of the 10 spaces on the vernier and one of the 9 spaces on the thimble is one-tenth of a division on the thimble?or one-tenth of one-thousandth, which is one ten-thousandth. To read a ten-thousandths micrometer, first obtain the thousandths reading, then see which of the lines on the vernier coincides with a line on the thimble. If it is the line marked "1," add one ten-thousandth; if it is the line marked "2," add two ten-thousandths: and so on.

EXAMPLE: REFER TO INSETS A AND B IN FIGURE 10-19.

The "2" line on sleeve is visible, representing Two additional lines, each representing 0.025" Line "0" on the thimble coincides with the reading line on the sleeve, representing The "0" lines on the vernler coincide w ~ t h lines on the thimble, representing The micrometer reading is Now you try to read Inset C The "2" llne on sleeve 1s visible, representing Two addit~onal llnes, each representing 0.025" The reading line on the sleeve lies between the " 0 " and "1"on the thimble, so ten-thousandths of an inch is to be added as read from the vernier. The "7" line on the vernier coincides with a line on the thimble, representing The micrometer readmg is 0.200 in. 2 X .025" 0.000 in. 0.000 in. 0.2500 in 0.200 in. 2 X 0.025' 0 . 0 5 0 in.

= 0.050 in.

7 X 0.0001" = 0.007 in. 0.2507 in.

However. owing to the lack of pressure control, micrometers can seldom be relied on for accuracy beyond 0.0005 in., and such vernier scales are not used extensively. On metric micrometers the graduations on the sleeve and thimble are usually 0.5 mm and 0.01 mm,respectively (see the Problems at the end of the chapter). Many errors have resulted from the ordinary micrometer being misread, the error being i0.025 or 20.5 mm. Consequently, direct-reading micrometers have been developed.Figure 10-20 shows a digital outsidemicrometer that reads to 0.001 in. on the digit counter and 0.0001 in. on the vernier on the sleeve.The range of a micrometer is limited to 1 in.Thus a number of micrometers of various sizes are required to cover a wide range of dimensions.Tocontrol the pressure between the anvil, the spindle, and the piece being measured, most micrometers are equipped with a ratchet or a friction device, as shown in Figures 10-19 and 10-20. Calipers that do not have this device may be overtightened and sprung by several thousandths by applying excess torque to the thimble.Micrometer

E 10-20 Dig~tal

neter for measurements lta 1 in., in 0.0001-In itions

SECTION 10.5 Measuring Instruments

235

E 10-22 Toolmaker's microscope w~th d~gital ~tsfor X and Y table movements.

?syof Nrkon )

FIGURE 10-23 Opt.cal comparator, measuring the contow on a vvordplere. LYy:t~l :ndicalor, nith ronvers onc add l o (he btility of optica. rornpararors.

curate measurements on small or dehcate parts.The base, on which the special microscope is mounted, has a table that can be moved in two mutually perpendicular, horizontal directions (X and Y ) by means of accurate micrometer screws that can be read to U.BOQ1 in. or,if so equipped, by means of the digital readout. Parts to be measured are mounted on the rable, the microscope is focused, and one end of the desired part feature i s aligned with the cross-line in the microscope.The reading is then noted and the table is moved until the other extremity of the part coincides with the cross-line. From the final reading, the desired measurement can be determmed. In addition to a wide variety of linear measurements, accurate angular measurements can also be made by means of a special protractor eyepiece.These microscopes are available with digital readouts. The optical projscfor or comparator (Figure 10-231is a large optical device on which both linear and angular measurements can be made. As with the toolmaker's microscope, the part to be measured is mounted on a table that can be moved in X and Y d~rections by accuratemicrometer screws The optical system projects the image of the part on a screen, magnifying st from 5 to more than 100 times. Measmments can be made directly by means of the micrometer dials, the digital readouts, or the d i a l indiiators, or on the magnified image on the screen by means of an accurate rule A very common use for this type of instrument 1s the checking of parts,such as dies and screws. A template is drawn to an enlarged scaleand is placed on the screen.Theprojected~mtour of the part is compared to the desired contour on the screen. Some projectors also function as low-power microscopes by providimg surface illumination.

MEASURING WITH LASERS

One of the earliest and most common metrological uses of low-power lasers has been in interferametry.The interferometer uses light interference bands to determine distance and thickness of objects (Figure 10-24).First, a beam splitter divides a beam of light Into a measurement beam and a reference beam. The measurement beam travels 6 a reflector (optical glass plate A), resting on the part whose distance is to he measured, wh~le the reference beam is directed at fixed reflectorB. Both beams arc reflected back through the beam splitter,where they are recombiued into a s~ngle beam hefore traveliag

Fringes as seen b y observe1

partially reflecting plate

FIGURE 10-24 Interference

bands can be used to measure the size of objects to great accuracy. (Based on the Micheison in 1882.) ~nterferorneter, ~nvented

Object to

be measured

Reflector B

FIGURE 10-25

Method of calibrating gage block by light-wave interference.

to the observer. T h ~ recombined s beam produces ~nterfer whether the waves of the two returning beams are in phase (called constructzv ference) or out of phase (termed destructwe mterference) series of br~ght bands, and out-of-phase waves produce dark bands. The n fnnges can be related to the slze of the object, measured in terms of light waves of a frequency.The following example will explain the basics of the method. To determine the slze of object U in Figure 10-25, a calibrated reference stan S, an optical flat, and a toolma source Optrcalf[ats are quartz diameter and about to 1~ n(12 . planes and nearly parallel. Flats (0.00003 mm) of true flatness. It is not essential that both surfaces be accurate o they be exactly parallel, but one must be certain that only the accurate surface 1s m maklng measurements. A toolmaker'sflat is similar to an optical flat but is ma steel and usually has only one surface that IS accurate.Amonochromatic l~ghtsource, of a single wavelength,must be used. Selenium, hehum, o commonly used along \nth hehum-neon lasers. The block to be measured is U and the calibrated block is S. Distances a must be known but do not have number of mterference bands sh determined. Because the difference in the distances be surface of U is one-half wavelength, each dark band wavelength m the elevation. If a monochromatic light source having a wave 23.2 pin. (0 589 mrn) IS used, each Interference band represents 11.6 pin (0. Then, by simple geometry, the difference in the helghts of the two blocks can puted.The same method 1s appli~able for maklng precis by comparing them with a known gage block. Accurate measurement o untll the development of laser per million over a distance of 6 1 m with 0.01 pm resolution. Such equpment is ularly useful in checking the movement of machine large assembly jigs, and making measurements of lntrlcate machined parts such as tread molds.

The Hewlett-Packard laser interferometer (Figure 10-26) uses a helium-neon laser beam split into two beams, each of different frequency and polarized. When the beams are recombined, any relative motion between the optics creates a Doppler shift in the frequency.This shift is then converted into a distance measurement.The laser light has less tendency to diverge (spread out) and is also monochromatic (of the same wavelength). A process that has been largely confined to the optical industry and the metrology lab is now suitable for the factory, where its extremely precise distance-measuring capabilities have been applied to the alignment and calibration of machine tools. The company's first two-frequency interferometer calibration system was introduced in 1970 to overcome workplace contamination by thermal gradients, air turbulence, oil mist: and so on, which affect the intensity of light. Doppler laser interferometers are relatively insensitive to such problems.The system can be used L o meas u e linear distances, velocities, angles, flatness, straightness,squareness, and parallelism in machine tools. Lasers provide for accurate machine tool alignment. Large. modern machine tools can move out of alignment in a matter of months, causing production problems often attributed to the cutting tools, the workholders, the machining conditions, or the numerical control part program.

etroreflector shown with he~ght adjuster, post and base

Retroreflector Twofrequency lase1

3.6

_f

3.1 MHz

and triggers

238

CHAFER10

Measurement and Inspection and Testing Direction of scan

Photodetec

Diameter of object

Recelver

FIGURE 10-27 Scann~ng laser measuring system. (Courtesy ofZYGO Corporat~on.)

Light interference also makes it possible to determine easily whether a surfact exactly flat.The achievement of interference fringes is largely dependent on the coh ence of the light used.The availability of highly coherent laser light (in-phase light o single frequency) has made interferometry practical in far less restrictive environme~ than in the past. The sometimes arduous task of extracting usable data from a clo: packed series of interference fringes has been taken over by microprocessors. The most widely used laser technique for inspection and in-process gaging known as laser scanning. At its most basic level, the process consists of placing an c ject between the source of the laser beam and a receiver containing a photodioi A microprocessor then computes the object's dimensions based on the shadow $ the object casts (Figure 10-27). The noncontact nature of laser scanning makes it well suited to in-process nii surement, including such difficult tasks as the inspection of hot-rolled or extruded4 terial, and its comparative simplicity has led to the development of highly porta! systems.The bench gage versions can measure to resolutions of 0.0001 mm.

10.6 VISION SYSTEMS FOR MEASUREMENT

If a picture is worth a thousand words, then vision systems are the tome of inspec@ methods (see Figure 10-28). Machine vision is used for visual inspection, for guidafi and control, or for both. Normal TV image formation on photosensitive surface$, arrays is used, and the video signals are analyzed to obtain information about the objq Each picture frame represents the object at some brief interval of time. Each frak must be dissected into picture elements (calledpiuels). Each pixel is digitized (has bin numbers assigned to it) by fixing the brightness or gray level of each pixel to produor bit-map of the object (Figure 10-29). That is, each pixel is assigned a numerical v based on its shade of gray. Image preprocessing improves the quality of the im by removing unwanted detail. The bit-map is stored in a buffer memory. By a and processing the digitized and stored bit-map, the patterns are extracted, edges lo and dimeosions detemined.

Image buffer

H
Monitor

lmage preprocessing

Image analys~s

I
tool or robot

n
lmage formation (lighting. OPtlCS. filters, polar~zersl la) Object lbi Video plcture frame

t=

Mircrocomputer Storage

computer Terminal

Sophisticated computer algorithms using artificial intelligence have greatly reduced the computer operations needed to achieve a result, but even the most powerful video-based systems currently require one to two seconds to achieve a measurement.This may be too long a time for many on-line production applications. Table 10.6 provides a comparison of vision systems to laser scanning. With the recent emphasis on quality and 100% inspection, applications for inspection by machine vision have increased markedly. Vision systems can check hundreds of parts per hour for multiple dimensions. Resolutions of i O . O 1 in. have been demonstrated, but 0.02 in. is more typical for part location. Machine vision is useful for robot guidance in material handling, welding, and assembly, but nonrobotic inspection and part location applications are still more typical. The use of vision systems in inspection, quality control, sorting, and machining tool monitoring will continue to expand. Systems can cost $100,000 or more to install and must be justified on the basis of improved quality rather than labor replacement.

(cl Digitized frame

Laser-ScanningSystems
Independent Object usually stationary Readily adapted; some limitations on robot motion speed or overall system operation Simple; computers often not required Very fast Readily handled; edges and features produce sharp transitions in signal Can size an object in a single scan per axis Limited three dimensiona1ity;needs ranging capability Submicrometer 0.001 to 0.0001 in. or better accuracy; highly repeatable

Video-Based

Systems

Dependent Multiple cameras or strobe lighting may be required Readily adapted; image-processing delays may delay system operation Requires relatively powerful computers with sophisticated software Seconds of computer time may be needed Requires extensive use of sophisticated software algorithms to identify edges Can size on horizontal axis in one scan; other dimensions requirc full-frame processing Uses two views of two caineras with sophisticated software or structured light Depends on resolution of cameras and distancc between camera and object systems with 0.004-in.precision and 0.006-in. accuracy are typical

IINATE MEASURING MACHINES

Precision measurements in three-dimensional Cartesian coordinate space can b with coordinate measuring machines (CMM) of the design shown in Figure 10-30. parts are placed on a large granite flat or the table.The vertical arm carries a probe can be precisely moved in x-y-z directions to produce 3D measurements. In this sign, the vertical column rides on a bridge beam and carries a touch-trigger probe. Su machines use digital readouts, air bearings, computer controls, and granit achieve accuracies of the order of 0.0002 to 0.0004 in. over spans of 10 to 30 in. These systems may have computer routines that give the best fit to featu ments and that provide the means of establishing geometric tolerances, discus er in this chapter. Figure 10-31 gives a partial listing of the results one can achieve. these machines.

-MEASURING INSTRUMENTS
Accurate angle measurements are usually more difficult to make than linear meas ments.Angles are measured in degrees (a degree is part of a circle) and de divisions of a degree (or in minutes and seconds of arc). The SI system calls measurements of plane angles in radians, but degrees are permissible. The grees will continue in manufacturing, but with minutes and seconds of arc possibly b replaced by decimal portions of a degree. The bevel protractor (Figure 10-32) is the most general angle-measur instrument. The two movable blades are brought into contact with the sides of angular part, and the angle can be read on the vernier scale to 5 minutes of A clamping device is provided to lock the blades in any desired position so that instrument can be used for both direct measurement and layout work. As previously, an angle attachment on the combination set can also be used to mea angles, similar to the way a bevel protractor is used but usually with somewhat accuracy.

FIGURE 10-30 Coordinate

measuring machine with inset showing probe and a part be~ng measured.

Straightness Reference line

Straightness Measured or previously calculated points may be used t o determine a "best fit" iine. The form routine establishes two reference lines that are parallel to the "best fit" line and that just contain all ofthe measured or calculated points. Straightness is defined as the distance D between these two reference lines. Flatness Measured or prev~ously calculated polnts may be used to determ~ne the "best f ~ t " plane The form routlne establishes two reference planes that are Parallel to the "best f ~ t " olane and that just contaln all ofthe measured or calculated ~ o i n t s . Flatness is deiined as the distance D between these two reference planes. Roundness Measured or previously calculated points may be used to determine the "best fit" circle. The farm routine establishes two reference circles that are concentric with the "best fit" circle and that just contain all of the measured or calculated points. Roundness is defined as the difference D in radius of these two reference circles. Cvlindricitv Measured or previously calculated Doints mav be used to determ~ne the "best fit" cylinder The form routine establishes two reference cylinders that are co-axial to the "best fit" cylinder and that just contain all of the measured or calculated points. Cylindricity is the difference D in
~~ ~

"Best fit" line Flatness Reference plane

plane

I D

Reference plane

Roundness
,,Best

Reference circle a D r c I e

Reference circle

Cylindricity Reference cylinder

Reference "Best fit" cylinder Con~c~ty dinder

radius of these two reference cylinders Also applicable to stepped cylinders. Con~c~ty

D
Reference cone

E 10-31 Examples of
'

Measured or prev~ously calculated polnts may be used to determ~ne the "best f ~ t " cone. The form routlne are co-axial two establishes withreference and similar cones to the that "best fit" cone and that just contain all of the measured or calculated points. Conicity is defined as the distance D between the side of these two reference cones.

,:\

form tolerances b y probing surface

"Best fit" cone Reference cone

U R E 10-32 Measuring an eon a part with a bevel ractor. (Courtesy of Brown &

242

Cwifi~ 10 Measurement and lnspectlon andTesting

0 Dial indicator (no d e v ~ a t ~ o wanted) n

FIGURE 10-33 Setup to measure an angle on a part

using a sine bar. The dial indicator is used to determine when the part surface X is parallel to the surface plate.

Gage block.

The toolmaker's microscope is very satisfactory for making angle measureme but its use is restricted to small parts. The accuracy obtanable is 5 minutes of Similarly, angles can be measured on the optlcal contour projector. Angular m surements can also be made by means of an angular interferometer w ~ t h the lr system. A rine bar may be used to obtain accurate angle measurements if the physicalc ditions will permlt. This device (Figure 10-33) consists of an accurately ground ba~ which two accurately ground pins of the same diameter are mounted an exact dista apart. The distances used are usually elther 5 or 10 in., and the resulting instrume~ called a 5- or 10-m. sine bar. Sine bars are also available with millimeter dimensic Measurements are made by using the principle that the sine of a given angle is then of the opposite side to the hypotenuse of the right triangle. The part being measured is attached to the sine bar, and the mclination of the sembly is raised until the top surface is exactly parallel with the surface p1ate.A st of gage blocks is used to elevate one end of the sine bar, as shown m Figure 10-33.7 height of the stack directly determines the difference m height of the two pms.The~ ference in he~ght of the pins can also be determined by a dial indicator gage or any ot type of gage The difference m elevation is then equal to either 5 or 10 times thes of the angle being measured, depending on whether a 5- or 10-in. bar is being us Tabulated values of the angles corresponding to any measured elevahon difference 5- or 10-in. sine bars are available in various handbooks. Several types of sine bars available to suit various requirements. Accurate measurements of angles to 1 second of arc can be made by means angle gage blocks.These come ln sets of 16 blocks that can be assembledin deslredcc b~nations.Angle measurements can also be made to i0.001" on rotary indexing tab havmg suitable numerical control.

In manufacturing, particularly m mass product~on, it may not be necessary to know1 exact dnnensions of a part, only that it is within previously established limits. Linnts4 often be determined more easlly than specific dimensions by the use of attribute-$ instruments called gages. They may be of elther fixed type or deviation type, may! used for both linear and angular dimensions,and may be used manually or mechanid (automatically).

FIXED-TYPECAGES A Fixed-type gages are designed to gage only one dimension and to indicate whetherr I larger or smaller than the previously established standard. They do not determine how much larger or smaller the measured dimenslo1 than the standard. Because such gages fulfill a simple and limited function, they arer 1 atively inexpensive and usually quick and easy to use. i Gages of this typeare ordinanly made of hardened steel of proper composit~ona are heat treated to produce dimensional stability. Hardness is essential to mininnzeq and maintam accuracy. Because steels of h g h hardness can become dimensionally I

Pla~nplug gage having the go member e left end (1,1250-ln.d~ameter) and no-go ber on the right end. (Courtesy of Sheff~eld)
RE 10.34

FIGURE 10-35 Step-type plug gage with go and no-go elements on the same end. (Courtesy of Shefield.)

stable,some fixed gages are made of softer steel, then given a hard chrome plating to provide surface hardness. Chrome plating can also be used for reclaiming some worn gages. Where gages are to be subjected to extensive use, they may be made of tungsten carbide at the wear points. One of the most common fixed gages is thepluggage. As shown in Figure 10-34, plug gages are accurately ground cylinders used to gage internal dimensions, such as holes.The gaging element of apIainplugguge has a smgle diameter.To control the minimum and maximum limits of a given hole, two plug gages are required.The smaller, or go gage, controls the minimum because it must go (slide) into any hole that 18 larger than the required minimum.The larger, or no-go gage, controls the maxlmum dimension because it will not go into any hole unless that hole is over the maximum permissible size. The go and no-go plugs are often designed with two gages on a slngle handle for convenience in use.The no-go plug is usually much shorter than the go plug, it is subjected to little wear because it seldom slides into any holes. F~gure 10-35 shows a step-type go/no-go gage that has the go and no-go diameters on the same end of a single plug, the go portion being the outer end.The user knows that the part is good if the go gage goes into the hole but the no-go gage does not go. Such gages requme careful use and should never be forced into (or onto) the part. Obviously these plug gages were spec~ally designated and made for checking a specific hole on a part. In designing plug and snap rlng gages, the key piinciple is. zt is better fo reject ir goodpart than declare a badpan to be within spec~ficateons.All gage design decisions are made with this principle in mind. Gages must have tolerances like any manufactured components. All gages are made with gage and wear tolerances. Gage tolerance allows for the permissible variation in the manufacture of the gage.It is typically 5 to 20% (depending on the industry) of the tolerance on the d~mension being gaged. Wear tolerances compensate for the wear of the gage surface as a result of repeated use. Wear tolerance is applied only to the go side of the gage because the no-go side should seldomsee contact with apart surface. It is typically 5 to 20% of the dimensional tolerance. Plug-type gages are also made for gaging shapes other than cyhndncal holes Three common types are taperpluggages, threadplug gages, and splzrte gages.Taper plug gages gage both the angle of the taper and its size. Any deviation from the correct angle is indicated by looseness between the plug and the tapered hole.The size is indicated by the depth to which the plug fits into thehole, the correct depth being denoted by a mark on the plug.Thread plug gages come in go and no-go types.The go gage must screw into the threaded holes, and the no-go gage must not enter. Ring gages are used to check shafts or other external round membemThese are also made in go and no-go types, as shown in Figure 10-36. Go ring gages have plain knurled exteriors, whereas no-go ring gages have a circumferential groove in the knurling, so that they can easily be distinguished. Rrngthread gages are made to be slightly adjustable because it 1s almost ~mpossible to make them exactly to the desired size.Thus they are adjusted to exact, final size after the final grinding and polishing have been completed. Snap gages are the most common type o f fixed gage for measuring external dimensions. As shown in Figure 10-37, they have a rigld, U-shaped frame on which are two or three gaging surfaces, usually made of hardened steel or tungsten carbide. In the RE and nO-gO adjustable type shown, one gaglng surface is fixed, and the other(s) may be adjusted ~ght) rlng gages for over a small range and locked at the desired position(s). Because in most cases one king a shaft,(Courtesy of wlshes to control both the maximum and the minimum dimensions, the progressive or motlon and ~~~~~~~~~~t Ion, Bendix Corporat~on.) step-type snap gage is used most frequently. These gages have one ftxed anvil and two

Adjustable go anvtl
Adjustable no-go anv~l

F~xed anvil

Light pressure ,Step

pin

and no-go snap gage. of Bendrx Corporation, ond Measurement Divi

FIGURE 10-37 Adjus

adjustable surfaces to form the outer go and the inner dogo openings, thus elimina the use O f separate go and no-go gages, Snap gages are available in several types and a widerange of sizesThe gaging% Workpiece to be gaged faces may be round or rectangular. They are set to the desired dimensions with the FIGURE 10-38 Flush-p~n gage of gage blocks. being used to check height of Many types of special gages are available or can be conskrncted for special ap step. cations The flush-pin gage (Figure 10-38) is an example for gaging the depth of a shb der.The main scction is placed on the higher' ofthe two surfaces,w ~ t h the movable pin resting on the lower surface If the depth between the two surfaces is sufficien not too great, the top of the pin, but not the lower step, wlll be shghtly above the top lace of the gage buddy. If the depth is too great, the top of the pin will be below the face. Similarly,if the depth 1 s not great enough, the lower step on the top of the pin be above the surface of the gage body. When a finger or fingernail is run across the of the pin, the pin's position with respect to the surface of the gage body can read11 determined. Several types of form gages are available for usem checking eheprofileofvari objects.Two of the most common types are radrusxage.~ (Figure 10-34) andscrew-thr pzrchgage~(Fipre 10-40).

FttURE 10-39 Set of radius gages, showing how they are used (Counesy of MTI CorporaDon)

DEVIATION-TYPE GAGES A large amount of g a p g , and some measurement, is done through the use of& type gaga, which determine the amount by which a measured part devrates,plus or from a standard dimension to which the ins&rument hasbeen set, Inmost cases, anon is indicated directly in units of measurement, but in some cases, the gage sh whether the deviation is within a permissible range. A good example of a deviatio gage is a flashlight batterycheckw, which shows whether the batteq is good (green (red), or borderline (yellow) but not how much voltage or current is generated, Such use mechanical. electrical, M fluidie rrmphficat~on techniques so that very small 1 deviations can be detected. Must are qulte rugged, and they ate available in a vafi designs, ampbfications, and sizes. Dtal indrcator~~, as shown m Figure 10-41,are a widely used form of deulabo gage. Movement of the gaging spmdle is amplified mechanically through arack ion and a gear train and 1s indicated by a y n t e r on a graduated dial. Most dia tors have a spindle travel equal to about 2Li revolutions of the inhcating pointer and read tn either 0.001 or 0.0001 in. (or 0.02 or 0.002 mm). The dial can be rotated by means of the knurled bezel ring to align the zero p with any posltioa of the pointer.The mdlcator is often mounted on an adjustable to perm~t its being brought into proper relationship with the work. It is Important the axis of the spindle be aligned exactly with the dimension being gaged if a m a to be achieved. Digital dial indicators are also readily available.

FIGURE 10-41 Digital dial indicator with I-in. range and 0.0001-in. accuracy. (Courtesy ofCDI.)

Dial indicators should be checked occasionally to determine il their gage capability has been lost through wear in the gear train. Also, it should be remembered that the pressure of the spindle on the work varies because of spring pressure as the spindle moves into the gage.This spring pressure normally causes no difficulty unless the spindles are used on soft or flexible parts Linear variable-differential transformers (LVDT) are used as sensory elements in many electronic gages, usually with a solid-state diode display or in automatic inspection setups. These devices can frequenlly be combined into multiple units for the simultaneous gaging of several dimensions.Ranges and resolutions down to 0.0005 and 0.00001 in. (0.013 and 0.00025 mm, respectively) are available. A i r gages have special characteristics that make them especially suitable for gaging holes or theinternaldimensions of various shapes. A typical gage of this type, shown earlier in Figure 10-12,indicates the clearance between the gaging head and the hole by measuring either the volume of air that escapes or the pressure drop resulting from the airflow.Thegage is calibrated directly in 0.0001-in. or 0.02-mm divisions.Air gages have an advantage over mechanical or electronic gages for this purpose in that they detect not only linear size deviations but also out-of-round conditions.Also, they are subject to very little wear because the gaging member is always slightly smaller than the hole and the airflowminimizes rubbing. Special types of air gages can be used for external gaging.

A variety of tests have been developed to evaluate product quality and ensure the prov~des one such means absence of any performance-impairing flaws. Destructive te~tmng of product assessment. Components or assemblies are selected and then subjected to conditions that ~nduce fallure Determin~ng the specific conditions where failure occurs can provide insight into the performance charactenstics and quahty of the remain~ng products. Statistical methods are used to determine the probability that the remaining products would exhib~t sirmlar behavior. For example, asqume that 100 parts are produced and then one is selected (randomly) and tested to failure.1~ it safe to assume that therema~nmg 99 w~ll perform the same way? A satisfactory test of another randomly selected part (or, more typ~cally, the first and last of the 100 parts) would further increase our confidence in the remalnlng 98. Additional tests would enhance this confidence,but the cost of destroy~ng each of the tested (I e ,destroyed) products must be borne by the rematning quantity. Regardless of the amount of testing, there w~ll st111be some degree of uncertainty since none of the remalmngproducts have actually been subjected to any form o E property asse5sment. Proof testmng is another means of ensurlng product qual~ty Here a product is subjected to a load or pressure of some determined magn~tude (generally equal to or greater than the designed capacity or the cond~t~on expected during operation). If the part remains intact, there is reason to believe that it will subsequently perform m an adequate fashion,provided it 15 not subjected to abuse or service conditions that exceed its rated

level. Proof tests can b e conducted under laboratory condlt~ons o r at the site of inst lahon o r assembly, as with large manufactured assemblies such as pressure vessels

Nondestruchve tests can detect internal or surface flaws, measure a product's

p r o b m g ( 3 ) a sensor to detect the response; (4) a device to indicate or record the sponse; and (5) a way to interpret and evaluate quallty.

Provides a dlrect and rellable measurement ot how a material or component w~ll respond to servlce cond~tlons 2 . Prov~des quantitat~ve results, useful for design. 3 Does not require interpretat~on of results by skilled operators. 4 Usually tlnds agreement as to meanmg and s~gn~ficanceoi test results Disadvantages I Appl~ed only to asample; must show that the sample a representatrve of the group 2 Tested parts are destroyed dur~ng testlug 3 Usually cannot repeat a test on the same Item or use the same speclmen for multiple tests 4 May be restr~cted tor costly or few I-number parts 5 Hard to predmct cumulat~ve effect of servlce usage. 6 Difficult to apply to parts m use, if done, testmg termlnates their useful life 7. Extensive mach~n~ng or preparahon of test specimens is often requ~red 8 Capital equlpment and labor cost? are often h ~ g h

1 Can be performed directly on product~on Items w~thout regard to cost or quantlty avadable 2. Cdn be performed on 100% of product~on lot (when hlgh variabil~ty is observed) or a representatwe sample (dsuffic~ent ?~milar~ty 13 noted). 3 D~ffereut tests can be apphed to the same item, and a test can be repeated on the same specrmen. 4 Can be performed on parts that are in servlce; the cumulahve effects of servlce llfe can be monitored on a smgle p 5. L~ltle or no spcclmen preparation 1s requ~red 6. The lest equ~pment IS often portable. 7 Labor costs are usually low.
Disadvantages 1 Results often requlre mterpretahon by skrlled operators 2 D~fierent observers may Interpret the test results d~fferently 3 Properties are measured mdirectly, and results are often qualaahve or comparative.

4. Some test equlpment requlres a large capital investment.

Regardless of the specific test, nondestructive testing can be a vital element in good manufacturing practice. Its potential value has been widely recognized as productivity and production rates increase, consumers demand higher-quality products, and product liability continues to be a concern. Rather than being an added manufacturing cost, nondestructive testing can actually expand profit by ensuring product reliability and customer satisfaction. In addition to its role in quality control, nondestructive testing can also be used as an assessment aid in product design. Periodic testing can provide a means of controlling a manufacturing process and reducing overall manufacturing costs by preventing the continued manufacture of out-of-specification, defective, or poorquality parts.

Probably the simplest and most widely used nondestructive testing method is visual inspection, summarized inTahle 10-8.The human eye is a very discerning instrument and, with training, the brain can readily interpret the signals. Optical aids such as mirrors, magnifying glasses, and microscopes can expand the capabilities of this system.Video cameras and computer syste~us, such as digital image analyzers, can be used to automate the inspection and perform quantitative geometrical evaluations. Bore scopes and similar tools can provide accessibility to otherwise inaccessible locations. Only the surfaces of a product can be examined, but that is often sufficient to reveal corrosion, contamination, surface-finish flaws, and a wide variety of surface discontinuities.

ole
lfuges oiom hllimiiarions

simple, easy la use, relatively inexpensive. Depend on skill and knowledge oTinspector.Limited to detection of surface flaws. None.

of evcryone associated wlth parls manufacture

Liquid penetrant testing, also calred dye penetrant inspection, is an effective method of detecting surface defects in metals and other nonporous materials; it is illustrated schematically in Figure 10-42. The piece to be tested is first subjected to a thorough cleaning and is dried prior to the test.Then a penetrant, a liquidmaterial capable of wetting the entire surface and being d r a m into fine openings, is applied to the surface of the workpiece by dipping, spraying, or brushing. Sufficient time is given for capillary action to draw the penetrant into any surface discontinuities,and the excess penetrant liquid is then removed by wiping, water wash, or solvent.The surface is then coated with a thin film of developer, an absorbent material capable of drawing traces of penetrant from the defects back onto the surface. Brightly colored dyes or fluorescent materials that glow under ultraviolet light are generally added to the penetrant to make these

FIGURE 10-42 ~ L ' O petierralit nting: (d) intta \~rf;rce wltrt open rrack; ( u ) netietrdlit :s a~091ed ano 's r)~llcd into rne crdrd hv cdoillarv dct on; (c) excrcs

1 s extraited, and pen'etrant 1 s rernoied, (d) developer IS applied, ~ome'~enktrar;t the product n inspected.

G H A ~ 10 R Measurement and Inspection and Tesung

Pmciple
Ad~anmg@s

A llquid peuetrmt eontaming hu01estant matenel or dye IS drawn mto surface flaws by capllary a a i m and subsequently revealed by developer material in con]unction with visual
Inspection Slmple, mnexpenave. versatde, portable, easaly rnterpreted, aud applicahleta complex
chrna"--y"",

L~milat~ons
Maferlal l~nnrmions

Can only detect flaws that are open to the surfacc;s~f*ws must be cleaned before and after mnspemou; deformed surfaces and surFace coa1tng.s may prevtntdetect1on;and the penetrant may be wped or washed out of largedefms, Cannot be usedm hqt products h ~ h e a b l to e all materials with a nonoornm surface

I
traes more visible, and the developer is often selected to provlde a contrasti ground Radioactive tracers can also be added and used in conjunction with graphicpaper to produee a permanent image of the defects. Cracks, laps,seams, bonding, pinholes, gouges, and tool marks can all be detected. After inspection, veloper and residual penetrant are removed by a second cleaning operation. To be su~cessful the mn-spection for surface defects must be correlated w i manufacturing operations IC previous processex such as shot peening, honing, bum machining, or vari~us forms of cold working produced plastic deformation of the s material, a chemical etching may be required to remove material that mig ing critical flaws. An alternative procedure is to perform a penetrant tes surface-finishing operatima when significantdefects will still be open and av detection. Penetrant inspection systems can ranEe from aerosal spray cans of c penetrant, and developer (for portable applications), to automated, mass-prod equipment using sophistrcated wmputer vislon systems.Tab1e 10-9is a summary process and its advantages and limitations.

A'fag~etdc paruck inspection, summarized in Table 10-10, is based on the princ~ple ferromagnetic materials (such as the alloys of iron, nickel, and cobalt), when m tized, will have distorted mametic fields in the vicinity of material defects.Pwsho F~gure 10-43,surface and subsurface flaws, such as cracks and inclusions, will magnetic anomalim that can be mapped with the aid of magnetic particles on imen surface. As with the previaus method, the specimen must he cleaned pnor spection. A suitable magnetic field ts then estabhhed in the part. AS $ Figure 10-44, orientation can be quite important.For a flaw ta be derected.i dnce a significapt disturbance of the magnetic field at or near the surface,

Gemertzciailmliatiom

Permaneni record

When magnehzed,ferrama~nct~c materials will have a dutorted magqetie held LDthe ulcin~ty of flaws and ddecfech Magneucparticles will &strongly attracted to rcpgicns w h m the m a p e h c hux breaks thg surface. Relatively s m p k , Last,easy-to-intewret, portable units ex1rt;can reveal both surfaceand subsunFace tPaws and inclus~ons (as much as 6mm d&p) and small, txht ccraks Pam must he relavely clean, alrgnment of the flaw and the field affects the sensltlvrv so that muluple lnspeetiom with mtferent magnetizatwns may be requ~red, can only detect defems at or near surfaces: m u s t demaguetm part after tsst,h~gbcwxent source is ~equmd',some su&w processes can maskdefect8,postckaoing may be requ~ted Must beferromagneac: Raferrous metaissuch as alum~num,magnes~um, capper, lead, tin, and tltau1um and the ferrous (but not ferrom+gneric) austenltlc stainless steels cannot bc rnspected %re and shape are almost unlimired,mo#trestnclians rehte t o the ahlllty to induce uolCorm magnefic fields wthm the p : hard to u% On rough sudace6 Photographs, uldeotapes, and inspectors' reports are most mmmon. Ined&tion,the defezt pattern can be preserved on thr: specimen by an apphahon of transparent laoqueror transferred toa plece of transparent tape that has been apphed to the apctmen and peeled off.

I
I

Magnet~c

(b)

(c)

RE 10-43 (a) Magnetic field show~ng d~srupt~on by a surfacecrack; (b) magnetlc particles are apphed and are if they are rent~ally attracted to f~eld leakage, (c) subsurface defects can also produce surface-detectabled~srupttons lently close to the surface

flaws will show

and 45flaws will show

crack w~ll

be detected

not show
ib)

steel 1s placed wlthin an energized coi1,a magnetic field will be produced whose lines of flux travel along the axis of the bar. Any defect perpendicular to this ax~s wlll sign~ficantly alter the field. If the perturbation is sufficiently large and cloae enough to the surface, the flaw can be detected. However, if the flaw is sn the form of a crack al~gned w~th the specimen axis, there will be l~ttle perturbanon of the lines of flux and the flaw 1s likely to go undetected. If the cylindrical specimen is then magnetized by passing a current through 11,a circumferenlial magnetic field will be produced. Any axla1 defect now becomes a signdlcant perturballon, and a defect perpendicular to the axis will likely go unnoticed. To fully inspect a product, therefore, a serkes of inspections may be required using vanous forms of magnetization. Passing a current between various points of contact 1s a popular means of inducing the deslred fields Electromagnetic co~ls of varlous shapes and slzes are also used. Alternating-current methods are most sensitive to surface flaws, whle direct-current inspections are better for detecting subsurface defects, such as nonmetallic inclusions. If the cylindncal specimen is then magnetized by passlug a current through it, a circumferential magnetic field will be produced. Any axial defect now becomes a signsficant perturbation, and a defect pmpendxcular to the axis w~ll likely go unnoticed.To fully Inspect a product, therefore, a series of inspections may be required using various forms of magnetizahon. Passing a current between vanous polnts of contact is a popular means of rnduclng the deslred flelds. Electromagnet~c coils of various shapes and sizes are also used. Alternating-current methods are most sensitive to surface flaws, whde d~rect-current inspections are better for detecting subsurface defects,such as nonmetallic inclusions. Once the specimen has been subjected to a magnetic fie14 magnetic particles are applied to the surlace in the form of either a dry powder or a suspension in a hqlud carrier These particles are attracted to places where the lines of magnetic flux break the surface,revealing anomalies that can then be interpreted To better reveal the orientation of the lines of flux, the particles are often made in an elongated form.They can also be

250

CHAPTER 10 Measurement and lnspect~on and Testlng

FIGURE 10-45 Front-axle k~nq pin tor a rrurk. (a) A$ n~ari.rfact.rrcd and a p p ~ r e n t . )

la!

soulid: (b) 'r~spccteo ~rlder corive~~rional rnagnetrc particle ~nspect~on to reveal numerous gr~nd~ng-~nduced cracks; (c) fluorescent particles and ultravrolet light make the cracks even more visible (Courtesy of Magnaflux Corporabon.)

treated with a fluorescent material to enhance observat~on under ultraviolet li coated with a lubricant to prevent ox~dation and enhance their mobihty, Figure shows a component of a truck front-axle assembly as manufactured, under st magnetic partrcle inspection, and under ultraviolet light w ~ t h fluorescent particles

Sound has long been used to provide an indication of pprducl quality.A cracked b not ring true, but a fine crystal goblet will have a dear ring when lightly tapped. S an object and listening to the characteristic ring 1s an ancient art but is limited t detection of large defects because the wavelength of audible sound is rather compared to the size of most defects. By reducing the wavelength of the signal t ultrasonlcrange, typically between IU0,OM) and 25 mllhon herb, ultrasonic inspectio be used to detect rather small defects and flaws A$ shown in Table 10-11, ultrasonzc raspection mvolves send~ng high-fr waves through a material and obsemmg the response.Within the speclmen,soun interfaces w i can be affected by voids,impur~ties, changes in density, delaminat~ons, terials having a different speed of sound, and other imperfectrons.At any interfac of the ultrasonic wave wiU be reflected and part wlll be transmtted. If the mncldent

High-frequency sound waves are propaghted through a test specimen, and the transmitted
reflected signal is monitored and interpreted. Can rzveal internai defects; high sensitivity to most cracks and flaws;high-speed test with immediate results; can be automated and recordeil; portable; high penetration in most important materirils (up to W ft in steel);indicates flaw size and 1ocation;access tooniy o side is required can also be used to measure Lhicknesh Poisson-s ratio,or elastic modulu nrecmlc no rndiatinn nr qafptv hararci. . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~, Difficult to use with complexshapes; external surfaces and test (may need dual transducer or multiple inspzctions); a coveiage i s smaU (inspection of large areas requires scanning); trained, experienced, and motivated technicians may be required. Few can be used on metals,plastics,cwami~ glass. rubber,graphite, and concrele, as well as joints and interfaces between materials. Small,thin, or cornpiex-shaped parls or parts with rough surfaces and nonhomogeneous strwture pose the greatest difficulty. Ultrasonic signals can be recorded For subsequent playback and analysis. Strip charts can also be used.
~~

Matcrul llrmrtarlonr
Ceometnc lrmitattvni Permanent record

I I

is at an angle to an interface where materials change, the t ~ a n s ~ t t portion ed s f the beam w i l l be bent to a new a n & by thephetlomenon of refractmn.Byreceiving andInterpretingeither trmsmirted OI reflected s@als,ultrasonic inspeetfon can be uved to detect flaws within the mabrid, measure thickness from only one side, or characterize metallurgisal snucture. An nltrasonicin6pecfion system hegins with a pulsed oscrlfator and transducer, a device that transforms electrical energy into mechanical vlbrabons. The pulsed ofcillatox gneratas aburst ofalternathgvolca~e, witba characteristicprindpalfreqmcy, duration, profile, and repetition rate. ?lWs burst is then applied to a sanding transducer, which uses a piezoelectric crystal to cmverf the electrical oscillations into mechanical vibrations. Becaus-e air is a p o r transmitter of ultrasonic waves, an acoustic coupli* medium-generally a liquid such as oil or water-is required to liik the transducer to the pi- to be inspected and transmit the vibrations into the part.The pulsedvibraticms then propagate through the part with a velocity that depends on the density and elasticay of* test materia1.k receiving transimcer i s h n usedlo convert the heransmtted 6r reflesaedvibrationsback into electrical si~n&.The rec6ivmn.g transducer is often identical . to the sending unit,and the s,lme tranjduccr can actually periorn~ both iunction\.A rccciving unit rhcn amplifies, riltcrs, and prucehses the sign31 ior displas, po~s~hlc recording. and final intcil)ret;ition.An clectrdnic clock is gcn~.rfiIl! intzgratcrl into t l i c <!.stern 'to time the responses and provide reference signals for comparison purposes. Depending r m the test objectives and part geometry, aeverd different inspection

L In thepulse*edza techmque, an ultrasonic pulse 1s introduced into the pmce to be inspected, and the echoes from opposing surfaces and any intervening flaws are deh e initial witted pulw and the tected by %he reCeiVw-The time interval between t various echoes can be displayed an the horizontal axis of a display smepn.Defects are identified by the position and amplitude of the various echoes. Figure 10-&shows a schematic of a single-transducerpulse-echo inspection and the companion signal as it would appear on a display. Figure 10-47a depids a dual-transduoerpulse-echoexamiflatian. Both cases require a w s to only one side of the specimen.

Sound echo

I
t

Introduced pulse Echo from [bottom surface defect

Time

(b)

Sendlng transducer

&

Receiving transducer

2. The through-transmtssron techncque requlres separate sending and rece~ving trz ducers As shown m Figure 10-47b,a pulse is em~tted by the sending transducer a detected by a receiver on the opposlte surface. Flaws in the mater~al decrease t amplitude of the tran~mitred signal because of back-reflection and scattering. 3. Resonurzce testing can be used to determine thc th~ckness of d plate or sheet f r c one side of the matenal. Lnput pulses of varying frequency are fed into the materi When resonance 1s detected by an Increase in energy at the transducer, the thicknf can be calculated fiom the speed of sound in the material and the time of traver Ultrasonic thickness gages can be cahbrated to prov~de d~rect dig~tal readout oft thickness of a material

Reference standards--speamens of known lhckness or containing various ty? and slzes of machmed "flaws"-are often used to ensure consistent results and ald interpret~ng any indicationcof internal discont~nmt~es.

Lfll 10.15 RADIOGRAPHY

Radiographic inspection, summarized in Table 10-12, employs the same principles 8 techniques as those of medical X-rays. A shadow pattern is created when cerrain tyj of radiation (X-rays, gamma rays, or neutron beams) penetrate an object and are d ferentially absorbed due to variations in thickness, density,or chemistry, or the presen of defects in the specimen. The transmitted radiation is registered on a photograps film that provides a permanent record and a means of analyzing the component. Fli rescent screens can provide direct conversion of radiation into visible light and enat fast and inexpensive viewing without the need for film processing.The fluorescent imaj however, usually does not offer the sensitivity of the photographic methods. Various types of radiation can be used for insp.ection. X-rays are an extrem! short wavelength form of electromagneticradiation that are capable of penetrating 4 materials that reflect or absorb visible 1ight.Tney are generated by a high-voltage elect cal apparatus-the higher the voltage, the shorter the X-ray wavelength and the gred the energy and penetrating power of the beam. Gamma rays, another useful for4 electromagnetic radiation, are emitted during the disintegration of radioactive nucf Various radioactive isotopes cm be selected as the radiation source. Neutron beams! radiography can be obtained from nuclear readors, nuclear accelerators, or radio4 topes. For most applications it is necessary to moderate the energy and collimate! beam before use. j~ The absorption of X-rays and gamma rays depends on the thickness, density,;y atomic structure of the material being inspected. The higher the atomic greater the attenuation of the beam. Figure 10-48 shows a radiograph of Liberty Bell.The famous crack is clearly visible, along with the internal spider to support the clapper in 1915) and the steel beam and bolts installed in yoke in 1929. Other radiographs disclosed previously unknown shrinkage and additional cracks in the bell, as well as a crack in the bell's clapper.

I'rincipic Advnnlage.~ Ltrniioiio,i.s

Macerial liniitaiions Gcomerric liniilirlions

IJermunenr record

Some Corm of radiation (X-mv.gamma ray, or neutron beam) is passcd through the sample and is diiCe~entIa'~ absorbed depending on the thickness, type of material, and thc presence of internal flaws or defects. Pnrbes the internal regions of a material; provides a permanent record of the inspection; can be used to determine the thickncss o l a rnateria1;very sensitive to density changcs. Most costly of the NDT methods (involves expensive equipment):radiation precautions arc necessary (potentially dangerous to human health); the defect must he at least2% of the total section thickness to be detrcted (thin cracks can he missed if oriented perpendicular to the beam): film processing requires time, faciiilies, and carc; the image is a two-dimensional piojection of a three-dimensianal ohjed, so the location of an internaldefect requires a second inspection at a different angle; complexshapes can present problem?: a high degree of operator training is required. Applicable to most engineering materials. Complex shapes can present problems in s t t i n g exposure conditions and obtaining proper orientation of source.specimcn,and film.Two-side accessibility is required. A photographic image is part of the standard test procedure.

i 10-48 Rad~ograph of

2rty Bell. The photo the famous crack, as well ron sp~der installed in I SuDoort the claooer and , , is beam ana \bpport(. lucre 5et into Ole )uke in Codrteryu i Larrnlorr .onlpo,iy

..

In contrast to X-ray absorphon, neutron absorphon valies w~dely from atom to atom, with no pattern in terms of atomtc number. Unusual contrasts can be obtained that would be tmpossible with other inspectton methods. For example, hydrogen has a high neutron absorpt~on.The presence of water in a product can be easily detected by neutron radiography. X-rays, on the other hand, are readlly transm~tted through water, and its presence could be missed. When a radiation beam m passed through an object, part of the rad~atlon i~ scattered In all direcbons This scatter produces an overall 'Yoggmg" of the radiograph, reduclng the contrast and sharpness of the image. The thlcker the material, the more troublesome the scattered rad~ation becomes. Photographic considerations relatrng to the exposure time and development also affect the qualtty of the radlographtc Image. Image-enhancing computer software can help reveal the subtle but important vartations in photographic denslty. A standard Lest piece, or penetrometer, 1 s often included in a radiographic exposure. Penetrameters are made of the same or s i d a r mater~al as the speclmen and contain features with known dimenstons.The image of the penetrameter is compared to the image of the product being Inspected Regtons of simtlar intens~ty are cons~dered to he of similar thickness Rad~ography is expensive, however. Many users. therefore, recommend extensive use only dunng ihe development of a new product or process, followed by spotchecks and stat~sttcal methods dnr~ng subsequent product~on.

When an electrically conductive material is exposed to an alternating magnetic field such as that generated by a cod of wlrc carrying an alternat~ng current, small electric currents are tnduced on or near the surface of the matertal (Figure 10-49) These Induced eddy currents, 1n turn, generate the11 own opposrng magnetic fteld, which then reduces the strength of the freld from the cotl. This change 1n magnetic field causes a change in the impedance of the cod, which m turn changes the magmtude

FIGURE 10-49 Relation of the

magnetizing coil, magnetizing current, and induced eddy currents. The magnetizing current is actually an alternating current, producing a magnetic field that forms, collapses, and re-forms in the opposite direction. This dynamic magnetic field induces the eddy currents, and the changes in the eddy currents produce a secondary magnetlcfleld that Interacts w~th the sensor toll or probe

Maanetizinq current

= J
w

n g z iit;!:

/ coil

Magnetizing

currents
I I

Magnetizing current

currknts

Induced Magnetic f~eld

bv the eddy currents

FIGURE 10-50 Eddy currents are constrained to travel within the conductive material, but the magnitude and path of the currents will be affected by defects and changes in mater~al properties. B y focuslng on the magnitude of the eddy currents, features such as differences In heat treatment can be detected.

I of the current flowing through it. By monitoring the impedance of the exc~ting a separate indicating toll, eddy-current testing can be used to detect any car that would affect the current-carry~ng ability (or conductivity) of the test spe, Figure 10-50 shows how the eddy-current paths would be forced to alter arc crack, thereby changing the characterisbcs of the induced magnetic field in that v Eddy-current testing, summarized in Table 10-13, can b e used to surface and near-surface flaws, such as cracks, voids, inclusions, and seams. concentrations, differences in metal chemistry, or variations in heat trez (i e., microstructure and hardness) will all affect the magnetic permeab~h conductivity of a metal and therefore alter the eddy-current characteristics. M mix-ups and processing errors can therefore be detected. Specimens can be by hardness, case depth, residual stresses, or any other structure-related prr

When an electr~cally conducuve matenal is brought near an alternatm~-current call that

defZTts can affictihe magnitude and d ~ e c t ~ot &t<e ~nduced eddy currenis a i d can be
detected hv thc elpctromcs.

Advantage

Limilarivns Material limitations Geometric limifotions

Permanen! record

standards are needed for cornpansan,trained operators are generally reqm~ed Only applicable to conductive materials, such as metals, Same d~fficulues may be encouniered with ferromagneticmaterials. Depth of penetration is limited:must have accessibility of coil or probe; constant separat distance between coils and specimen is required for good results Electronic signals can be recorded using devices such as strip-chart recorders.

Thickness (or variation in thickness) of platings, coatings, or even corrosion can be detected and measured. Eddy-current test equipment can range from simple. portable units with handheld probes to fully automated systems with computer control and analysis. Each system includes a source of changing magnetic field capable of inducing eddy currents in the part being tested, a means of sensing the field, and a means of measuring and interpreting the resulting impedance changes. When comparing alternative techniques, eddy current is usually not as sensitive as penetrant testing in detecting small, open flaws, but it requires none of the cleanup operations and is noticeably faster. In a similar manner, it is not as sensitive as magnetic particle inspection to small subsurface flaws, but it can be applied to all metals (ferromagnetic and nonferromagnetic alike). In addition, eddy-current testing offers capabilities that cannot be duplicated by the other methods, such as the ability to differentiate between various chemistries and heat treatments.

LO. 17 ACOUSTIC EMISSION MONITORING

Materials experiencing the dynamic events of deformation or fracture emit stress waves in frequencies as high as 1MHz. Whiie these sounds arc inaudible to the human ear, they are detectable through the use of sophisticated electrouics.Transducers,amplifiers,filters, counters, and computers can be used to isolate and analyze the sonic emissions of a cracking or deforming material. Muchlike the warning sound of ice cracking underneath boots or skates, the acoustic emissions of materials can be used to provide a warning of impending danger.They can detect deformations as small as 10-IZ in./in. (which occur in short intervals of time), initiation or propagation of cracks (including stresscoorrosiou cracking), delamination of layered materials, and fiber failure in composites By using multiple sensors,it is possible to accurately pinpoint the source of these sounds by a triangulation method similar to that used to locate seismic sources (earthquakes) in the earth. Acoustic emission monitoring, summarized in Table 10-14, involves listening for indications of failure. Temporary monitoring can be used to detect the formation of cracks in materials during production operations, such as welding and subsequent cooling of the weld region. Monitoring can also be employed to ensure the absence of plastic deformation during preservice proof testing. Continuous surveillance may be used when the product or component is particularly critical, as with bridges and nuclear reactor pressure vessels. The sensing electronics can be coupled to an alarm and safety system to protect and maintain the integrity of the structure. In contrast to the previous inspection methods, acoustic emission cannot detect an existing defect in a static product. Instead, it is a monitoring technique designed to detect a dynamic change in the material, such as the formation or growth of a crack or defect, or the onset of plastic deformation.

Principle

Advantages

Limitations

Almost all materials will emit high-frequency sound (acoustic emissions) whzn stressed, deformed, or undergoing structural changes, such as the formation or growth of a crack or defed.%ese emissions can now be detected and provide an indication of dynamicchange within the material. III,:. ! ; 1 1,: ~n.,nlt~,rc.l nlth n , l. I-III*~l r ~ t . t o . . ~ J. ~ I .L;~~.,II an.: r. I h c unurr. ,Irll:tt.r<: 1 ..~nt!nu~>t.~w ~Il;,n.: ~\. 1 , [pt ~ \ ! h . ~ . d < t ~ in .t .c.:~~cr~lc \ I,, ,,'I~cr n ~ ~ l l ~ . . :.r. . l \ I!<, cI~l:ctt J, In,pc:tl(ln c In lr. ill ll.%r>ll:mlr mrncnl, ?#,.I 111: 1 1;111 111 Ill. c n l ~ \ . h l ~ . Il) ~ ~ r i L.~II : hc determined. Only growing or "active" flaws can be deteclcd (the mere presence of defects is not detectable); background signals may cause difficulty; there is no indication of the size or shape of the flaw; expensive equipment is required; and experience is required to interpret the simals and shape of the component affect the strength of the emlsxon signals that reach the detector.

256

CHAPTER 10 Measurement and Inspection and Test~ng

LEAK TESTING Leak testing is a form of nondestructive testing designed to determine the existence absence of leak sites and the rate of material loss through the leaks. Various testii methods have been developed,ranging from the rather crude bubble-emission test (pre surize, immerse, and look for bubbles), through simple pressure drop tests with eith air or liquid as the pressurized media, to advanced techniques involving tracers, dete tors, and sophisticated apparatus. Each has its characteristic advantages,limitations,ar sensitivity. Selection should be on the basis of cost, sensitivity, reliability, and compa bility with the specific product to be tested. THERMAL METHODS Temperature-sensing dev~ces (including thermometers, thermocouples, pyrometel temperature-sens~tivepaints and coatmgs, hquid crystals, infrared scanners, u~fraan film, and others) can also be used to evaluate the soundness of engineer~ng materia and components Parts can be heated and then Inspected during cool-down to reve abnormal temperature distributions that are the result of faults or flaws.The idenW cation of "hot spots" on an operating component is often an indication of a flaw or defe and may provide advanced warnmg of Impendmg failure For example, faulty electri: components tend to be hotter than defect-free devices Composite materials (difficult1 inspect by many standard techniques) can be subjected to bnef pulses of intense heat ar then inspected to reveal the temperature pattern produced by the subsequent the? conduct~vity.Thermal anomahes tend to appear in areas where the bonding between$ components is poor or incomplete. In another technrque, ultrason~c waves are used! produce heat at ~nternal defects, wh~ch are then detected by mfrared examinat~on 1 STRAIN SENSING (( Although used primarily during product development, strain-sensing techniques cg also be used to provide valuable insight into the stresses and stress distribution witd a part. Brittle coatings,photoelastic coatings, or electrical resistance strain gages cad applied to the external surfaces of a part, which are then subjected to an applied The extent and nature of cracking, the photoelastic pattern produced,or the sistance changes then provide insight into the strain at various locations. X-ray diifrac methods and extensometers have also been used. ADVANCED OPTICAL METHODS Although visual inspection is often the s~mplest and least expensive of the nond tlve inspection methods, there are also several advanced optical methods. Mo matic laser hght can be used to detect differences in the backscattered patte a part and a master.Tne presence or absence of geometrical features such as gear teeth is readily detected. Holograms can provlde three-diiens~onal Images object, and holographic interferometry can detect minute changes in the shape object under stress. RESISTIVITY METHODS The elecrfrcal resuhvlty of a conductive material is a function of its chemistry,proc ing h~story, and structural soundness Measurement of resistivity can therefore be for alloy ident~fieation, flaw detection, or the ensurance of proper processing.Test be developed to evaluate the effects of heat treatment, the amount of cold work, th tegrity of welds, or the depth of case hardening.The development of sensitive mlcroo meters has greatly expanded the possibilities in this area. COMPUTED TOMOGRAPHY While X-ray radiography provides a singleimage of the X-ray intensity being trans ted through an object, X-ray computed tomography (CT) 1s an Inspection techmque

provides a cross-sectional view of the interior of an object along a plane parallel to the X-ray beam. Thls is the same technology that has revolutionized medical diagnostrc tmaging (CAT scans), with the process parameters (such as the energy of the X-ray source) being adapted to permlt the nondestructive probmg of industrial products. Basic systems include an X-ray source, an array of detectors, a mechanical system to move and rotate the test object, and a dedicated computer system.The intensity of the received signal 1s recorded at each of the numerous detectors wrth the part in a variety of orientations. Complex numencal algorithms are then used to construct an Image of the interior of the component Internal boundaries and surfaces can be determined clearly, enabling inspection and dimensional analysis of aprodnct's interior.Thepresence of cracks, voids, or inclusions can also be detected, and their precise location can be determined CTinspections are slow and costly, so they are currently used only when the component is cr~tical and the more standard inspection methods prove to be Inadequate due to features such as shape complexity, thick walls, or poor resolution of detail.The vldeo Images of the CT technique also permit easy visualizahon and interpretation. Acoustzc holography 1s anothercomputer reconstmction technique, thls time based on ultrasound reflections from withln the part.

CHEMICAL ANALYSIS AND SURFACE TOPOGRAPHY Whlle nondestructive Inspectton IS usually associated wrth the detection of flaws and defects,various nondestructive techniques can also be employed to determine the chemical and elemental analysis of surface and near-surface material These techniques include Anger electron spectroscopy (AES), energy-dlspernve X-ray analysls (EDX), electron spectroscopy for chemlcal analysis (ESCA), and various forms of secondary-ion mass spectroscopy (SIh4S). Because of its large depth of focus, the scanning electron microscope has become an extremely useful tool for observing the surfaces of materials. More recently, the atomic-force microscope and scanning tunneling microscope have extended information about surface topography with resolution t h ~capability s and can now prov~de to the atomic scale.

There was a time when the detection of a flaw was considered to be sufficient cause for rejecting a material or component, and material specifications often contained the term flaw-free. Such a criterion, however,is no longer pract~cal, because the sensitlliity of detection methods has increased dramat~cally. If materials were rejected upon detecbon of a flaw, we would find ourselves rejecting nearly all commercial engineering maferials. If a defect is sufficiently small, it is possible for it to remarn dormant throughout the useful lifet~me of a product, never changing in size or shape. Such a defect is clearly allowable. Larger defects, or defects of a more undesirable geometry,may grow or propagate under the same (cyclic) cond~txons of loading,often causing sudden or catastrophic failure.These flaws would be clearly unacceptable.The objective (or challenge), therefore, is to ident~fy the conditions below which a flaw remains dormant and above which it becomes cntzcal and a cause for rejection. This issue 1s addressed in the sectlon on "Fracture Toughness and the Fracture Mechanics Approach" in Chapter 2.

I I Key Words
sacy

istic holography emission ~stic Nance (ere butes lela rance fit puted tomography d~nate measuring lachine

coupl~ng med~um flaw cr~tlcal destructive teshng dormant flaw dr~t eddy-current testmg electncal reslstlvity flaw-free gage blocks geometric tolerances hardness testlng

impedance Interference bands Interference bt laser interferometer lay leak testlng length Imearlty hquld penetrant testing machinist's rule magnetlc particle lnspectlou

magnification

mass metrology mrcrometer caliper nondestruct~ve testlng (also nondestruct~ve mspechon) optical comparator penetrameter penetrant plug gage prewsion

258

CHAPrrR 10

Measurement and Inspection and Test~ng srne bar snap gage stab111 ty super micrometer temperature through transmission technique time tolerance tomography toolmaker's flat toolmaker's mlcraswpe transducer ultrasonic inspection variables vernier callper vision system v~sual InspectIan

proof test pulse-echo method rad~ographic inspcctton resolut~on resonance testmg ring gage rule of 10

. IReview Questions
I What are some of the advantages to the consumer of standard~zahon and of ~nterchangeable parts? 2 DFM stands for "design tor manufacturtng " Why IL lt mnlportantfor designers to interface wlth manufacturrng as early as posable w ~ t h the design phase? 3 Explain the d~fference between attrthutes and variables inspect~on. 4 Why have so many var~able-typedevices m autos been replaced wtth attribute-type devses? 5. What are the four h a w measures upon w h ~ c h all others depend?

6. What is a pasca1,and how IS ~tmade up of the bas~cmeas 7 What are the dfferent grades of gage blocks, and why d come in sets? 8. When gage blocks are "wrung together:' what keeps th together? 9 What is the difference between tolerance and allowance? 10. Here is a table that provldes a descr~pt~on of fits from c ance to mterference. Try to think of an example of eac these fits.

I S 0 Symbol Hole Basis Hl licll H9ldY Shaft Basis

Cl llhll D9lh9

Exanlple

Loose-running fif. for wide wmmercial tolerances or allowances an external mernbr~s Free.runningfit-not for use where accuracyis essential,but good for large temperature

3 .g
@

variations
Close-running fit: for funning on accurate machines and for accurale location a1moderate

H81fl H7lg6 H7lb6

F8lh7 G7lh6
H7ihh

ifk

speeds and journal pressures

Slidingfif:not intended to run freely, but to move and turn freely and locate accurately Locmional-clearunce fir: provides snug fit for locating stationary parts, but can be f m l y assembled and disassembled

.9

H7lk6
H7iu6

K71h6
N7lhfi

Locarional-trunsitianfi~: for accurate location; a compromise between clearance and interference

Locoiional-rramirionfi*- for more accurate locatian where greater interference is permissible

3$
8

.-lz
M

82
C

H7ls6 H71u6

S7lh6 U7lh6

Medim-drivefit for ordinary steel parts or shrink fits on light sections;the tightest fitusable with cast iron Force fir: for highly stressed parts or for shfink fits where the heavy pressing forces t-equiredare impractical

11. What type of fit would describe the following situations. a. The MP 01 u hall-point pen b The lead in a mechanical lead pencil, at the tip c. The bullet in a barrel of a gun 12. What does the word shrink imply in a shrink fit? 13. Why might you use ashrink fit to join the wheelsof trains to the axle rather than welding them? 14. Explain the difference between accuracy and precision. 15. When measuring time, is it moreimportant to be accurate or precise? Why? 16. Into which of the five basic kinds of inspection does interferometTy fall? 17. What facton should be considered in selecting measurement equipment? 18. Explain what is meant by the statement that usablemagnification is limited by the resolution of the device.

19. What is parallax? (Why do linesmen in t e ~ i sit s lookingq the line?) i ; 20. What is the rule of lo? 21. How does the vernier caliper work to make measurzmzB 22. What are the two most likely sources of error in us'd micrometer caliper'? !, 23. What is the major disadvantage of a micrometer calii 1, compared with a vernier caliper? J 24. What is the main advantage of a micrometer over thev caliper? 25. What would b e the major difficulty in obtaining an measurement with a micrometer depth gage if it equipped with a ratchet or friction device for thimbie') 26 Why 1 s the toolmaker's mcroscope particularly useful for; mg measurements on deltcate parts? I

I : In what two ways can linear measuremenls be made using an

b
.

optical projector? What type of instrument would you select for checking the accuracy of the linear movement of a machine tool table ' through a distance of 50 inches? What are the chief disadvantages of using a vision system for measurement compared to laser scanning? What is a CMM (coordinate measuring machine)? , What is the principle of a sine bar? : How can the no-go member of a plug gage be easily distinguished from the go member? What is the primary precaution that should be observed in using a dial gage? What tolerances are added to gages when they are being .designed'? Explain how a golno-go ring gage works for check a shaft. Why are air gages particularly well suited for gaging the diameter of a hole? , t Explain the principle of measurement by light-wave interfer-

L i E d o e s a toolmaker's flat chfEer from an optical flat7 ' Why must quality decisions derived from destructive testing be made on a statistical basis'? . What is a proof test, and what assurance does it provide? What quality-related features can a hardness test reasonably

. What exactly is nondestructive testing, and what are some

I attractive features of t h e approach?

What are some possibleobjectives of nondestructive testing? What are some factors that should be considered when se1 lecting a nondestructive testing method? : How might the costs of nondestructive testing actually bemnsidered as an asset rather than a liability? . Why should visual inspection be considered as the initial and primary means of inspection? , What is the primarylimitation of a visual inspection?

49. What is the primary materials-related Limitation of magnetic particle inspection? 50. Describe how the orientation of a flaw uith respect to a magnetic field can affect its detectability during magnetic particle inspection. 51. What is themajor limitation of sonic testing, where one listens to the characteristic ring of a product in an attempt to detect defects? 52. What is the role of a coupling medium in ultmsonic inspection? 53. What are three types of ultrasonic inspection methods? 54. What types of radiation can be used in radiographic inspection of manufactured products? 55. What are penetrameters, and how are they used in radiographic inspection? 56. While radiographs offer a graphic image that looks like the part being examined, the technique has some significant limitations. What are some of these limitations? 57. Why would we not use eddy-current inspection with ceramics o r polymeric materials? 58. What types of detection capabilities are offered by eddycurrent inspection that cannot he duplicated by the other methods? 59. Why can't acoustic emission methods be used to detect the presence of an existing but static defect? 60. How can acoustic emission be used to determine the location of a flaw or defect? 61. How can temperature be used to reveal defects? 62. What kinds of product features can be evaluated by electrical resistivity methods? 63. What type of information can he obtained through computed tomography? 64. What are some of the techniques that can he used to determine the chemical composition of surface and near-surface material? 65. Why is it important to determine the distinction between allowable and critical flaws, as opposed to rejecting all materials that contain detectable flaws?

. Read

thc 25-diviaon vernter graduated in English (Flgure 10-A)

2. Read the %-division vernier graduated in metnc (direct read~ n g (Rgure ) 10-B).

Engltsh Reading

Metric Reading

FIGURE 10-6

u
FIGURE 10-C

FIGURE 10-D

3. Convert the larger of the two readings to units o f the smaller and subtract. 4. Suppose that in Figure 10-33 the height of the gage blocks are 3.2500 in. What is the angle H assuming that the dial indicator is reading zero? 5. What is the eslimated error in lhis measurement, given that Grade 3 working gage blocks are being used? 6. InFigure 10-C.the sleeve-thimbleregionafthreemicrometers graduated in thousandths of an inch are shown. What are the readings for these three micrometers? (Hint:Tnink of the various units as if you were making change from a $10 bill. Count the figures on the sleeve as dollars, the vertical lines on the sleeve as quarters, and the divisions on the thimble as cents. Add up your change,and put a decimal poinl instead of a dollarsign i n front of the figures.) 7. Figure 10-D shows the sleeve-thimble region of two micrometers graduated in thousandths of an inch with a vernier for an addif onal ten-thousandths. What are the readings? 8. In Figure ICE, two examples of a metric vernier micrometer are shown.?he micrometor is graduated in hundredths of a millimeter (0.01 mm), and an additional reading in two-thousandths of a millimeter (0.002 mm) is obtained from vernier on the sleeve. What are the readings?

9. Progress in machine tool technology over the last 100 y has led to the continual redefinition of precision as shoa Figure 10-F, developed by Taniguchi. The trend here is clear-that precisio~~ in machiningcontinues to improve time and approaches some limit. Discuss this figure, add ing such issues as 9.1 What is the limit in machining precision? 9.2 The vertical axis of the plot uses the term accul while the curves use the term precision. Are these same? Explain. 9.3 What is nanoprocessing? Give some examples. 9.4 What is the current industrial level of precision for chine tools'! 9.5 What is the correct title for the vertical axison this Lig

- ---

--_ -

Thimble

Thimble

U)

45

45

1 0 3 --------\

Sleeve

Sleeve
FIGURE 10-E

1 nm 10-3 - - - - - - J
0.3 nm --------lmic lattice distance) 1900 1920

I
I

I I 1----

---- I - - l - - L I I I -7--rI I I

I T I I

1940

1960

1980

2000

Year
FIGURE 10-F

at processes might be grouped into the nanotechnology field? For example, what Level of precision is needed in a CD player or an artificial joint? Suppose you had a 2-ft steel bar in your supermicrometer. Could you detect a length change if the temperature of the bar changed by 20F? Rgure 10-G shows a section of a vernier caliper. What is the reading for the outside caliper?

13. For each of the inspection methodslisted below, citeone majo limitation t o its use. a. Visual inspection b. Liquid penetrant inspection c. Magnetic particle inspection d. Ultrasonic inspection e. Radiography f. Eddy-current testing g. Awustic emission monitoring 14. Which of the major nondestrnctive inspection methods might you want to consider if you want to detect (1)surface flaws and (2) internal flaws in products made from each of the following materials? a. Ceramics metal matrix (Consider various fiber materials) 15.Discuss the application of nondestructive inspection methods to powder metallurgy (metallic) products with low, average, and high densitty.

Measuring an Angle

lgure CS-lOa, shows a part drawing. After the part is made, 8 needs to be ~nspectedThe quality engineer, Kav~t, suggested the setup shown in F~gure CS-lob (No sine plate was ava~lable ) (a) Determlne the angle from the part draw~ng and the value of Xfor the stack of gage blocks

In the stack to get the total t o " X 7 (You will have to flnd a box of gage blocks.) (c) Suggest another way to check thls angle (no slne plate, or gage blocks) (d) Show the setup you would use if you had a sine plate
(b) What blocks would you use

FIGURE CS-IOA Part drawing

FIGURE C5-106 Setup for checklng the angle 0

Surface plate

FUNDAMENTALS OF CASTING
1 1 . 1INTRODU TO ~ ~MA.IFRIAL ON PROCESSING
~ ~

11.2

~TRODUCTION TO CASTING

11.3 CASTlNG TERMINOLOGY

11.4 THESOL~IFICATIONPROCES~ Cooling Curves

lltl 11.1 INTRODUCTION TO MATERIALS PROCESSING

ing might be "whatever must be done to convert stuff into things." A primary ohjcctivc of malerials processing is the production of a desired sha in the desired quantity. Shape-producing processes are often grouped into four ba

upon cooling. 7he material reniovul processes remove selected segments from an i tially oversized piece. Traditionally, thesc processes have often been relerred to

involve aspects of casting and forming. Each of the four basic families has distincl advantages and limitations, and various processes within the families have their own unique characteristics. For ex ple. cast products can have extremely complex shapes, but also possess structures al-c produced by solidification and arc therelore subject to such defects as shrinka

from simple shapes. but the joint areas are often affected by thc joining process a may possess characteristics different from the original basc material.

Family

Subgroup

Typical processes

Expendable mold Casting Multiple-use mold

Sand casting Shell castlng Investment casting Lost foam casting

<~ ~ ~ , ' i , " ~ ~ m o , d

Turning Milling Drilling Boring Sawing
r Etchina

Mechanical machining Material remaval

L Nontraditional
machining

~lectro-polishing Electrodischarse machining Water jet Laser beam Forging Rolllng Extrusion

Deformation processes Cold forinlng

-
L

W~re drawing Swag~ng Roll formlng Deep drawlng Oxyfuel

Weldjng Plasma

Consolidation processes

Resistance Laser

1 The four

xessing families, ups and typical

Soidertng Adhesive bonding Mechanical join~ng

-t

Discrete fasteners Integral fasteners Press fit Shrink fit

When selecting theprocess or processes to br used in obtaining a desired shape and achieving the desired properties"decisions 6bould be made \nth the knowledgeof &available alternatives and their associated assets and Innitations A large portion of t h s book is dedicated to presenting the various processes that can be applied to engineering materials. They are grouped according to the four basic ~ategories, with powder metallurgy being included at the end of the section on deformation processThe emphasis is on process fundamentals, deScriptions of the various alternatives, and an assessment of associated irswts and limitation9,Wewill begin with a survey of the casting ~rocesqes

JN TO

CA~TING

In the casrrng processes, a material is. fsrrst melted, heated to proper temperature, ancl sometimes treated to modify i t s chemical composition. The molten mater~al is then poured into aca'sity or mold that holds it in the desired shapeduring cool-down and solidification. Ln a single step, simple or complex shapes can be made from any material that can be melted. By proper design and process control, the resistance to working stresses can be optimized and a pleasing appearance can be produced. Cast parts range in size from a fraction of a centimeter and a fraction of a gram (such as the individual teeth on a zipper) to over 10meters and many tons (as in the huge

Fundamentals of Casting propellers and stern frames of ocean liners). Moreover, the casting processes have 8 tinct advantages when the production involves complex shape$ parts having hollow tlons or internal cavities, parts that contain irregular curved surfaces (except those can be made from thin sheet metal), very large parts, or parts made from metals that a d~fficult to machine. It is almost impossible to des~gn a part that cannot be cast by one or more of commercial casting processes. However, as with all manuf&cturingtechniques, the be results and lowest cost are only achieved if the designer understands the various o tlons and tailors the des~gn to use the most appropriate process m the most effic~ manner.The variety of cashng processes use different mold materials (sand, metal, orv ious ceramics) and pouring methods (gravity,vacuum, low pressure, or high pressu All share the requirement that the mater~al should solidify in a manner that will m a g mize the properties and avoid the formation of defects, such as shrinkage vo~ds, porosity, and trapped inclusions.

BASIC REQUIREMENTS OF CASTING PROCESSES SIXbasic steps are present m most casting processes:
1. A contalner must be produced with a mold cavity, having the desired shape and s' wlth due allowance For shrinkage of the solidifying material. Any geometncal ture desired m the finlshed casting must be present in the cavity. The mold ma must provide the desired detall and also wthstand the high temperatures an contaminate the molten material that it will contain. In some processes, a new is prepared for each casting (single-use molds) while in other processes, them made from a matenal that can withstand repeated use, such as metal or graphite. multiple-use molds tend to be quite costly and are generally employed with pro where large quantities are desired. The more eeonomlcal single-use molds are u preferred for the production of smaller quantities but may be required when cas the h~gher-melting-temperature materials. 2. A meltrng process must be capable of providing molten material at the proper te perature, in the desired quantity, with acceptable quality, and at a reasonable cost 3. Apouring technique must be devised to introduce the molten metal into the mo Provision should be made for the escape of all air or gases present in the cavi to pouring, as well as those generated by the introduction of the hot metal.The material must be free to fill the cavity, producing a high-quality casting that is dense and free of defects. 4. The solrdrficatcon process should be properly designed and controlled Castings sho be designed so that solidification and solidification shrinkage can occur withou ducing internal porosity or voids In addition,the molds should not provide exce restrmt to the shrinkage that accompanies cooling, a feature that may cause the ing to crack when it is still hot and its strength is low. 5. It must be poss~ble to remove the casting from the mold (i.e., mold removal). single-use molds that are broken apart and destroyed after each casting,mold remo presents no serious difficulty. With multiple-use molds, however, the removal o complex-shaped casting may be a major design problem. 6. Various cleuning;finishing, und inspection operations may be required after the iug is removed from the mold. Extraneous materlal is usually attached where metal entered the cavrty, excess material may be piesent along mold parting ' and mold material may adhere to the casting surface. All of these must be rem from the finished casting.

Each of the six steps will be considered in more detail as we move throng chapter. The fundamentals of solidification, pattern design, gating, and risering all be developed. Various defects will also be considered, together with their ca and cures.

[INOLOGY

Before we proceed to the process fundamentals,it is helpful to first become famih a bit of casting vocabulary. F~gure 11-2shows a two-part mold, its cross section, and ety of features or components that are present in a typical casting processTo produce mg, we begin by constructing aparrern,an approximate duplicate of the fmal castmg.M materral will then be packed around the pattern and the pattern is removed to creat part of the mold cavity.The iigid metal or wood Frame that holds the molding aggre called aflask In a horizontally parted two-part mold, the top half of the pattern mold, or core is called the cope The bottom half of any of these features is called th A core is a sand (or metal) shape that is inserted into amold to produce the intern tures of a casting,such as holes orpassages for water coolmng. Cores are produced in metal, or plastic tooltng, known as core b0xes.A coreprcnr is a feature that is added to tern, core, or mold and is used to locate and support a core within the mold.The mo terial and the cores then comb~ne to produce a completed mold cavity, a shaped ho whlch the molten metal is poured andsolidlfied to produce the desired casting.A an additional void in the mold that also fills with molten metal. Its purpose is to pro reservoir of addihonal liquid that can flow Into the mold cavity to compensate f shrinkage that occurs during solidification.By designing so the nser contains the la terial to solidify,shrinkage voids should be located in the riser, not the final castlng

Parr~ng line

Pauring cup

11-2 Cross sectlon c I t w c)-part sand mold, 1 g vi~rious mold lents,and term~nology.

Pattern

Core

Casting

lentals of Casting The network of connected channels used to deltver themolten metal to themol cavity is known as the gatrng system. The pouring cup (or pouring basm) is the portio of the gatlng system that receives the moltenmetal from the pouring vessel and contr its delivery to the rest of the mold. From the pouring cup, the metal travels down a spr (the vertical portion of the gating system), then along horizontal channels, called ru ners, and finally through controlled entrances, or gates, into the mold cavlty.Addition channels,known as vents, may be included in a mold or core to provide an escape for gases that are originally present in the mold or are generated during the pour. (Thes and other features of a gating system will be discussed later in the chapter and ar illustrated in Figure 11-9.) The parhng lrne or partrng surface is the interface that separates the cope and dr halves of a mold, flask, or pattern, and also the halves of a core in some core-maki processes. Draft 1s the term used lo describe the taper on a pattern or castmg that mits it to be withdrawn kom the mold. The draft usually expands toward the par line. Finally, the term castzng is used to describe both the process and the product wh molten metal is poured and solid~fied in a mold

3ATION PROCESS
Casting is a solrdrficanon process where the moltenmaterial is poured Into a mold then allowed to freeze tnto the desiredfinal shape. Many of (he structural features ultimately control product properties are set durtng solid~fication. Furthemore,m casting defects, such as gasporosity and solidrficanon shrinkage, are also solidif~ca phenomena, and they can be reduced or eliminated by controlling the solid~fica process. Solidificat~on is a two-stage,nucleation and growth, process, and it is important control both of these stages. Nucleation occurs when stable particles of solid form fr w~thin the molten liquid. When a material is at a temperature below its melting pot the solid state has a lower energy than the hquid. As solid~fication occur? internal ergy is released.At the same tlme, however,interface surfaces must be created betwe the new solid and the parent liquid. Formation of these surfaces requires energy. order for nucleation to occur, there must be a net reduct~on or release of energy,A result, nucleation generally begins at a temperature somewhat below the equillbr melting point (the temperature where the internal energies of the ltqutd and sol~d equal).Thedifference between the melt~ng pomt and the actual temperature of nu ation is known as the amount of undercoolmg. If nudeation can occur on some form of existing surfacehitno longer requlres creation ot a full, surrounding interface, and the requlred energy is reduced. Such faces are usually present in the form of mold or container walls, or sohd impunty p cles contained w i t h the molten 11qutd.When ice cubes are formed in a tray, the in sohd forms on the walls of the container. The same phenomena can be expected metals and other englneenng matenals. Each nucleation event produces a crystal or grain m the final casting. Since grained matenals (many small grains) possess enhanced mechanical proper ' may be made to promote nucleat~on. Particles of existmg solid may be intro the liquid before it IS poured into the mold. These particles provide the surfaces requ for nucleation and promote the formation of a umform,fine-grained product.Thisp tice of introducing solid particles is known as inoculatron or gram refinement The second stage ~nthe solidification process is growth, which occurs as the of fusion is extracted from the liqu~drnaterialThe direction, rate, and type of grow be controlled by the way in wh~ch this heat is removed. Dcrectronal solidrficat whtch the solidification interface sweeps continuously through the materia1,can be to assure the production of a sound casting.Themolten material on the liquid side o f interface can flow into the mold to continuously compensate for the shrinkage that curs as the material changes from liqutd to solid.The relative rates of nucleation growth control the size and shape of the resulting crystals. Faster rates of cooling erally produce products with finer grain size and superior mechanical properties

SECTION 11.4 The Solidification Process

Equil~brlum

freezing temperature Thermal arrest

L E 11-3 Cooling c u m lure metal or eutedlczition alloy (metals with a 't freesjng point), ting major features related dilication.

c"

-=cooling

rate

Local solidification time

Total solidificationtime Time ( t )

COOLING CURVES

Cooling curves, such as those introduced in Chapter 4, can be one of the most us tools for studying the solidification process. By inserting thermocouples into a cas and recording the temperature versus time, one can obtain valuable insight into wh happening in the various regions. Figure 11-3shows a typical cooling curve for a pure or eutectic-composition terial [one with a distinct melting point) and is useful for depicting many of the featu and terms related to solidification.Thepo~rring temperature is the temperature of the uid metal when it first enters the mold. Superheat is the difference between the pour temperature and the freezing temperature of the material. Most metals are poured temperatures of 100-200C (200-400F) above the temperature where solid begin form. The higher the superheat, the more time is given for the material to flow into intricate details of the mold cavity before it begins to freeze.The cooling rate is the at which the liquid or solid is cooling and can be viewed as the slope of the cooling cu at any given point.The thermal arrest is the plateau in the cooling curve that occurs d ing the solidification of a material with fixed melting point. At this temperature, the ergy or heat being removed from the mold comes from the latent heat of fusion th being released during the solidification process.The time Goni the start of pouring to end of solidification is known as the total solidification tinze.The time from the star solidification to the end of solidification is the local solidification time. If the metal or alloy being cast does not have a distinct melting point, such as one shown in Figure 11-4, solidification will occur over a range of temperatures. liquidus temperature is the lowest temperature where the material is all liquid, and solidus temperature is the highest temperature where it is all solid.The region betw the liquidlrs and solidus temperatures is known as the freezing runge.The onset and mination of solidification appear as slope changes in the cooling curve.

$4

11-4 Phase diagram

I
I I

panlon coollng curve bran a loy with a freez~ng !&ye. The slope changes e the onset and atlon of solidificat~on.

Cu

%Ni

NI

Time IT)
Cooling curve

Phase d~agram

The actual form of a cooling curve w l l depend on the type of materlal be poured, the nature of the nucleation process, and the rate and means of heat remo from the mold. By analyzing experimental cooling curves, we can galn valuable ins Into both the castlng process and the cast product.Fast coolmgrates and short soli cation times generally lead to finer structures and ~mproved mechanical properhe.

PREDICTION OF SOLlDlFlCATlON TIME: CHVORINOV'S RULE The amount of heat that must be removed from a casting to cause it to solidlfy depe upon both the amount of superheating and the volume of metal in the casting. Conver the ability to remove heat from a castlng is d~rectly related to the amount of exposed face area through whch the heat can be extracted and the environment surrounding molten matenal (I e., the mold andmold surroundmgs).These observations are reflec In Chvorinov'srule,' wh~ch states that the total solid~ficahon tlme, &,canbe computed
t, = B (V/A)"where n = 1.5 to 2.0

The total solidification time, t,, is the hme from paurmg to the completion of lldification; V is the volume of the castlng: A is the surface area through whch hea extracted; and B is the mold constant. The mold constant, B, incorporates the c teristlcs of the metal being cast (heat capacity and heat of fusion), the mold m (heat capaclty and thermal conductivity), the mold thickness, in~tial mold tempera and the amount of superheat. Test specimens can be cast to determine the value of B for a glven mold mat casting material, and con&t~on of casting.This value can then be used to compute solidification times for other castings made under the same conditions. Since a nser casting both lie within the same mold and fill with the same metal under the same ditlons, Chvorinov's rule can be used to compare the sohd~ficatlon times of ea thereby ensure that the riser will solidify after the casting.This condition is essential if the liquid w~thin the riser is to effectively feed the cashng and c for sohd~ficat~on shrinkage. Aspects of rlser des~gn, including the use of Chvor~n rule, will be developed later in this chapter. Different cooling rates and solidification times can produce substantla1variaho the structure and properhes of the re~ulting casting, Dle casting, for example,uses w cooled metal molds, and.the faster cooling produces higher-strength products than casting, where the moId material is more thermally ~nsulatingEven variations in the and condit~on of sand can produce different cooling rates. Sands wth hlgh moisture tents extract heat faster than ones with low mo~sture.Table11-1 presents a comparu the properties of aluminum alloy 443 cast by the three different processes of sand c (slow cool), permanent mold castmg (intermediate wolingrate), and die casting (fast

THE CAST STRUCTURE The products that result when molten metal is poured into a mold and permitted hdify may have as many as three distinct regions or zones.The rapid nucleation th curs when molten metal contacts the cold mold walls results in the production of a tone, a narrow band of randomly orlented crystals on the surtace of a casti addit~onal heat is removed, the grams of the chill zone begm to grow inward, and t of heat extrachon and sohddication decreases. Slnce most crystals have directlo

ermanenr mold
Die cast

16

33

'N Chvorinov, "Theory of Casting S o i ~ d ~ f ~ c d', f~ Glessere~, on Val

27,1940, pp 177-180,201-208,

rapid growth, a select~on process beglns. Crystals with rapld-growth direction perpendicular to the casting surface grow Past and shut off adjacent grain3 whoqe rapid-growth dvection is at some intersect& angle.The favorably oriented crystals continue to grow, producing the long, thin colu~nnar grains of a columnar tone.The properhes of this region are highly direct~anal,smce the selection process has converted the purely random structure of the surface ~ u t o one of parallel crystals of slrnilar orlentatlon. Figure 11-5 shows acast structure containing both chill and columna~ zones. In many materials, new crystals then nucleate in the interlor of the casting and grow to produce another region of spherical, randomly onenfed crystals, known as the cquiaxed zone. Low pouring temperatures, alloy addiaons and the addition of inoculants can be used to promote the formation ofthis region, whose isotropic properties (uniform in all direchons) are ar more desirable than those of columnar grams.

MOLTEN METAL PROBLEMS

Castings begin with molten metal, and there are a number of chemical reactions that can occur between molten metal and ils snrroundingsThese reactions and their products can often lead to defects m the final casting. For example, oxygen and molten metal can react to produce metal oxides (a nonmetallic or ceramic material), which can then be carried with the molten metal during the pouring and filling of the mold. Known as dross or slag, this material can become trapped m the casting and impair surface finish,machinability, and mechanical properties. Materlal eroded from the linings of furnaces and pouring ladles and loose sand particles from the mold surfaces can alsocontribnte nonmetallic components to the casting. Dross and slag can be controlled by using special precautions durrng melting and pouring, as well as by good mold design. Lower pouring temperatures or superheat slows the rate of dross-forming reactions. Fluxes can be used to cover and protect molten metaI during melting, or the melting and pouring can be performed under a vacuum or protective atmosphere. Measures can be taken to agglomerate the dross and cause it to float to the surface of the metal, where it can be sklrnmed off prior to pouring. Special ladles can be used that extract metal from beneath the surface, such as those dep~cted to trap any dross, sand, or eroded mold ting systems can be de~igned materlal and keep it from flowing into the mold cavity. In addition, ceramic filters can be inserted into the feeder channels of the mold.These filters are available in a variety of shapes, sizes, and materials. amounts of d~ssolved gas.When these maLiquid metals can also contain sign~ficant terialssolidify, the solid structurure cannot accommodate the gas, and the rejected atoms form bubbles, or gas poroszty, within the casting. Figure 11-7 shows the maximum

<
0

"loo0

0,
E m
01
72 Z

I 075 070 ml Hz/lOOg

>

Bottom pour ladle

. c

Tea pot ladle

11-6 Two types of ladles used t6 pour .Note how each extracts molten mater~al from om, avoid~ng transfer of the Impure material e top of the molten pool.

m > 0 u
Y) (O

0.50

E
ji

025 -

5
FIGURE 11-7 The maxlrnum solubrl~ty of hydrogen in alum~nurn as a functton of temperature.
0
s?d

-3

/ Melt~ng po~nt /

0 03 ml H21100g

Temperature

FIGURE 11-8 Demonstration casting made from aluminum that has been saturated in dissolved hydrogen. Note the extensive gas porosity.

solubility of hydrogen in aluminum as a function of tempcrature. Note the substan decrease that occurs as the material goes from liquid to solid. Figure 11-8shows a sn demonstration casting that has been made From aluminum that has been saturated w dissolved hydrogen Several techniques can be used to prevent or minimize the formation of gas por ity. One approach is to prevent the gas from initially dissolving in the molten met Melting can be performed under vacuum, in an environment of low-solubility gases, under a protective flux that excludes contact with the air. Superheat temperaturesc be kept low to minimize gas solubility. In addition. careful handling and pouring can much to control the flow of molten melal and minimize the turbulence that brings and molten metal into contact. Another approach is to remove the gas from the molten metal before it is po into castings. Vncuum degassing sprays the molten metal through a low-pressure e n ronment. Spraying creates a large amount of surface area, and the amount of dissolv gas is reduced as the material seeks to establish equilibrium with its new surroundin (See a discussion of Sievcrt's law in any basic chemistry text.) Passing small bubbles inert or reactive gas through the melt, known as gnu flushing, can also be effective. seeking equilibrium. the dissolved gases enter the flushing gas and are carried a w a Bubbles of nitrogen or chlorine, for example, are particularly effective in removi hydrogen from molten aluminum. The dissolved gas can also be reacted with something to produce a low-dens; compound, which then floats to the surface and can be removed with the dross or Oxygen can be removed from copper by the addition of phosphorus. Steels ca deoxidized with addition of aluminum or silicon.The resulting phosphorus, alumin or silicon oxides are then removed by skimming or are left on the top of the contain as the remaining high-quality metal is extracted froin beneath the surface.

FLUIDITY AND POURING TEMPERATURE

When molten metal is poured to produce a casting, it should first /low into all regio of the mold cavity and then freeze into this new shape. It is vitally important that the two functions occur in the proper sequence. If the metal begins lo freeze before it h completely filled the mold, defects known as misruns and cold shutc. are produced. The ability of a metal to flow and fill a mold, its "sunniness," is known as flui and casting alloys are often selected for this property. Fluidity affects the minimum tion thickness that can be cast, the maximum length of a thin section, the finenessof tail, and the ability to fill mold extremities. While no single method has been accep to measure fluidity, various "standard molds" have been developed where the results sensitive to metal flow. One popular approach produces castings in the form of a lo thin spiral that progresses outward from a central sprue.The length of the final cast will increase with increased fluidity. Fluidity is dependent on the composition, lreezing temperature, and frcezingr of the metal or alloy as well as the surface tension of oxide films.The most important trolling factor, however, is usually the pouring tenzperuture or the amount of superhe The higher the pouring tempcrature, the higher the fluidity. Excessive temperat should be avoided, however.At high pouring tempcratures, chemical reactions betw the metal and the mold, and the metal and its pouring atmosphere, are all accelerat and larger amounts of gas can be dissolved. If the metal is too runny, it may not only fill the mold cavity, but also flow into small voids between the particles that compose a sand mold.Tne surface or the resul casting then contains small particles of embedded sand, a defect known as penetrali

THE ROLE OF THE GATING SYSTEM

When molten metal is poured into a mold, the gating system conveys the material delivers it to all sections of the mold cavity.Tlie speed or rate of metal movement is portanl as well as the amount of cooling that occurs while it is flowing. Slow filling high loss of heat can result in misruns and cold shuts. Rapid rates of filling, on the o hand, can produce erosion of the gating system and mold cavity, and might result in entrapment of mold material in the final casting. It is imperative that the cross-secti

Sprue
Gates to castlng

Runner extension

Runner

into

\
Sprue well

/
Runner well

areas of the varIouscharinels be selected to regulate flow The shape and length of channels affect the amount of temperature loss. When heat loss is to be mmlmized, s channels with round or square cross sections (minimum surface area) are the most slrable.The gates are usually attached to the thickest or heaviest sections of a castin control shrinkage and to the bottom of the casting to minimize turbulence and spl ing. For large castings, rnult~ple gates and runners may he used to introduce meta more than one point of the mold cavity. Gating systems should be designed to mn~mize turbulentflow,whlch tends to p mote absorption of gases, oxidation of the metal, and erosion of the mold. Figure shows a typical gating system for a mold with a hor~zontal parting line and can be u to identify some of the key components that can be u p t i m e d to promote tbe smo flow of molten metal. Short sprues are desirable, since they minlmize the distance the metal must fall whenentering the mold and the kinetic energy that the metal acqu dnnng that fall. Rectangular pourlng cups prevent the formation of a vortex or sp ing funnel, which tends to suck gas and oxides into the sprue.Tapered sprues also vent vortex f0rmation.A large sprue well can be used to dissipate the kinetic energ the falling stream and prevent splashing and turbulence as the metal makes the t into the runner. The choke, or smallest cross-sectional area in the gating system, serves to con the rate of metal flow. I f the choke is located near the base of the sprue, flow thro the runners and gates is slowed and flow 1s rather smooth. lf the choke is moved to gates, the metal might enter the mold cavity with a fountain effect, an extremely tur lent mode of flow, but the small connecting area would enable easier separation of casting and gating system. Gating systems can also be designed to trap dross and sand particles and k them from entering the mold cavity. Given sufficient time, the lower-density wnta nants will rise to the topuf the molten metal Long, flat runners can be beneficial these promote cooling of the metal), as well as gates that exit from the lower portio the runners. Since the first metal to enter the mold is most hkely to contain the for matter (dross from the top of the pouring ladle and loose particles washed from walls of the gating system), runner extensions and wells (see Flgure 11-9) can be use catch and trap t h ~first s metal and keep it from entering the mold cavity.These feat are particularly effective with aluminum castings since aluminum oxide has appr mately the same density as molten aluminum. Screens or ceramic filters of various shapes, sizes, and materials can also be serted into the gating system to trap foreignmaterial. Wire mesh can often be used w the nonferrous metals, but cerarmc materials are generally requ~red for irons and s Figure 11-10 shows several ceramic filters and depicts the two basic types-extrude andfoam.The pores on the extruded ceramics are uniform in size and shape and pro

nentals of Casting

FIGURE 11-10 Various types

of ceramic filters that may be inserted into the gating systems of metal casting! parallel channels. The foams contain interconnected pores of various size and orien lion,forcing the material to change direction as it negotiates its passage through the ter. Since these devices can also rcstricl the fluid velocity, streamline the fluid flow, reduce turbulence, proper placement is an important consideration.To ensure remov of both dross and eroded sand, the filter should be as close to the mold cavity as po ble, but since a filter can also act as the choke, it may be positioned at other locatio such as at the base of the pouring cup, at the base of the sprue, or in one or more of runners. The specific details of a gating system often vary with the metal being cast. bulent-sensitive metals (such as aluminum and magnesium) and alloys with low m ing points generally employ gating systems that concentrate on eliminating turbule and trapping dross. Turbulent-insensitive alloys (such as steel, cast iron, and most c per alloys) and alloys with a high melting point generally use short,open gatingsyst that provide for quick filling of the mold cavity.

SOLIDIFICATION SHRINKAGE
Once they enter the mold cavity and begin lo coo1,most metals and alloys undergo ticeable volumetric contraction. Figure 11-11 shows the typical changes experience a metal column as the material goes from superheated liquid to room-temperature s There arc three principal stages of shrinkage: (1) shrinkage of rhe liquid as it c the temperature where solidification begins, (2) solidification shrinkage as th turns into solid, and (3) solid metal contraction as the solidified material cools tor temperature.

.. m
. + .

! i

_i
o

Soiidification shrinkage

0) .a ,

I :

Solid

! /

ti^^ point

Temperature

FIGURE 11-11 Dimensional changes experienced by a metal column as the material cools from a superheated liquid to a room-temperature solid. Note the significant shrinkage that occurs upon solidification.

SECTIoN 11.4 The Solidification Process

27

The amount of liquid metal contraction depends on the co ficient of thermal contraction (a property ol the metal being ca and the amount of superheat. Liquid contraction is rarely a pro lem, however, because the metal in the gating system continues flow into the mold cavity as the liquid already in the cavity cools a contracts. As the metal changes state from liquid to crystalline solid, t new atomic arrangement is usually more efficient, and signific amounts of shrinkage can occur. The actual amount of shrinka varies from alloy to alloy, as showninTable 11-2.As indicated in th table, not all metals contract upon solidification.Some actually e pand,such as gray cast iron, where low-density graphite flakes fo as part of the solid structure. When solidification shrinkage does eccur, however, it is i portant to control the form and location of the resulting void. Metals and alloys with sh freezing ranges, such as pure metals and eutectic alloys, tend to form large cavities pipes. These can be avoided by designing the casting to have directional solidificat where freezing begins farthest away from the feed gake or rlser and moves progressiv ly toward it As the metal solidifies and shrinks, the shrinkage void is continually fil with additional liquid metal. When the flow of additional liquid n exhausted and sol ~fication is complete, we hope that the final shrinkage void iilocated external to the d sired casting in eitker the riser or the gating system. Alloys with large freezlng ranges have a perlod of time when the material is in slushy (liquid plus sohd) condit~on. As the material cools between the liquidus a solidus,the relative amount of solid increases and tends to trap small, isolated pock of liquid. It is almost impossible for additional 11qu1dto feed Into these locations, and t resultant casting tends to contain small but numerous shnnkage pores dispersed throug out.Thn lype of shrinkage IS far more difficult to prevent by means of gatlng and ris ing, and a porous product may be inevitable. If a gas- or liquid-t~ght product is require these castings may need to be impregnated (the pores filled with a resinous material lower-melting-temperature metal) in a subsequent operation. Castings with dispers porosity tend to have poor duchlity, toughness, and fattgue life. After solidification is complete, the casting will contract further as it cools to roo temperature. This sohd metal contraction is often called patternmaker's contractio since compensation for these dimensional changes should be made when the moId ca ity or pattern 1s designed. Examples of thesecompensatlons w~ll be provided later in th chapter. Concern anses, however, when the casting is produced in a ngid mold, such the metal molds used in d ~ casting. e If the mold provides constraint during the time contraction, tenslle forces can be generated within the hot, weak casting, and cracki can occur (hot tears). It is often desirable, therefore, to eject the hot castings as soon solidification is complete.

RISERS AND RISER D E S I G N Rlsers are added reservoirs designed tofill with liquid metal, whch k then fed to the ca ing as a means of compensatingfor solidification shrinkage To effect~vely perform t function, the risers must solid~fy after the casting. If the reverse were true, liquid me would flow 6.om the casting toward the solidifying riser and the casting shrrnkage wou be even greater Hence, castings should be designed to produce directional solidificat that sweeps from the extrem~ties of the mold cavity toward the rlser. In this way, t riser can continuously feed molten metal and will compensate for the solidificati shrinkage of the entire mold cavity. Figure 11-12 shows a three-level step block cast aluminum with and wlthout a nser. Note that the riser is positioned so directional lidification moves from thtn to thick and the shrinkage void is moved from the casti to the riser. If a slngle directional sohdification is not possible,multiple risers may be quired, with various sections of the casting each sohdifylng toward their respectwe ns The nsers should also be designed to conserve metal If we define the yceld of a ca 1ng as the casting weight divided by the total weight of metal poured (complete gati

274

. 11 R Fundamf:ntals of Casting

FIGURE 11-12 A three-tier step-block aluminum casting made with (top) and without (bottom) a riser. Note how the riser has moved the shrinkage void external to the desired

system, risers, and casting), it is clear that there is a motivation to make the ris small as possible, yet still able to perform their task.This is usually done throug er consideration of riser size, shape, and location, as well as the type of connec tween the riser and casting. A good shape for a riser would be one that has a long freezing time. A Chvorinov's rule. this would favor a shape with srnall surface area per unit vol a sphere would make the most efficient riser, this shape presents considerab e to both patternmaker and moldmaker.The most popular shape for a riser, the cylinder, where the height-to-diameter ratio is varied depending upon the nat alloy being cast, the location of the riser, the size of the flask, and other variables. to-one height-to-diameter ratio is generally considered to be ideal. Risers should be located so that directional solidification occurs from th tremities of the mold cavity back toward the riser. Since the thickest regions of a ing will be the last to freeze,risers should feed directly into these 1ocations.Variou of risers are possible. A fop riser is one that sits on top of a casting. Because of th cation, top risers have shorter feeding distances and occupy less space within the They give the designer more freedom for the layout of the pattern and gating sy Side risers are located adjacent to the mold cavity, displaced horizontally along the ingline.Figure 11-13depicts both a top and a side riser. If the riser is contained e ly within the mold,ir is known as a blind riser. If it is open to the atmosphere, itis an open riser. Blind risers are usually larger than open risers because of the addi heat loss that occurs where the top of the riser is in contact with mold material. Live risers (also known as hot risers) receive the last hot metal that enters th and generally do so at a time when the metal in the mold cavity has already begun t and solidify. Thus, they can be smaller than dead (or cold) risers, which fill with that has already flowed through the mold cavity.As shownin Figure 11-13,top ris almost always dead risers Risers that arc part of the gating system generally live
Top

riser (open-type)

Side riser (blind-type)

FIGURE 11-13 Schemat~c of a sand casting mold, showing (a) an open-type top rlser and (b) a bl~nd-type side riser The s~de riser is a llve rlser, recelvlng the last hot metal to enter the mold. The top rlser 1 s a dead rlser, receiving metal that has flowed through the mold cav~ty.

Mold cavity
(3)

Mold cavity
I b)

The minimum size of a n s a can be calculated from Chvorinov's rule by setting t total solidificat~on tlme for the nser to be greater than the total sohdification time f the casting. Since both cavities receive the same metal and are in the same mold, the mo constant, B, wlll be the same for both regions. Assumlng that n = 2 and that a sa difference in solidification time 1s 25% (the riser takes 25% longer to solid~fy than t castmg), we can write this condition as

tr,,,
or

1.25t,,,,,,,

(11-1)

(I 1-2) (V/~)?lrer = 1-25 ( V / A ) f a u m g CalculaUon of the riser size then requires selection of a riser geometry, which a ge erally cylindrical.For a cylinder of diameter D and he~ght H, the volume and surface ar can be written as: V A
=

rrD2H/4 = TDH + 2(vD2/4)

Selecting a specific height-to-diameter ratio for the riser then enables equation 11-2 be written as a simple expre5sian with one unknown, D.The volume-to-area ratio for t casting is computed for rts particular geometry, and the equation can then be solved provide the size of the required riser. One should note that if the riser and casting sha f the common surface should be subtra a surface, as with a blind top riser, the area o ed from both components slnce it will not be a surface of heat loss to either It shou be noted that there are actually a number of methods t o calculate riser size. T Chvorinov's rule method w ~ lbe l the only one presented h e ~ e . A final aspect of riser design is the connection between the riser and the castin Since the rlser must ultunately be separated from the castlng, it is desirable that the co nection area be as small as possible. On the other hand, the connection area must be su ficiently large so that the lmk does not freeze before sol~dification of the casting complete. If the risers are placed close to the casting with relatively short connectio the mold material surrounding the link recelves heat from both the castlng and the ris Tt should heat rapidly and remain hot throughout the cast, thereby preventing solid cation of the metal in the channel

RISERING AIDS

Various methods have been developed to assist the rlsers in performing their job. Som are intended to promote directional solidificat~on, whlle others seek to reduce the nu ber and slze of the risers, thereby increasing the yield of a casting.These techniques ge erally work by e~ther speeding the solidification of the casting (chrlls) or retarding t solidification of the riser (sleeves or toppmgs). External chills are masses of hgh-heat-capacity, h~gh-thermal-conductivity ma rial (such as steel, iron, graphite, or copper) that are placed in the mold, adjacent to t casting, to absorb heat and accelerate the coollng of various regions. Chills can effective promote directional sohdificatlon or lncrease the effectlve feeding dlstance of a ns They can also be used to reduce the number of risers required for a casting. Extern chills are frequently covered with a protective wash, slllca flour, or other refractory m terial to prevent bonding with the casting. Internal chllls are pleces of metal that are placed withln the mold cavity to abso heat and promote more rapid solidificatlon. When the molten metal of the pour su rounds the chill, ~tabsorbs heat as it seeks to come to equlhbrium with its surroundin Internal chills ult~mately become part of the final castlng, so they must be made From alloy that is the same as or compatible with the alloy being cast. The cooling of rlsers can be slowed by methods that include (1) swltchmg &om blind riser to an open rlser, (2) placing ~nsulating sleeves around the riser, and ( surrounding the sides or top of the riser with exothermcc rnnter~al that suppl~es added he to just the riser segment of the mold.The objective of these techniques is generally reduce the riser size rather than promote directtonal solidification

~damentals of Casting I t isimportant to note that risers are not always necessary or functional. For alloys large h e z i n g ranges, risers would not be particularly effective, and one generally acc the fine, dispersed porosity that results. For processes such as die casting, low-pressureper maneut molding, and centrifugal casting, the positive pressures associated with the proces provide the feeding action that is required to compensate for solidification shrinkage.

Casting processes can be divided into two basic categories: ( I ) those for which a ne ~nold must be created for each casting (the expr.nduhle-~~zoldpro~e~s.~e~s) and (2) those th employ a permanent, reusuhle mold. Most of the cxpcndable-mold processes begin wit some form of reusable pattern-a duplicate of the part to be cast, modified dimcnsio ally to reflect both the casting process and the material bcing cast. Patterns can be mad from wood, metal,foam, or plastic, with urethane now being the material of choicef nearly hall of all casting patterns. The dimensional modifications that are incorporated into a pattern are called lowtmce.~, and the most important of these is the sllrinkage nllownncr. Following soli fication, a casting co~itinues to contract as it cools to room temperature, the amount this contraction being as much as 2% or 1 1 4 in./ft.To produce the desired final dimensi the pattern (which sets the dimensions upon solidification) must bc slightly lar than the room-temperature casting.The exact amount of this shrinkage compensatio which dcpcnds on the metal that is being cast, can be estimated by the cquati 1 length = length u AT, where n is the coefficient of thermal expansion and ATis t difference between the freezing temperature and room temperature.Typica1 allowan for some common cnginccring metals are: Cast iron Steel Aluminum Magnesium Brass 0.8-1 .O% 1.5-2.0'Yo 1.0-1.3% 1.0-1.3% 1.5%

Shrinkagc allowances are often incorporated into a pattern through use orsp cia1 slirink rules-measuring deviccs that are larger than a standard rule by an appr priatc shrink allowance. For example, a shrink rule for brass would designate 1 foo a length that is actually 1 foot 3/16 inch,sincc the anticipated 1.5% shrinkage will red the length by '111, inc11.A complete pattern made to shrink rulc dimensions will prod a proper size casting after cooling. Caution should be exercised when using shrink rule compensations, however, thermal contractiou may not be the only factor affecting the rinal dimensions.Thev ious phase transformations discussed in Chapter 4 are often accompanied by signific dimensional expansions or contractions. Examples include cutcctoid reaclions,malie sitic reactions, and graphitization. In many casting processes, mold material is formed around thc pattern a the pattern is then extracted to create the mold cavity. To facilitate pattern remo molds are often made in two or more sections that separate along mating surfa called the porting line or parting plane. A flat parting linc is usually preferred. but casting design or molding practice may dictate the use of irregular or multiple part surfaces. If the pattern contains surfaces that arc perpendicular to the parting li (parallel to thc dircction 01"pattern withdrawal). fl-iction between the pattcrn and mold material as well as any horizontal movement of the pattcrn during extractionco induce damage to the mold. This dalnagc could he particularly severe at the corn where the mold cavity intersects the parting surface. Such extraction damage can minimized by incorporating a slight taper, or druj't, on all pattern surfaces that arep allcl to thc dircction of withdrawa1.A slight withdrawal of the pattcrn will free it fr the mold material on all surfaces, and it can then be filrther removed without dama to the mold. Figure 11-14 illustrates the use of draft to facilitate pattern removal.

Mold cavlty

14 Two-part mold pafflng llne ari d the n of a draft 3 vertlral surfal

The size and shape of the pattern, the depth of the mold cavlty, the method used to wlthdraw the pattern, the pattern material, the mold material, and the mold~ng procedur all influence the actual amount of draft required Draft ir seldom less than lo or 118 in./ft w~th a minimum taper of about 111binch over the length of any surface. Since draft al lowances increase the slze of a pattern (and thus the sEe and we~ght of a casting), it i generally des~rable to keep them to the minimum that w~ll permit satisfactorypattern re moval. Molding procedures that produce higher-strength molds and the use of mechanl cal pattern withdrawal can often enablereduct~ons ~ndraft allowances. By reduc~ng th taper, casting we~ght and the amount of subsequent mach~nrng can both be reduced. When smooth machined surfaces are requ~red, it may be necessary to add an ad dit~onal mach~n~ng allowance,or finrsh allownnce,to the pattern.The amount of thls al lowance depends to a great exlent on the castlng process and the mold material. Ordinar sand castlngs have rougher surfaces than those of shell-mold casbngs. Die casllngs hav smooth surfaces that may requlre httle or no metal removal, and the surfaces of invest ment castlngs are even smoother. It is also important to consider the locat~on of the de sired machining and the presence of other allowances, since the draft allowance may provide part or all of the extra metal needed for mach~ning. Some castlng shapes require an additional allowance for dutortlon. Cons~der U-shaped sect~on where the arms are restra~ned by the mold at a time that the base o the U is free to shrink The result will be a findl casting w~th outwardly slop~ng arm$ I the arms are des~gned to or~ginally slope inward, however, the subsequent distortion wlll produce the deslred straight-shape casting Distorlion depends greatly on the par tlcular configuration of the castlng,and castlng designers must use experience and judg ment to provide an appropnale distortion allowance F~gure 11-15 illustrates the manner in wh~ch the varlous allowances are incorpo rated into a casting pattern Srmilar allowances are apphed to the cores that create th holes or rnterior passages of a casting. If a castlug is to be made In a multluse metal mold. all of the "pattern allowances d~scussed above should be incorpol aled Into the mach~ned cavity. The dunensions o this cav~ty will Curther change, however, as sequential casts raise the mold temperatur to an equ~hbrium level. An add~tional correct~on~hould be added to compensate fo this effect.
I

Orlglnal outllne w~th shrlnk rule

3 m m ( $ ) all around for m a c h ~ n ~ n g

V slot to be
machined

lTdraft

allowance

FIGURE 11-15 Various

a allowances incorporated ~nto castrng pattern.

Fundamentals of Cast~ng

CONSIDERATIONS IN CASTINGS
To produce the best quality designers of castings give careful atten common for mmor and simplify the casting of a One of the first fe orientation of the partzng plane, an or separable molds.The location o (2) the melhod ot supporting the (4) the welght of the final castin of molding In general, it 1 s desirable to mlmrnize the use of coresA change III the locab orlentation of the parting plane can often asslst in this objectlve.The change illust In Figure 11-16 not only e of the casting by eliminating the how a core can be eliminated b speciflaition of draft can act to f ~ the x parting plane.This figure also shows that s noting the deslred shape and th able deslgn freedom. S should also be given to t greater variat~on than those th Controlling the solidification process is of prime importance in obtaining cast~ngs, and this control high ratio of surface area to vol and harder than the other reg

Parting line

FIGURE 11-16 Ei~mlnatlon of the locabon a core by chang~ng or orientation of the partlng plane.

wlth draft permitted

by note

Parting line

FIGURE 11-17

Elimim3tion of a dry-sand core by a change in part design

FIGURE 11-18 (Top left) Deslgn where the location of the

Part wlth dra parting plane IS specified by the draft (Top r~ght) unspecified. (Bottom) Various optlons to produce the top-rlgh part, lncludlng a no-draft deslgn

rn/
11-19 Typical e s forsection change ns in castfngs.
then

r = Dl3

Blend in

l f D > 15"andd<2013,

r = Dl3 w ~ t h a 15" slope between

the two parts

IfD>15and d>213 D,then r = Dl3

IfDd15"and

d>213D,then r = 112"

may contain shrinkage cavities and porosity, or have weaker, large grain-size structu Ideally, a casting should have uniform thickness at a11 locations. Instead of thicker tions, ribs or other geometric features can often be used to impart additional stren while maintaining uniform wall thicknessWhen the section thickness must change,it is b if these changes are gradual, as indicated in the recommendations of Figure 11-19. When sections of castings intersect, as in Figure 11-20a. two problems can ar The first of these is stress concentration. Generous fillets (inside radii) at all interior ners can better distribute stresses and help to minimize potential problems, includ shrinkage cracks. If the fillets are excessive, however, the additional material can a

11-20 (a)The"hot sedion rz i s caused by jng sections. (b) A n

Met and exterior radius nore uniform th~ckness e un~form cool~ng.

ment thesecond problem,known as hotspvts.Tnick secrious, hke those at the inters tion in Figure 11-20a and those illustrated in Figure 11-21, cool more slowly than ot locations and tend to be sites of localized ~hnnkage. Shrinkage voids can be sites ot s sequent failure and should be prevented IE at all possible. Where thlck sections m exist,an adjacent riser is offen used ro feed the section during solidification and shri age. If the riser is designed properly, the shrinkage cavity will lie totally within the ri as illustrated in Figure 11-22,and will be removed when the riser is cut off. Sharp e rior corners rend to cool faster than the other sections of a casting. If an exterior rad is provided, the surface area can he reduced and cooling slowed to be more cons~ w~th the surrounding material. Figure 11-20b shows a reeommended modification Figure 11-20s.

11-27
i .

Hot spots often rewlt from lntersectlng sections of varlous

FltURE 11-22 Attached risers can move shr~nkage cavlty external to the actual casti

Bad

[m!
Better
FIGURE 11-23 Uslng staggered ribs to prevent

crack~ng dur~ng cool~ng.

When sectlons intersect to form contrnuous ribs, l ~ k those e x n Fxgure 11-23,c tractlon occurs in opposite dtrections as each of the arms cools and shr~nks As ac sequence, cracking frequently occurs at the intersectrons. By staggering the rib8 shown in the second portion of F~gure 11-23, there 1s opportunity for distort~o~ occur that would provide relaxation to the high residual stresses that m~ght othen Induce cracking The locatlon of the parting line may also be an appearance consideration. As1 amount of fin, or flash,is often present at the partrngline. and when the flash is rema (or left in place if it 1s small enough), a line of surface imperfection results. If the b tion is in the mrddle of a flat surface, a will be clearly vlsrble rn the product If the p ing line can be moved to coincrde with a corner, however, the assoc~ated "defect" go largely unnoticed. Thm-walled castmgs are often desired because of the~r reduced wetght, but walls can often oresen1 manufacturingproblemsrelated to mold fillmg (pren n set ing before complete fill). M~ntmur designing castings Specrfic values are with the shape and size of the casting, the type of metal being cast, the method ofc ing, and the practice of an individual foundry. Table 11-3 presents typical minim thickness values for several cast materials; casting processes. Zinc die casting cannow] duce walls as thin as 0.5 mm.

Castlnz Method
bana casung

- .

M~nimum Section Th~ckness [mm) Aluminum Magnesium Steel

puter simulatron. The mathematics of fluld I

modeling. The mathematical tools of mite , , , ment or fm~te dlfft pled with the use or nrgn-speea computer permit beneficial design changes beforu manufacture of patterns or molds

- ." 15
i

1 YO

.",
3.18

4 13

"

qr

Permanent mold Dle cast Investment cast Plaster mold

2 36

157 157 2 03

2 35 157

2 36
-

ltl 11.7 THE CASTING INDUSTRY

The US. metal-casting industry ships over 14 nnllion pounds of castlngs every year,; erav " ,iron. ductile iron. aluminum allovs. and coouer-base metals eomprrsine them portlon Thirty-five percent of the manufacture. The average 2005 pounds) of iron castlngs and 115 kg (250 pounds) of cast alummum. Magnesium cast are also beginning to achieve a presence. Metal castings form primary components in: ag~~cultural implements; const tlon equipment; mining equ~pment; valves and fittings; metalworking machinery;pc tools; pumps and compressors, r d r o a d equipment;power transm~ssion equipment; heating, refrigeration, and alr-conditioning equipment Ductlle Iron prpe is a maid for conveymg pressurized fluids, and household appliances and electronlcs all ut metal castings.

$ I I
allowance hl~nd rner castlng
chill

Key Words
coohng rate cope core cose box core prlnt dead rlser
deformation processes direct~onal soltdrfication distortion draft drag dross

chill zone

choke

Chvor~nov's rule cold shut columnar zone computer sirnulatlon consohdatlon processes cooling curve

process

insulating sleeve internal chill liquidus live riser local solidification time machining machining allowance material removal materials processing misruns mold cavity mold constant mold material multiple-use mold nucleation oven riser parting lme (parting surface)

pattern penetration pouring cup pouring temperature powder metallurgy reuseahle mold riser runner runner extension shrink rule shrinkage allowance side riser single-use mold slag sleeves solidification solidification modeling

sohd~ficationshrinkage solidus sprue sprue well stre% concentrators superheat thermal arrest top riser total sohdlfication tlme turbulent flow undercoolmg vacuum degassing vent yield

I I I Review Questions
25. What is dross or slag, and how can it be prevented froni becoming part of a finished casting? 26. What are some of the possible approaches that can be taken to prevent the formation of gas porosity in a metal casting? 27. What is fluidity, and how can it be n~easured? 28. What is a misrun or cold shut, and what causes them to form? 29. What defect can form in sand castings if the pouring temperature is too high and fluidity is too great? 30. Why is it important to design the geometry of the gating system to control the rate of metal flow as it travels from the pouring cup into the mold cavity? at types of problem or defect can occur if the mold material 31 What are some of theundesrrable consequences that could result &om turbulence of the metal m thegatlngsystem andmold ides too much restraint to the solidifying and cooling metal? cavity? at is a casting pattern? Flask? Cote? Mold cavity? Riser? at are some of the components that c o m b i ~ ~ toemake up 32. What is a choke,and how does its placement affect metal flow? 33. What features can be incorporated inlo the gating system to gating system of a mold? at is a parting line o r parting surface? aid in trapping dross and loose mold material that is flowing at is draft and why is it used? with the molten metal? at are the two stages of solidification, and what occurs dur- 34. What features of the metal being cast tend to influence whether the gatingsystemis designed to minimize turbulence i s it that most solidification does not begin until the temand reduce dross or promote rapid filling to minimize temperature loss? ture falls somewhat below the equilibrium melting tem35. Whatare the threestages of contraction or shrinkage as a liqrature (i.e., undercooling is required)? uid is converted into a finished casting? sirable to promote nucleation in a casting 36. Why is it more difficult toprevent shrinkage voids from forminoculation or grain refinement processes? ing in metals o r alloys with large freezing ranges? enegus nucleation begins at preferred sites withimamold. 37. What type of flaws o r defects form during the cooling of an tare some probable sites for heterogeneous nucleation? already-solidified casting? might directional solidification b e desirable in the pro38. Why is it desirable to design a casting to have directional 'on of a cast product? solidification sweeping from the extremities of a mold toward scribe some of the key features observed in the cooling a riser? rve of a pure metal. 39. Based on Chvorinov'srule, what would be an ideal shape for a casting riser? A desirable shape from a practical perspective? at is the freezing range for a metal or alloy'? uss the roles of casting volume and surface area as they re- 40. What is "yield,"and how does it relate to the number and size of the specified risers? e to the total solidification time and Chvorinov's rule. at characteristics of a specific casting process are incorpo- 41. Define the following riser-relaled terms: top riser, side riser, d into the mold constant, B, of Chvorinov's rule? vpen riser, blind riser, live riser, and dead riser. at is the correlation between cooling rate and final prop- 42. What assumptions were made when using Chvorinov's rule to calculate the size of a riser in the manner presented in rties of a casting? the text? Why is the mold constant, B, not involved in the at is the chill zone of a casting, and why does it form? calculations?' ich of the three regrons of a cast structure is least desir7 Why are its propertres h~ghly directional7
tare the six activities that are conducted onalmost every ufactured product? at is "materials processing"? at are the four basic families of shape-production processCite one advantage and one limitation of each family. rihe the capabilities of ihe casting process in terms of size hape of the product. might the desired production quantity influence the seon of a single-use or multiplc-use molding process? y is it important to provide a means of venting gases from

282

CHA~ER 11

Fundamentals of Casttng
48. Why isit desirable to make the pattern allowances as sm

43. Discuss aspects rclating to the connection between a riser and the casting.
45. What types o l modifications or allowances are generally incorporated into a casting pattern? 46. What is a shrink rule, and how does it work? 47. What is the purpose of a draft or taper on pattern surfaces?

directly related to the location of the parting plane? 50. Whal are "hot spots" and what sort of design features them to form? 51. Are metal castings used in passenger cars and light trucks what extent?

111 Problems
I. Using Chvorinov's rule as presented in the tex1;with n = 2,calculate the dimensions of an effective riser for a casting that is

of the surface of the casting, and recompute the size and

fraction of the combined weight of the riser and casting? 2. Reposition the riser in Problem 1 so that it sitsdirectly on top of the flat rectangle,withits bottom circular surface being part

drilling industry, using an air-set or no-bake sand for both the mold and the core. A silica sand has been used in combination with a catalyzed alkyd-oillurethane binder. The figure shows a cross section of the mold with the core in place (a) and a cross section of the finished casting (b). The final casting contains several significant defects. Gas bubbles are observed in the bottom section of the horizontal tee. A penetration defect is observed near the bottom of the inside diameter, and there is an enlargement of the casting a t location C. 1. What i s the most likely source of the g a s bubbles? Why are they present only a t the location noted? What might you recommend as a solution?

2. What factors may have caused the penetration defect Why is the defect present on the inside of the casting but not on the outside? Why is the defect near the bottom o f the casting but not near the top? 3. what factors led to the enlargement of the casting a t point C? whatwould you recommend to correct this problem?
4, Another producer has noted penetrationdefects on all surfacesof his castings, both interiorand exterior,What would be some possible causes? What could you recorn mend as possible cures? 5 , Could these moldsand cores be reclaimed (i,e,,recycle breakout? Discuss,

PENDABLE-MOLD CASTING PROCESSES

No-Bake, Air-Set, or Chemically Bonded Sands Shell Molding Other Sand-Based Molding Methods 12.3 CORES AND CORE MAKING 12.4 OTHER EXPENDABLI-MOLD W m MULTIPLE-USE PROCESSES PATTERNS Plaster Mold Casting Ceram~c Mold Castmg Expendable Craphite Molds Rubber-Mold Casting
125 EXPENDABLE-MOLD PROCESSES USING SINGLE-USE PATTERNS Investment Casting Counter-Gravity Investment Cast~ng Evaporative Pattern (Full-Mold and Lost-Foam) Casting 12.6 SHAKEOUT, CLEANING,AND F~ISHING 12 7 SUMMARY Case Study: MOVABLE AND FIXED PIECES FORA HEAVY-DUTY JAW BENCH VISE

Sodium Silicate-CO, Molding

2.1 INTRODUCTION
The versatility of metal castmg ib made possiblc by a number of d~stinctiy different processes, each w ~ t h its own set of charactenst~c advantages and benefits. Selection of the best process requlres a famihar~atlon with the varlou5 opttons and capabiht~es as well as an understanding of the needs of the spec~fic product. Some [actors to be cons~dered include the desired d~mensional precision and surface quality, the number of castlngs to be produced, the type of pattern and core box that will be needed, the cost of maklng the requ~red mold or d ~ eand , rcstrictions Imposed by the selected material. As we begin to survey the various casting processes. it is helpful to have some form of process classilicat~on. One approach focuses on the molds and patterns and utilizes the following three categories. 1 . Single-use molds with multiple-use patterns 2. Single-use molds with single-use patterns 3. Multiple use molds Categories 1 and 2 are often combined under the more general headmg expendable-mold castrngprocesses,and these processes w~ll be presented m this chapter. Sand, plaster,ceramlcs, or other refractory materials are combmed w ~ t h binders to form the mold. Those processes where a mold can be used multiple tlmes will be presented m Chapter 13.The multiple-use molds are usually made from metal. Since the castlng processes are primar~ly used to produce metal products the emphasis o f thc casting chapters will be on metal castlng The metals most frequently cast are aou, steel, stainlesssteel, aluminum alloys, brass, bronze and other copper alloys,magnesium alloys, certain zinc alloy&and nickel-based superalloys.Among these, cast Iron and aluminum are the most common, primarily because of their low cost, good tlu~&ty, adaptability to a variety of processes, and the wide range of product properties that are available. The processes used to fabricate products from polymers, ceramics (~ncluding glass), and composites, including castmg processes, w11l be discussed in Chapter 15

k2.2 SANDCASTING
Sand casting 1s by far the most common and possibly the most versatile of the casting processes, accounting for over 90% of all metal cashngs. Granular refractory material (such as sihca, zircon, ohvine, or chromte sand) is mixed w~th small amounts of other

283

I
materials, such as clay and water, and is then packed around a pattern that has the
of the desired casting. Because the grams can pack into thin sections and can be eco lcally uaed in large quantities, products spanning a wrde range of slres and detail ca made by thlq method. If the pattern is to be removed before pouring, the mold I made in two or more segments. An opening called a sprue /tole 1 s cut from the tl mold through the sand and connected to a system of channels called runners.Th metal i s poured down the sprue hole, flows through the runners, and enters then ity through one or more openings, called gates. Grav~ty flow is the most commc of introduc~ng themetal into the mold.?he metal 13 allowed Lo sohdrfy, and the mo broken t o permit removal of the f~nibhed casting. Because the mold ISdestroyed uct removal, a new mold must be made for each casting. Figure 12-1 shows the

FIGURE 12-1 Sequential steps in making a sand casting. (a) A pattern board is placed between the bottom (drag) and top (cope) halves of a flask, with the bottom side up. (b) Sand is then packed into the bottom or drag half of the mold. (c) A bottom board is positioned on top of the packed sand, and the mold is turned over, showing the top (cope) half of pattern with sprue and riser pins in place. (d) The upper or cope half of the mold is then packed with sand. (e) The mold is opened, the pattern board is drawn (removed), and the runner and gate are cut into the bottom parting surface of the sand. (e') The parting surface of the upper or cope half of the mold is also shown with the pattern and pins removed. (f) The mold is reassembled with the pattern board removed, and molten metal is poured through the sprue. (g) The contents are shaken from the flask and the metal segment is separated from the sand, ready for further processing.

SECTION 12.2

Sand Casting

285

steps and basic components of a sand casting process. A two-part cope-and-drag mold is illustrated. and the castine incornorates both a core and a riser (discussed in C h a ~ t e 11). r

PATTERNS A N D PATTERN MATERIALS

The first step in making a sand casting is the design and construction of apattevn. This is a duplicate of the part to be cast: modified in accordance with the requirements of the casting process, the metal being cast, and the particular molding technique that is being used. Selection of the pattern material is determined by the number of castings to be made, the size and shape of the casting, the desired dimensional precision, and the molding process. Wood patterns are relatively easy to make and are frequently used when small quantities of castings are required. Wood, however, is not very dimensionally stable. It may warp or swell with changes in humidity, and it tends to wear with repeated use. Metal patterns are more expensive but are more stable and durable. Hard plastics, such as urethanes, offer another alternative and are often preferred with processes that use strong, organically bonded sands that tend to stick to other pattern materials.In the full-mold and lost-foam processes, expanded polystyrene (EPS) is used, and investment casting uses pattcrns made from wax. In the latter processes, both the pattern and the mold are single-use, each being destroyed when a casting is produced.

TYPES OF PATTERNS
Many types of patterns are used in the foundry industry. with selection being based on the number of duplicate castings required and the complexity of the part. One-piece or solidpatterns, such as the one shown in Figure 12-2, are the simplest and often the least expensive type.They are essentially a duplicate of thc part to be cast, modified only by the various allowances discussed in Chapter 11 and by the possible addition of core prints. One-piece patterns are relatively cheap to construct, but the subsequent molding process is usually slow. As a rcsult, they are generally used when the shape is relatively simple and the number of duplicate castings is rather small. If the one-piece pattern is simple in shape and contains a flat surface, it can be laced directlv on a follow hoard.The entire mold cavitv will bc creatcd in one segment " of the mold, with the follow board forming the parting surface. If the parting plane is to be more centrally located, special follow boards are produced with inset cavities that position the one-piece pattern at the correct depth for the parting line. Figure 12-3 illustrates this technique, where the follow board again forms the parting surface. Split patterns are used when moderate quantities of a casting are desired. The pattern is divided into two segments along what will become the parting plane of the mold. The hottom segment of the pattern is positioned in the drag portion of a flask, and the bottom segment of the mold is produced. This portion of the flask is then inverted, and the upper segment o l the pattern and flask are attached.Tapered pins in the cope half of the pattcrn align with holes in the drag segment to assure proper positioning. Mold material is then packed around the full pattern to form the upper segment (cope) of the mold.The two segments of the flask are separated, and the pattern pieces are removed to produce the mold cavity. Sprues and runners are cut, and the mold is then reassembled, ready for
Pattern

RE 12-2

Single-piece

rn for a pinion gear.

Pattern

+4,
Follow board

Method of using a follow board to position a single-piece f the pattern and locate a parting surface. The final figure shows the flask o previous operation (the drag segment) inverted in preparation for construction of the upper portion of the mold (cope segment).
FIGURE 12-3

CHAPIZR 12 Expendable-Mold Castinq Processes

light-colored port~ons are core prints

opposite sides of a single-pattern board.)

12-4 shows a split pattem that also contams several core prints (lighter pour. F~gure Match-platepatterns,hke the one shown in Figure 12-5,further simphfy and can be coupled with modern molding machines to produce large quantities plicate molds.The cope and drag segments of a split pattem are permanently fasts to opposlte sides of a wood or metal matchphte.The match plate is positioned bet\$ the upper and lower flask segments. Mold material is then packed on both sides of match plate to Corm the cope and drag segments of a two-part mold. The mold secti are then separated and the match-plate pattern is removed.The plns and gulde holes sure that the cavities in the cope and drag will be in proper alignment upon reasseq The necessary gates, runners, and risers are usually incorporated on the match plal well. This guarantees that these features will be uniform and of the proper size ~nf mold, thereby reducing thepossibiity of defects Figure12-5 further illustrates theo man practlce of including more than one pattern on a single match plate. I When large quantities of identical parts are to be produced, or when the cash; quite large, it may be desirable to have the cope and drag halves of split patterns attad I to separate pattern boards. These cope-and-drug putterns enable independent moldii the cope and drag segments of a mold. Large molds can be handled more easily in d I rate segments, and small molds can be made at a faster rate if a machine 1s only pro1 ing one segment.Figure 12-6 shows the mating pieces of a typical cope-and-drag pat I When the geometry of the product 1s such that a one-piece or two-piece pa could not be removed from the molding sand, a loose-piece pattern can sometnnes bd

ulagral

and ba muller. loosen wheels cornpc

lnterno

12.2 SECT~ ~ N Sand Casting

287

veloped Separate pieces are joined to a pnmary pattern segment by beveled grooves or prns (Figure 12-7).After molding, the prlmary segment of the pattern 1s withdrawn.The hole that 1s created then permits the remainmg segments to be sequentially extracted Loose-p~cce patterns are expensive.They requlre careful maintenance, slow the molding process, and increase moldlng costs They do, however, enable the sand casting of complex shapes that would otherwise require the full-mold, lost-foam, or mvestment processes.

12-7 Loose-piece a large or rnold~ng ear. After suffic~ent sand een packed around the rn to hold the pleces in ion, the wooden pins are rawn The mold 1 s then leted, after wh~ch the of the pattern can be ved in a desiqnated

SANDS AND SAND CONDITIONING The sand used to make molds must be carefully prepared if rt is to provide satisfactory and unitorm results. Ordinary silica (SiO,), zircon. ohvlne, or chromlte sands are compounded with add~tives to meet four requirements:
1. Refi~~ctonness: the ablllty lo withstand hlgh temperatures without melting, fracture, or deterloration 2. Cohescveneas (also referred to as bond): the ability to retain a glven shape when packed Into n mold 3. Permeability: the ability of mold cavity,mold, and core gaves to escape through the sand 4. Collnps~bil~tj: the ability to accommodate metal shrinkago after solidification and provide for easy removal of the casbng through mold disintegration (shakeout) Refractormess is provided by the baslc nature of the sand Cohes~veness, bond, or strength is obtained by coating the sand grains wlth clays, such as bentonite, kaolinite, or ilhte, that become wheslve when moistened Collapsibtllty 1s sometimes enhanced by addlng cereala or other organic materials, such as cellulose, that burn out when they come m contact with the hot metal.The combustion of these materials reduces both the volume and strength of therestrainingsand Permeablhty is afunction of the slzeof the sand particles, the amount and type of clay or bond~ng agent, the molsture content, and the compacting pressuie. Good molding sand always represents a compromlse between competing factors. The size of the sand particles, the amount of bonding agent (such as clay), the moisture content, and the organlc additives are all selected to obtain an acceptable compromlse among the four basic requuements The overall composition must be carefully controlled to ensure satisfactory and consistent results Slnce molding material is often reclaimed and recycled, the temperature o l the mold during pounng and solidification is also important. If organic matenals have been incorporated into the mix to provide collapsibility, a portion w~ll burn during the pour.Adjuslments will be necessary, and ultimately some or all of the mold matenal may have to be discarded and replaced wlth new. A typical green-sand mlsture contains about 88% silica sand, 9% clay, and 3% water. To achieve good molding, it is Important for each grain of sand to be coated uniformly with the proper amount of additlve agents. This 1s achieved by puttlng the Ingredients through a muller, a device that kneads, rolls, and stirs the sand.Figure 12-8 shows both a continuous and batch-type mulier, with each producing the deslred mixing
Hopper

Sand entrance

Plow blade (1 of 2) Muller wheel (1 of 2)

Plow blade
Sand exlt
11 of 2)

~le-Mold Cast~ng Procesres through the use of rotating blades that l&,fluff,and red~stribute the material and wh that compress and squeeze. After mixing, the sand is often discharged through an a tor, which fluffs it for further handling.

SAND TESTING If a foundry is to produce high-quality products, it is important that it maintain a

sistent quality ~n~ tmolding s sand.The sand itself can be characterized by grain sue, shape,surface smoothness,density,and contaminants.Blended moldmg sand can be acterued by moisture content, clay content, and cornpactu6rlrty. Key properties of c packed sand or finished molds include mold hardness,permeabdzty,and strength Stan tests and procedures have been developed to evaluate many of these properties. Grain size can be deterrmned by shaking a known amount of clean, drys downward through a set of 11 standard screens or sieves of decreas~ng mesh size. being shaken for 15 minutes, the amount of materlal remaining on each sieve is we1 and these weights are used to compute an AFS (American Foundry Society) grain ness number Mowture content can be determined by a special device that measures the el cal conductivity oi a small sample of compressed sand. A more direct method is to sure the weight lost by a 50-gram sample after it has been subjected to a tempera of about llO0C(23O0F)for sufficient hme to drive off all the water. Clay content is determined by washing the clay from a 50-gram sample of ing sand, uslng water that contains sufficient sodium hydrox~de to make it alkaline. era1 cycles of agitation and washing may be required to fully remove the clay remaining sand is then dned and weighed to determine the amount of clay remo from the or~ginal sample. Permeability and strength tests are conducted on compacted sands, using a s dard rammedspectrnen. An amount of sand is first placed into a 2-inch-diameter tube. A 14-pound we~ght is then dropped on it three tlmes from a helght of 2 inches the height of the resulting specimen must be within & inch of a targeted 2-lnch he' Permeabrlrty 1s a measure of how easily gases can pass through the narrow v between the sand grainb.Air in the mold before pounng,plns the sleam that is prod when the hot metal contacts the moisture in the sand along with various comhu gases,must all be allowed to escape,rather than prevenl mold filling or be trapped in casting as poros~ty or blowholes. During the permeability test, shown schemat~ Figure 12-9, a sample tube containrng the standard rammed specimen is subjecte air pressure of 10 g/cm2. By means of either a flow rate determination or a mea ment of the steady-state pressure between the orifice and the sand specunen, an AFS meab~lity numberi can be computed. Most test devices are now calibrated to prov direct readout of this number. Sand Pressure measured between ortflce and sand Caltbrated ortfice
0 rlng

Ramming cyl~nder

FIGURE 12-9 Schemat~c of a permeab~llty tester m operation. A standard sample in a metal sleeve o sealed by an 0-nng onto the top of the unlt wh~le alr is passed through the sand. (Courtesy of D~etert Equipment Foundry Test~ng Inc, Detroit, MI)

orced in under a constant pressure

'The AFS permeablllty number isdefined as
AFS number
=

(V

H)/(P X A

T)

where Vts the volume of mr (2000 cm')),H~s the he~ght of thespecimen (5 08cm),P1s the pressure (10gi A IS the cross-aectlon area of the speclrnen (20 268 c m ' ) . and T ir the time m seconds to pass a flow m' of arr through thcspec~men Substituhng each aCihe above constants, thepermeab~hty number be equalto 3000 ZIT

12-10 Sand mold

tester. (Courtesy of

All mold~ng matenal must have suffic~ent strength to retaln the integrity of the mold cavity while the mold is being handled between molding and pounng. The mold material must also withstand the erosion of the hquid metal as it flowsinto the mold and the pressures induced by a column of molten metal. The compres~ive strength of the sand (also referred to as green compressive ~trength) is a measure of the mold strength at this stage of processing. It is delerm~ned by removing the rammed specimen from the compacting tube and placing it in a mechanical test~ng dev~ce A compressive load is then applied until the specimen breaks, whlch usually occurs m the range of 10 to 30 psi (0.07 to 0.2 MPa). If there is too little moisture in the sand, the grams w~ll be poorly bonded and strength will be poor. If there 1s excess moisture, the extra water acts as a lubncant and strength r s agam poor. In between, there is a condition of maxlmum strength with an optimum water content that will vary w ~ t h the content of other materials in the mix. A simllar optimum also applies to permeabihty, slnce unwetted clay blocks vent passages, as does excess water. Sand coated with a uniform thin film of mom clay provides the best molding properties A ratlo of one part water to three parts clay (by weight) is often a good startmg point. The hardnesc. of compacted sand can give addit~onal Insight into the strength and permeablhty characteristics of a mold. Hardness can be determ~ned by the resistance of the sand to the penetration of a 0.2-in. (5 08-mm)-d~ametar spnng-loaded steel ba1l.A typical test instrument is shown in Figure 12-10. Compactibility is determined by siftlng loose sand lnto a steel cylinder, leveling off the column, striklng it three times with the standard weight (as in making a standard rammed specimen), and then measuring the final he~ght The percent compactrbzlcty is the change in height div~ded by the onginal height, tlmes 100%. This value can often be correlated m t h the moisture content of the sand, where a compact~bility of around 45% indicates a proper lcvel of moisture. A low compactibihty is usually associated with too little moisture.

SAND PROPERTIES AND SAND-RELATED DEFECTS The characternst~cs of the sand granules themselves can be very influentla1 in determin~ng the properties of foundry molding material. Round grains glve good permeability and minimize the amount ol clay requlred because of their low surface area.Angular sands glve better green strength because of the mechan~cal mterlock~ng of the grams Large grains provlde good permeab~llty and better reststance to high-temperature melting and expansion,while fine-grained sands produce a better surface finish on the final casting. Un~form-siw sands glve good permeabllity, whiie a distribution of sizes enhances surface f i s h . S~lrca sand is cheap and hghtweight,but when hot metal is poured into a silica sand mold, thesand becomes hot, and at or about 585C (1085F) ~tundergoes a phase transformation that is accompanied by a substantial expansion in volume. Because sand is a poor thermal conductor, only the sand that is adjacent to the mold cavity becomes hot and expands. The remaining mater~al stays fairly cool, does not expand, and often prov~des a h~gh degree of mechan~cal restraint. Because of t h ~ uneven s heating, the sand at the surface of the mold cavity may buckle or fold. Castings with large, flat surfaces are more prone to sand expansion defectv since a considerable amount of expanslon must occur m a s~ngle direction. Sand expansion defects can be minimized in a number of ways. Certain particle geometries permlt the sand grams to shde over one another, thereby reliev~ng the expansion stresses. Excess clay can be added to absorb the expansion, or volatile additlves, such as cellulose, can be added to the mix. When the casting n poured, the cellulose burns, creating vo~ds that can accommodate the sand expanslon. Another alternative is the use of oliv~ne or zircon sand in place of silica. Since these sands do not undergo phase transformatlans upon heating, their expansion 1s only about one-half that of s ~ h ca sand. Unfortunately, these sands are much more expensive and heavier m weight than the more commonly used silica. Trapped or evolved gas can create gas-related vords or blows in finished castings. The most common causes are low sand permeabllity (often associated with angnlar,f~ne, or wide-size distribut~on sands, fine sand addit~ves, and overcompaction) and large

290

C H A ~12 R Ex~endable-Mold Castina Processes

* m a
a*e

amounts of evolved gas due to hi mold-material moisture or excessi amount5 of volatiles. If adjustments 1 Is lwxpenstve In bulk quanmties the mold composition are not suffici 2 Retains properties through transportaoon and storage to eliminate the voids, vent pas 3 Umfarrniy f~lls a flask or contamer may have to be cut into themold, 4 Can be compacted or set by simple methods cedure that may add significantly 5 Hassuff~clent eladlclty to rcmaln undamaged durrng pattern u?rthdrawal cost. mold-maklng 6 Canwrthetand high temperatures and mamiam, its d ~ m e n s ~ o n unul s the metal has sol~dlfied Molten metal can also penetr 7 Is sull~ciently perrneablc to allow the escape o f gases between the sand grains, causlng 8 Is suffimently den% to prevent metal penettauon mold material to become embedded 9 Is suffiaently coheslve to prevent wash-out of mold mater~al into the pour stream the surface of the eastlng This def 10 1schemlcally Inert to the metal belng cast known as penetratLon, can be the re 11 Can y ~ e l d to sohdlbcatlon and thcrmal shnnkege,thereby prevenhng hot tears and cracks of high pouring temperatures (exc 12 Has good collapslb~hty t o permxt easy removal andseparatlon of the castlng flludity), hlgh metal pressure (pas ' 13 Can ba recycled due to excessive cope helght or pou from too high an elevation above mold), or the use ot hlgh-permeability sands with coarse, uniform particles. Fine-gra materials, such as silica flour, can be blended in to fill the voids, hut this reduces meabllity and increases the likelihood of both gas and expansion defects. Hot tears or cracks can form in castmgs made from metals or alloys with 1 amounts of sohdification shrinkage. As the metal contracts during solidificatton cooling to room temperature, it may find Itself restrained by a strong mold or core. sile stresses can develop while the metal is still partially llquid or fully solidified but hot and weak If these stresses become great enough, the castrng will crack. Hot tears often attributed to poor mold collapsibility. Addit~ves, such as cellulose, can be use improve the collapsib~lity of sand molds. Table 12-1 summarizes the many desirable properties of a sand-based m ing material.
.--ve.a.-a

ir.r.rrr
e?,
fb

THE MAKING OF SAND MOLDS

When only a few castlngs are to be made, hand ramming is oftcn the preferred me of packing sand to make a sand mold. Hand ramming, however, is slow, labor inte and usually results m nonumform compaction. For normal production, sand mold generally made using specially designed molding machines The various methods m the type of flask, the way the sand is packed within the flask,whether mechan~ slstance is provided to turn or handle the mold, and whether a flask is even require all cases, however, the molding machnes greatly reduce the labor and required and also lead to castings wlth good dimensional accuracy and consistency. Molding usually begins \nth a pattern, like the match-plate pattern discussed e and aflnsk. The flasks may be straight-walled containers with gu~de plns or rem jackets, and they are generally constructed of lightweight aluminum or magnes F~gure 12-11shows a snap flask, so named because it 1s designed to snap open for removal after the mold matenal has been packed in place. The mixed sand (mold material) can be packed in the flask by one or more techniques. A sand slinger uses a rotating impeller to fllng or throw sand agains

FIGURE 12-11 Bottom and top halves of a snap flask. (Left) drag segment in closed postion (r~ght) cope regment w~th latches opened for easy removal

S~cnoN 12.2 Sand Casting

291

Platen

Jolting
table

/ ; I

/Table

stops

where the greatest packing a expected.)

FIGURE 12-12 Jolting a mold section. (Nofe.The pattern 1 s on the bottom,

FIGURE 12-13

Squeezing a sand-f~lled mold sectlon Wh~le the pattern 1 s on the bottom, the h~ghest packrng will be d~rectly under the squeeze head

pattern. The slinger is man~pulated to progressively deposlt compacted sand into the mold. Sand slinging 1s a common method of achieving uniform sand compaction when making large molds and large castings In a method known as jolting, a flask is positioned over a pattern, filled with sand, and the pattern, flask, and sand are then lifted and dropped several times, as shown m Fignre 12-12.The we~ght and kinetic energy of the sand produces optimum packlng at the bottom of the mass, that is, around the pattern. Jolt~ng machines can be used on the first half of a match-plate pattern or on both halves of a cope-and-drag operation. Squeezzng machrnes use an air-operated squeeze head, a Llexible diaphragm, or small, individually act~vated squeeze heads to compact the sand. The squeezing motion provides firm packing adjacent to the squeeze head, wlth denslty dimimshlng as yon move farther into the mold. Flgure 12-13 illustrates the squeezing process, and Figure 12-14 compares the denstty achieved by squeezing with a flat plate and squeezmg with a flexible diaphragm. In match-plate molding, a combination of joltmg and squeeziug is often used to produce a more uniform density throughout the mold.The match-plate pattern is positioned between the cope and drag sections of a flask, and the assembly is placed drag side up on the molding machine. A parting compound IS sprmkled on the pattern, and the drag section of the flask is filled with mixed sand.The entire assembly is then jolted a speciof the pattern. A squeeze fied number of times to pack the sand around the drag s ~ d e head is then swung into place, and pressure is applied to complete the drag portion of the mold. The entire flask is then inverted, and a squeezmg operation 1s performed to compact loose sand m the cope segment. (Note Joltmg here m~ght cause the already-compacted sand to break free of the tnverted drag section of the pattern!) S~nce the drag segment sees both jolting and squeezing, while the cope is only squeezed, the pattern side \nth the greatest detail is generally placed in the drag. If the cope and drag segments of a mold are made on separate machines (using separate cope-and-drag patterns), the cambinatlon of jolting and squeezmg can be performed on each segment of the mold The sprue hole is most often cut by hand, w ~ t h this operation being performed before removal of the pattern to prevent loose sand from falling into the mold cavity.The
Air piston

Air pressure

Flask

Diaphragm

Pattern

292

CHAPTER 12 Expendfable-Mold Casting Processes

FIGURE 12-15 Activ~ty sequence for automatic matchplate molding. Green sand ts blown from the side and The final compressed vert~caliy. mold is ejected from the flask and poured in a flaskless condit~on. (From "F~ve Considerattons for Automatic Matchplate Molding," Engfneered CostfngsSoluhons, Winter, 2001, American Foundry Society, Schaumberg, IL )

pouring basin may also be hand cut, or it may he shaped by aprotruding segment o squeeze board.The gates and runners are usually included on the pattern.

then moved to other molds.

the production sequence for one of the variations of automatic match-plate ma where the sand is introduced into the cope-and-drag mold segments from the side

Since each block contains both a right-hand cavity and a left-hand cavity, an entire is made with each cycle of the machine. (Nofe: Previous techniques required two sep molding operations to produce the individual cope and drag segments of a two mold.) A vertical gating system is usually included on one side of the pattern, an

FIGURE 12-16 Vert~catly

parted flaskless moldrng wrth inset cores. Note how one mold block now contains both the cope-and-drag impressions

Blow

Squeeze

Draw

Set cores

dssembled molds are usually poured individually If a common horizontal runner is used to connect multiple mold segments, the method is known as the H-process. Since metal cools as it travels through long runners, the individual cavities of the H-process often fill with different-temperature metal. To assure product uniformity, most producers reject the H-process, preferring to pour their vertically parted molds individually. In stack molding, sections containing a cope impression on the bottom and a drag impression on the top are piled vertically on top of one another. Metal is poured down a common vertical sprue, which is connected to horizontal gating systems at each of the parting planes. For molds that are too large to be made either by hand ramming or by one of the previously discussed molding processes, large flasks can be placed directly on the foundry floor.Various types of mechanical aids, such as a sand slinger, can then be used to add and pack the sand. Pneumaticrammers can provide additional tamping. Even larger molds can be constructed in sunken pits. Because of the size, complexity, and need for strength, pit molds are often constructed by assembling smaller sections of baked or dried sand. Added binders may he required to provide the strength required for these large molds.

CREEN-SAND, DRY-SAND, AND SKIN-DRIED MOLDS Green-sand casting (where the term green implies that the mold material has not been fired or cured) is the most widely used process for casting both ferrous and nonferrous metals. The mold material is composed of sand blended with clay, water, and additives, and the molds fill by gravity feed.Tooling costs are low, and the entire process is one of the least expensive of the casting methods.Almost any metal can be cast, and there are few limits on the size, shape, weight, and complexity of the products. Over the years, green-sand casting has evolved from a manually intensive operation to a mechanized and automated system capable of producing over 300 molds per hour. As a result, it can he economically applied to both small and large production runs. Design limitations are usually related to the rough surface finish and poor dimensional accuracy-and the resulting need for finish machining. Still other problems can be attributed to the low strength of the mold material and the moisture that is present in the clay-and-water binder.Table 12-2 provides a process summary for green-sand casting, and Figuse 12-17 shows a variety of parts that have been produced in aluminum. Some of the problems associated with the green-sand process can be reduced if we heat the mold to a temperature between 150' and 300C (300" to 575F) and bake it until most oI the moisture is driven oE'Ihis drying strengthens the mold and reduces the volume of gas generated when the hot metal enters the cavity. Dry-sand molds are very durable and may be stored for a relatively long period of time.They are not very popular, however, because of the long time required for drying,the added cost of that operation, and the availability of alternative processes. An attractive compromise may be the production of a skin-dried mold, drying only the sand that is adjacent to the mold cavity.Torches are often used to perform the drying, and the water is usually removed to a depth of about 13 mm (i inch).

r: Almost no limit on size, shape, weighl, or complexity: law cost; almost any metal can be cast. lerances and surface finish are poorer than in other casting processes; somc machining is often required; relatively slow production g line and draft aic needed lo facilitate pattern removal; due to sprues, gates, and risers, typical yields range from 50% to 85%. eraL7: Cast iron,steel,stainlcss streh and carting alloys of aluminum,copper,magnesium, and nickel. 30 g to 3000 kg (1 uz to 6000 l h ) ss iimirr As tllin as 0.25 cm ($in.), with no maximum. oleroncer: 0.8 mm for first 15 cm i n . for first 6 in.).0.003 cm For each additional cm: additional increment for dimensions across the parting line

(A

finish: 2.5-25 microns 1100-1000 "in.! rms.

C H A 12 ~ Expendable-Mold ~ Cast~ng Processes

FIGURE 12-17 A vanetv of ~ rrtn ; sand cast a l ~ r n i r ~ p (Codrtesyo f Bod:ne Alun~,~lunl Inc,, St. LOUIS, MO) The molds used for the casting of large steel parts are almost always skin drie cause the pouring temperatures for steel are s~gnificantly higher than those for cas These molds may also be given a high-sibca wash prior to drying to increase the r tormess of the surface, or the more thermally stable zircon sand may be used as ing. Additional binders, such as molasses, linseed oil, of corn flour, can be added t facing sand to enhance the strength of the skin-dried segment.

SODIUM SILICATE-CO, MOLDING

Molds [and cores) em also be made from sand that receives its strength frbm the t i ~ of n 3% to 6% snd~um silzcute, an inorganic liquld btnder, commonly known as glnss.The sand can be mixed with the liquid bodium silicate m a standard mu1 be packed into flasks by any of the methods discussed previously in this ~emains soft and moldable until it is exposed to a flow of CO, gas. It then har matter of seconds by the reaction: Ma2Si03 + C 0 2 + Na,CO,

+ SIO,

(collaldal)

The CO, gas is nontoxic, nonflammable, and odorless, and no heating is re to initiate or drive the reaction. The sands achieve a tensile strength of about (0.3 MPa) after five seconds of GO, gassing, with strength increasing to 100-2 (0.7-1.4 MPa) after 24 hours of aging The hardened sands, however, have extr poor mllapsibhty, m a h g shakeout and core removal quite difficult, Unhke mos sands, the heating that accurs as a result of the pour actually serves to make th strongfir (a phenomenon similar ro the firlng of a ceramic material). Addit' will burn out during the pow are frequently used to ehhance the collapslhility o

silicate molds. Care must also be taken to prevent the carbon dioxide m the air from ha& ening the premixed sand betore the mold-making process is complete. A mod~fication of this process can be used when certan portlons of a mold require better accuracy, thinner sections, or deeper draws than can be achievcd with ordinary molding sand. Sandmixed with sodium sil~cate IS packed around a spec~al metal pattern to a thickness of about 1 in., lollowed by regular molding sand as a backing matenal. After the sand 1s fully compacted, CO, 1s mtroduced through vents in the metal pattern. The adjacent sand is further hardened, and the pattern can be w~thdrawn with less possibility of damage to the mold,

NO-BAKE, AIR-SET, O R CHEMICALLY BONDED SANDS An alternative to the sodium sihcate-CO, process ~nvolves room-temperaturc chemlcal reactions that can occur between organic or inorganic resin binders and liquid curing agents or catalysts.The two or more components are mixed w ~ t h sand just pnor to the molding operation, and the curing reactions bean immed~ately.The molds (or cores) are then made in a reasonably rap~d fashion,s~nce the mix remains workable for only a short period of tlme. After a few mmutes to a few hours at room temperature (depending on the speclfic b~nder and curlng agent), the sands harden sufficiently to permlt removal from the pattern w~thout concern for d~stortion. After time for addit~onal curmg and the possible application ot a refractory coatlng, the molds are then ready for pour. No-bake molding can be used w ~ t h \~irtuallyall engineenng metals over a wtde range of product sizes and weights. Since the time for mold curing slows production, no-bake molding 1s generally limited to low to medium-product~on quantItles.The cosl of no-bake molding is about 20-30% greater than green-sand mold~ng, so no-bake is generally used where offsetting savings can be ach~evedProducts can also be designed with thinner sections, deeper draws, and smaller draft, and the r~gid molds enable h~gh d~rnens~onal precision, along w ~ t h good surface finish. Since no-bake sand can be compactcd by only light vibrations, patterns can often be made h a m wood, plastic, fiberglass, or even Styrofoam, thereby reducing pattern cost A wid= variety of no-bake and syrtems are avalldble,with select~on bemg based on the metal being poured, the cure time desired, the complexity and thickness of the casting,and possible desire for sand reclamation. Like the molds produced by the sodium s~licate process,no-bake offers good hot strength and high resistance to mold-related casting defects In contrast to the so&um sil~cate material, however, the no-bake molds decompose readily after the metal has been poured, providing excellent shakeout characteristics Permeab~lity must be good,since the heat causes the resins to decompose to hydrogen, water vapor, carbon oxides, and various hydrocarbons-all gases that must be vented. Acr-setrnoldcngand chemically bonded ~ a n dare s other terms that have been used to describe the no-bake process. SHELL MOLDING Another popular sand casting process 1s shell molding, the basic steps of whlch arp descr~bed below and illustrated in F~gure 12-18

1 . The mdiv~dual grains of fine silica sand are first precoated with a thin layer of thermosetting phenolic resln and heat-sensitive liquid catalyst T h ~ s materlal is then dumped, blown,or shot onto ametal pattern (usually some form of cast iron) that has been preheated to a temperature between 230" and 315C (450" and 600F). During a period of sustained contact, heat from the pattern part~ally cures (polymerizes and crosslinks) a layer of material.This forms a strong, solid-bonded reglon adjacent to the pattern. The actual th~ckness of cured material depends on the pattern temperature and the time of contact but typically ranges between 10 and 20 mm (0 4 to 0 8 nl ). 2. The pattern and sand mixture are then inverted, allowing the excess (uncured) sand to drop free. Only the layer of partially cured mater~al remains adhered to the pattern. 3. The pattern with adhering shell is then placed in an oven, where adktiondl heat~ng completes the curing process

CHAPTER 12

Expendable-Mold Casting Processes

resin binder

(b)

Box righted

Ejector

(c)

(dl

FIGURE 12-18 Schematic of the dump-box version of shell molding. (a) A heated pattern is placed over a dump

box containing granules of resin-coated sand. (b) The box is inverted, and the heat forms a partially cured shell around the pattern. (c) The box is righted, the top is removed, and the pattern and partially cured sand is placed in an oven to further cure the shell. (d) The shell is stripped from the pattern. (e) Matched shells are then joined and supported in a flask ready for pouring.
4. The hardened shell, with tensile strength between 350 and 450 psi (2.4-3.1 MPa) then stripped from the pattern. 5. Two or more shells are then clamped or glued together with a produce a mold, which may be poured immediately or stored almost indefinit 6. To provide extra support during the pour, shell molds are often placed in a po jacket and surrounded with metal shot,sand, or gravel.

Because the shell is formed and partially cured around a metal pattern, the pr offers excellent dimensional accuracy.Tolerances of 0.08 to 0.13 mm (0.003 to 0.00 are quite common. Shell-mold sand is typically finer than ordinary foundry sand combination with the plastic resin, produces a very s machining, and other finishing costs can be significantly reduce offers an excellent level of product consistency. Figure 12-19 shows a set of metal patterns, the two shells before clamping, and resulting shell-mold casting. Machines for making shell mold far small operations to large, completely automated devicesfor of a metal pattern is often rather high, and its design must in system, since these cannot be cut after molding. Large amou required, but the amount of matcrial actually used to form a thin shell is not that gr High productivity, low labor costs,smooth surfaces, and a lev the amount of subsequent machining all combine to make even moderate quantities.The thin shell provides for the easy escape of gases that e during the pour, and the volume of evolved gas is rat moisture in the mold material. When the shell becomes h burns out, providing excellent collapsibility and shakeou both the molding sand and completed shells can be stored for indefinite periods of ti Table 12-3 summarizes the features of shell molding.

OTHER SAND-BASED MOLDING METHODS Over the years, a variety of processes have been proposed to overcome some of the itations or difficulties of the more traditional methods.While few have become mercially significant, several are included here to illustrate the nature of thes,e,eff

S ~ c n 12.2 o ~ Sand Casting

297

In the V-process or vacuum moldmg, a vacuum performs the role of the sand b~nder. Figure 12-20 depicts the production sequence, which begins by draping a thin withsheet of heat-softened plastrc over a special vented pattern. A vacuum is appl~ed in the pattern, drawing the sheet tight to its surface.A spenal vacuum flask IS then placed over the pattern; the fka5k is filled with vibrated dry, unbonded sand: a sprue and pourlng cup are formed; and a second sheet of plastic is placed over the mo1d.A vacuum is then drawn on the flask itself,compactmg the sand to prov~de the necessary strength and hardness The pattern vacuum is released. and the pattern is then w~thdrawn.Theother segment of the two-part cope-and-drag mold 1s made in a similar fashlon, and the mold halves are assembled to produce a plastic-lined cavity.The mold 1s then poured with a

Advantuges: Faster production talc than sand molding, high dimsnsianal accuracy with smooth surfaces. Limilalions: Requires expensive metal paltcrns. Plastic resin adds to cost: part size is Limited. Conirnon rnerais; Cast irons and casting alloys of aluminum and copper. si~e limro: , , 30 g (1 or) minimum; usually less than 10 kg (2s lb); mold area usually less than 0.3 ma (500 inz). Thickness llimiiu. Minimums range from 0.15 t a 0.6 cm to in.), depending on material. Typicnlrolerancc.~: Approximately I J . W . 5 c d c m or inlin. DraJiallowance: j or degree. Surfacefinish:~4.0 microns (5&150pin.) rms.

(A

--

Plastic sheet Pattern vacuum

lai

Sprue and ouring cup

Sand Mold vacuum Pattern vacuum

lbl

(ci

FIGURE 12-20 Schematic of the V-process or vacuum molding. (a) A vacuum is pulled o a pattern, drawing a heated shrink-wrap plastic sheet tightly against it. (b) A vacuum flask placed over the pattern and filled with dry unbonded sand, a pouring basin and sprue are'
formed; the remaining sand is leveled; a second heated plastic sheet is placed on top; and. mold vacuum is drawn t o compact the sand and hold the shape. (c) With the mold vacuu being maintained, the pattern vacuum is then broken and the pattern is withdrawn.The cope and drag segments are assembled, and the molten metal is poured.

dask.DltFmg the pmrtthe &ate@ b y ntetal, dlowing the;m u ~ u n ta eo east in^ has coerled %ad
k 3 r n W ~

def-; hider cost is reusable.\hiith no cl

In the Eff-satpracefs,wet sand

yMlnda-&rnm
with liquid nitrogen. The ice that farms serves as the binder, and the molten m~

poured ~nto the mold while the surfaceisin its frozen c o d d i t i ~ nprocess . ~ ~ offel binder mrf and ex&ntsh&wul bvt is not being used in a comme~cial operatiaa

12.3 C O ~ ~ AND E S&RE MAKING

Casting processes are unique in their ability ta easily incorporate somplex ' cavities ur reentrant sections.To produce these fealures, however, it is often n to usecores as part of themold. Rgure 22-21 shows an example of a product tha extensive use o f cores to produce the various cylinders, cooling passages, an internal features. While cores constitute an added wst, they sigdficantly exp ~apbilities of the p m m . and optimize pro Cores can often be used to improve casting des~gn in Figure 12-22, Various 8ide~ the simple belt pulley shown ~chematicaI& fabci~ation are suggested in the foursketch@, beginhingwith the casting of a and the suhssquent machining of the through-hole far the drive shaft. A largev of metal would have to be removed by a secondary machining process A mo namical approach would be to make the ptd6y with a h t - i n hhol In Figwe each half of the pattern include* a tapered hols, whi& filk with th bemgused for ?he remainder o;ftkmol&%ese pro'trudi~g sections

SECTION12.3 Cores and Core Mak~ng

299

i Parting llne

Id)

IE la-21 V-8 engine block (bottom center)

and ti.

FIGURE 12-22 Four methods o f mak~ng a hole in a cast pulley. Three involve the use o f a core.

rotlon, Dehoit, MI.)

of the mold, but they me also known as prea-sand cores. Oreen-sand cores have a a d qtively low strength. If the protrusions are long or narrow, it might be difficult t o withdraw the pattern without breaking them, or they m~ght not have enough strength to even support their awn weight.For long cares,a considerable amount of machmin&may still be required to remove the draff that must be provided on the pattern. In addition, green-sand cores are not an option for more complex shapes, where it might beimpassihle to withdraw the pattein. Dry-sand cores c@n overcome some of the cited difEculties.Thesecores are produced sepamke from the remainder of the mold and are then inserted into e m prints that hold them in posttion. The sketches in Figures 12-22cand 12-Z2d show dry-sand cares in the vertical a d RorizontaI pitions. Dry-sand cores can be made in a nnmbdr of ways. In each, the sand,mixed with some form of binder, is packed into a wood or metal Fore box tbat contains a cavity of'the desired shape. A dumpcore box such a8 the one shown in F~gure 12-23 offers the simplest approach. Sand is packed into

FlGURE 1223 (Upper right) A dumptype cme box; (bottom) two core halves ready for baklng; and (upper left) a completed core made by glu~ng two opposing

halves together.

the cavlty and scraped level with the top surface (which acts like the parting h e traditional mold) A wood or metal plate is then placed over the top of the box, the box 1s inverted and lifted, leaving the molded sand resting on the plate After ing or hardening, the core segments are assembled with hot-melt glue or some bonding agent. Rough spots along the parting l i e are removed with files or sa belts, and the final core may be given a t h n coating to provrde a smoother surface greater resistance to heat Graph~te, silica, or mica can be sprayed or brushed o the surface, Srngle-piece core.7 can be made in a split-core box. Two halves of a core box clamped together, with an opening m one or both ends through which sand is in duced and rammed. Aher the sand is compacted, the halves of the box are separate permit removal of the core. Cores wtfh a uniform cross section can be formed by a co extruding machine and cut to the desired length as the product emerges,The iudivid cores are then placed in core supports for subsequent hardenmg. More complex co can be made in core-blowing machines that use separating dies and receive the san a manner similar to injection molding or die casting. Cores are frequently the most frag~le part of a mold assembly.To provide the essary strength, the varlous core-mak~ng processes utilize a number of special bin In the core-ozlprocess, sand 1s blended with about 1%vegetable or synthetic oil, with244b water and about I % cereal or clay to help develop green strength (i e., t retain the shape prior to cunng).The wet sand is blown or rammed Into a relatively ple core box at room temperature.The fragile uncured cores are then gently transfe to flat plates or special supports and placed in convection ovens at 200" to 260C to 500F) for curlng The heat causes the brnder to cross-link or polymerize,produ a strong organic bond between the grams of sand. While the process is simple and materials areinexpensive, the dimensional accuracy of the resultant coresis often d~% to mamtain. In the hot-box method, sand blended w t h a liquid thermosetting blnder and alyst is packed Into a core box that has been heated to around 230C (450F).When sand is heated, the mmtial stages of curing begin within 10 to 30 seconds. After this b period, the core can he removed from the pattern and will hold its shape during su quent handling. For some materials, the cure completes through an exothemccu reaction. For others, further baking 1s required to complete the process. In the above methods, cores must be handled in an uncuied or partially state, and breakage or distortion is not uncommon. Processes that produce fin~shed while still in the core box and do not require heating operations would appear to distinct advantages. In the cold-boxprocess,bider-coated sand is first blown into a room-tempera core box, which can now be made from wood,metal, or even plastic.The box is sea and a gas or vaporized catalyst is then passed through the permeable sand to poly ize the resin. In a vatlation of the process, hollow cores are produced by introdu small amounts of curing gas through holes in the core-box pattern, with the nnc sand in the center being dumped and reused Unfortunately, the required gases to be e~ther toxic (an amlne gas) or odorous (SO,), making special handlrng of incoming and exhaust gas a process requirement. Room-temperature cores can also be made with the arr-set or no-bake sands. systems ehminate thegassing operation of the cold-box process through the use o active organic resin and a curing catalyst. As discussed previously, there is only a period of hme to form the core once the components have been mlxed. Shell m is another core-making alternative, producing hollow cores with excellent stren permeabihty. Selectingthe actual method of core production is usually based on a number o s~derations, rncluding production quantity, prducbon rate,required precision,req surface finish, and the metal being poured. Certaln metals may be senwtive to gase are emitted from the cares when they come into contact with the hot metal. Othe terials with low pouring temperatures may not break down the hluder sufficien provide collapsib~lity and easy removal from the final casting,

To functlon properly, castmg cores must have the following characteristics:

1 . Sufficient strength before hardening if they will be handled in the "green"condition.

2. Sufficient hardness and strength after hardening to withstand handling and the forces of the casting process.As metal fills the mold, most cores want to "float."The cores must be strong enough to resist the induced stresses, and the supports must be sufficient to hold them in place. Flowing metal can also cause surface erosion. Compressive strength should be betucen 100 and 500 psi (0.7 to 3.5 MPa). 3. A smooth surface 4. Minimum generation of gases when heated by the pour 5. Adequate permeability to permlt the escape of gas. Smce cores are largely surrounded by molten metal, the gases must cscape through the core. 6. Adequate refractormess. B e ~ n g surrounded by hot metal,cores can become qulte a blt hottcr than the adjacent mold material.They should not melt or adhere to the casting.

7. Collapsibility.After pouring, the cores must be weak enough to permit the casting to shrink as it cools, thereby preventing cracking. In addition, the cores must be easily removed from the interior of the finished product via shakeout.
Various techniques have been developed to enhance the natural properties of cores and core materials. Additional strength can be imparted by the addition of internal wires OT rods. Collapsibility can be enhanced by producing hollow cores or by placing a material such as straw in the center. Hollow cores may be used to provide for the escape of trapped or evolved gases.Vent holes can be formed by pushing small wires into the core, and coke or cinders are sometimes placed in the center of large cores to euhance venting. Since the gases must be expelled from the casting, and the core material itself must be removed to produce the desired hole or cavity: the cores must be connected t o the outer surfaces of the mold cavity. Recesses at these connection points, known as core prints, are used to support the cores and hold them in proper position during mold Eilling.The dry-sand cores in Figures 12-22c and 12-22d are supported by core prints. II the cores do not pass completely through the casting, where they can be supported on both ends, a single core print may not be able to provide sufficient support. Addilional measures may also be necessary to support the weight of large cores or keep lighter ones from becoming buoyant as the molten metal fills the cavity. Small metal supports, called chrrplets,can be placed between cores and the surfaces of a mold cavity, as illustrated in Figure 12-24. Because the chaplets are positioned within the mold cavity, they become an integral part of the finished casting. Chaplets should therefore be of the same, or a t least comparable, composition as the material being poured. They should be large enough that they d o not completely melt and permit the core to move, but small enough that their surface melts and fuses with the

k 12-24

(Left) Typical &.(Right) Method of Bfirq a core by use of Q (relative size of the ,pts is exaggerated).

Drag

Cheek

FIGURE 12-25 Method of

Inset section by using a three-piece flask. metal being cast. Since chaplets are one more source of possible defects and become a location of weakness in the finished casting, efforts are generally ma minimize their use. Additional sections of mold material can also be used to produce castings reentrant angles. Figure 12-25 depicts a round pulley with a recessedgroove aroun perimeter. By using a third segment of ing plane, the entire mold can be made by conventional green-sand molding aro withdrawable patterns. While additional molding operations are required, this m an attractive approach for small production runs. If we want to produce a large number of identical pulleys, rapid machine mbl of a simple green-sand mold might would be modified to include a seat for duced at the expense of a core box and a separate core-making operation.

FIGURE 12-26 Molding an

inset sec.tion using a dry-sand core.

~NDABLE-MOLD PROCESSES :PLE-USE PATTERNS

PLASTER MOLD CASTING
In plaster molding the mold material is plaster of paris (also known as calcium or gypsum), combined with various additives to meability, and castability Talc or magnesium oxide can be added to prevent and reduce the setting time. Lime or cement helps to reduce expansion during Glass fibers can be added to improve strength, and sand can be used as a filler. The mold material is first mixed over a metal pattern (wood patterns tion of the plaster produces a hard mold that (Note: Flexible rubber patterns can be used trant angles are required.The plaster is stro removal.)l%eplaster mold is then b With metal patterns and plaste curacy are both excellent. Cooling is slow, si thermal conductivity. The poured metal s replicate fine detail, which can often redn ing is limited to the lower-melting-t copper, magnesium, and zinc). At t plaster would first undergo a phase tran dration can cause the mold to explode. Table 12-4 summarizes the features of mold casting.

S m o 12.4 ~ Other Expendable-Mold Processes with Multiple-Use Patterns

303

,cr,sr Z . . u r n ) ' <,I p..8rlir.\vdltr, acld \;trlous il~lclit~vic 19 plrured 0,er :I pdllr'rti nnJ allsw-d 10 >:I. lllr pauenl ir rr-n~nbc;l an:l 11% nwld i, 1.akr.d 10 r.,nlu\c ;?;cos. ualrr.Ailsr p o ~ ~ l nnnJ , , i.lli.l~f~c~i~~>n.thc ,,\.,Id isba rkcn d n u lh. :.,it,n?. tcml,vid. Advanmagm High dimensland accuracy and smooth surface fimsh,can reproduce thin sect~ons and rntncate dm11to make net- or near-net-shaped parts
I't

C m o n metals Pnma~lly alumrnum and nppw Slzs hrnlt, Asmall as3Og (1 02) but usually less than 7 kg (151b). Thrckrrass Ilrnirs Section thickness 8ssmalLas 0 06cm (Qd25m 1. 73prca1 tolerances 0 01 cm on first 5 cm LO 005 in on First 2 10.). 0 002 cm per addlilonal em (0 002 in per ad&hont11in ) DraJ allowance 4-1 degree Surfneef2nuh 1 3-4 mlcrons (50-125 pin ) rmr

The Anrioch process is a variation of plaster mold casttng wheze the mold materiaI is comprised of 50% plaster and 50% sand, nuxed with water. An autoclave process 1s used to prepare the molds, which offer improved permeability and reduced solidification time The additlon of a foamlng agent ro a plaster-water mix can add fine air bubbles that increase the material volume by 50-100%.The resulring molds have much ~mproved permeability compared to the conventional process.

CERAMIC MOLD CASTING Ceramic maid casting (summarized inTable 12-5) is similar to plaster mold casting, except that the mold is now made from a ceramic material that can withstand the highermeltmg-temperature metals. Much like the plaster process, ceramic molding can produce thin sections,fine detail, and smooth surfaces, thereby elimmating a considerable amount of finish machimng. These advantages, however, must be weighed against the greater cost of the mold material. For large molds, the ceramic can be used to produce a facing around thepattern, which is then backed up by a less expensive material such as reusable fireclav. One of the most p ~ p u l a of r the =ram= noldmg techn~ques is the Shaw process. A reusable pattern IS positioned inside a slightly tapered flask, and a slurry-likemixture of refractory aggregate, hydrolyzed ethylsilicate,alcohol, and a gelling agent is poured on top.This mixtute sefs to a rubbery state that permits removal of both the pattern i t h a torch. Most of the volatiles are and the flask. The mold surface is then Ignited w consumed during the "burn-or and a three-dimensional network of micrmcopic cracks (microcrazmg) forms in the ceramic The gaps are small enough to prevent metaI penetration but large enough to provide venting of air and gas (pehneability) and to accommodate both the thermal expansLon ofthe ceramic particles durmg the pour and the subsequent shrinkage of the solidified metal A baking operation then removes a l lof the remaining volatile$,making the mold hard and rig~d. Ceramic molds are often preheated prior to pouring to ensure proper filling and to control the sohdification characteristics of the metal.

Proo~w Stable ceramic powders are combmedwlth binders andgelllng azents r o produce the mold mntenal Advanfaget Lnir~cate detnal,cloSe tnleI&cs, and smooth finish Limtlnr~onrMold material a costly and not reusable. Camman metals. Ferrous and high temperature nonferrrm.; metah are most common: can ~ S be Q used with alloys of alumloum.copper,magnesium,bmium,andzrnc Szze lmur 105 gtaas to several thousand kilograms (several ounces to Mvernl tons) Thrckms lwnitr A s thin as 0 U em (0.050 I&), no maxlmum Typical r o f m r e i 001 cm on the Rrst 2 5 cm (0.mm.on the first m.), 0.003 m per each add~tlonal cm (0.003 &.per caoh add~tronalin). DrajZ ollawwces I ' preferred Surfecefrrush 2 4 mrcram (75-150pin ) rms
-

FIGURE 12-27 Croup of intricate cutters produced by ceramic mold casting. (Courtesy of Avnet Shaw Division01 Avnet, Inc., Phoenix, AZ)

Like plaster molding, the ceramic molding process can effectively produce s m a size castings in small to medium quantities. Figure 12-27 shows a set of intricate cutt that wcrc produced by this process.

EXPENDABLE GRAPHITE MOLDS

For metals such as titanium, which tend to react with many of the Illore common ma materials, powdered graphite can be combined with additives, such as cement, starch,a water, and compacted around a pattern.After "setting." thc pattern is removed andt mold is fired at 1000C (1800F) to consolidate the graphite.The casting is poured,a the mold is broken to remove the product.

RUBBER-MOLD CASTING
Artificial elastomers can also be compounded in liquid form and poured over a patte to produce a semirigid mold. These molds are sufficiently flexible to permit stri from an intricate shape or patterns with reverse-taper surlaces. Unfortunately, ru molds are generally limited to small castings and low-melting-point materials.The patterns used in investment casting are often made by rubber-mold casting, as are srn quantities o l finished parts made from plastics or metals that can be poured at tempe aturcs below 250C (500F).

:-MOLD PROCESSES USING PATTERNS

INVESTMENT CASTING
Investmrnt casting is actually a very old process-used in ancient China and Egypta more recently performed by dentists and jewelers for a number of years. It was not the end of World War 1 1 , however, that it attained a significant degree of industria portance. Products such as rocket components and jet engine turbine blades required t fabrication o l high-precision complex shapes from high-melting-point metals that are easily machined. Investment casting offers almost unlimited freedom in both the c plexity of shapes and the types of materials that can be cast, and millions of investm castings are now produced each year. lnvestment casting uses the same type of molding aggregate as the ceramicmo ing process and typically involves the following sequential steps:

I. Produce a masterputtern-a

modified replica of the desired product made from me wood, plastic, or some other easily worked material.

2. From the master pattern, produce a master die. This can he made from low-meltin point metal, steel, or possibly even wood. If a low-melting-point metal is used, die may be cast directly from the master pattern. Rubber molds can also be made rectly from thc master pattern. Steel dies are often machined directly. eliminating need for step 1. 3. Produce wax patterns. Patterns are made by pouring molten wax into the master or injecting it under pressure (injection molding), and allowing it to harden. Re1 agents, such as silicone sprays, are used to assist in pattern removal. Plastic and 6.02 mercury are alternate pattern materials. The polystyrene plastic may be preferr for producing thin and complex surfaces, where its higher strength and greater du bility are desired. Frozen mercury is seldom used because of its cost. handlingpr

SECTION 12.5

Expendable-Mold Processes Using Single-Use Patterns

305

lems, and toxicity. If cores are required, they can generally be made from soluble wax or ceramic. The soluble wax cores are dissolved out of the patterns prior to further processing, while the ceramic cores remain and are not removed until after solidification of the metal casting.

4. Assemble the wax patterns onto a conznzon wax sprue. Using heated tools and melted wax, a number of wax patterns can be attached to a central sprue and runner system to create a pattern cluster, or a tree. If the product is sufficiently complex that its pattern could not be withdrawn from a single master die, the pattern may be made in pieces and assembled prior to attachment.

5. Coal the cluster or lree with a thin layer ofinvestment muterial. This step is usually accomplished by dipping into a watery slurry of finely ground refractory materia1.A thin but very smooth layer of investment material is deposited onto the wax pattern, ensuring a smooth surface and good detail in the linal product.

6. Form additional investment around the coated cluster. After the initial layer has dried. the cluster can be redipped. but this time the wet ceramic is coated with a layer of sand or coarse refractory, a process called stuccoing. After drying, the process is repeated until the investment coating has the desired thickness (typically 5 to 15 mm or 4 to 5 . g inch with up to eight layers). As an alternative, the single-dipped cluster can be placed upside down in a flask and liquid investment material poured around it. The flask is then vibrated to remove entrapped air and ensure that the investment material now surrounds all surfaces of the cluster. 7. Allow the investment to fully harden. 8. Remove the wapatternfrom the mold by melting or dissolving.Molds or trees are generally placed upside down in an oven where the wax can melt and run out, and any residue subsequently vaporizes.This step is the most distinctive feature of the process because it enables a complex pattern to be removed from a single-piece mold. Extremely complex shapes can be readily cast. (Note: In the carly years of the process, only small parts were cast, and when the molds were placed in the oven, the molten wax was absorbed into the porous investment. Because the wax "disappeared," the process was called the lost-waxprocess, and the name is still used.) 9. Heat the mold in preparrttionforpouring. Heating to 550" to 1100C (1000" to 2000F) ensures complete removal of the mold wax, cures the mold to give added strength, and allows the molten metal to retain its heat and flow inore readily into all of the thin sections and details. Mold heating also gives bettcr dimensional control because the mold and the metal can shrink together during cooling. 10. Pour the molten metal. While gravity pouring is the simplest, other methods may be used to ensure co~nplete filling of the mold. When complex, thin sections are involved, mold filling may be assisted by positive air pressure, evacuation of the air from the mold, or some form of centrifugal process.

11. Remove the solidified castingfrom the mold. After solidification, techniques such as
mechanical chipping or vibration, high-pressure water jet, or sand blasting are used to break the mold and remove the mold material from the metal casting. Figure 12-28 depicts the investment procedure, where the investment material fills the entire flask, and Figure 12-29 shows the shell-investment method. Table 12-6 summarizes the features of investment casting. Compared to other methods of casting, investment casting is a complex process and tends to be rather expensive. However, its unique advantages can often justify its use, and many of the steps can be easily automated. Extremely complex shapes can he cast as a single piece.Thin sections, down to 0.40 mm (0.015 in.), can be produced. Excellent dimensional precision can be achieved in combination with very smooth as-cast surfaces.Machining can often be completely eliminated or greatly reduced. When machining is required, allowances of as little as 0.4 to 1 mm (0.015 to 0.040 in.) are usually ample.These capabilities are especially attractive when making products from the high-melting-temperature, difficultto-machine metals that cannot be cast with plaster- or metal-mold processes.

Mold material

out of mold.

vacuum, ar centrifugal force.

Wax or plastic is injected into die t o make a pattern.

Patterns are gated to a central sprue.

from sprue, and gate stubs are ground off.

FIGURE 12-28

Investment-casting steps for thefla

Castings are removed from sprue, and gate stubs are ground off.

3 Pattern clusters are dipped in ceramic slurry.

Refractory grain is sifted onto coated patterns. Steps 3 and 4 are repeated several times to obtain desired shell.

After mold material has set and dried, patterns are melted out of mold.

Hot molds are filled with metal by gravity, pressure, vacuum, or centrifugal force.

Mold material is broken away from castings.

FIGURE 12-29

Investment-casting steps for the shell-casting procedure. (Courtesy of investment Cast~ng institute, Dallas, TX.)

3 A metal flask is placsd around the pattern cluster.

4 Flask is filled with investment-mold slurry.

5
After mold material has set and dried, patterns are melted out of mold

6 Hot molds are filled with metal by gravity, pressure, vacuum, or centrifugal 1

Mold material is broken away from castings.

I
1

u
2 Patterns are gated to a central sprue.
stubs are ground off

Wax or plastic is Injected Into dle to make a pattern.

steps for the flask-cast method. (Courtesy of Investment Cast~ng Insbtute, Dallas, TX! FIGURE 12-28 Investment-cast~ng

Wax or plast~c is injected Into die to make a pattern.

1

Panerns are gated to a central sprue.

2

Castings are removed from sprue, and gatestubs are ground off.

3
Pattern clusters are dipped in ceramic slurry.

4 Refractory grain is sifted onto coated patterns. Steps 3 and 4 are repeated several times to obtain desired shell.

5 After mold material has set and dried, patterns are melted out of mold.

6 Hot molds are filled with metal b y gravity, pressure, vacuum, or centrifugal force.

7 Mold material is broken away from castings.

FIGURE 12-29

Investment-casting steps for the shell-casting procedure. (Courtesy of Invest! Institute, Dallas, TX )

ges. EXcellent surhce Aatsh,high dimsmiond aceuracy;almosl unl~mted tntnoacy, almost any me1al can hi? caot,no flash or parting line eon-

ns Costly patterns and molds,laborcosfe can be high, lunited sue

on tnetuls Jusl about any castable metal. Alumbum, coppez, and steel dommate, also performed with stamless steel, mckel, magnesium, and

its As small as 3 g

(A 03) but usually less than 5 kg (I0 16)

css h i f s As tho as 0 . 0 6 cm (0 025 m

), but less than 7 5 cm (3 0 m) taleransm 0 01 cm for the first25 cm (0 OD5 in for the hrst mch) and 0 002 cmfat each additional cm j0.002 m lor eaeh addinmal m )

While mast investment castings are less than 10 cm (4 in.) in size and weigh less than $ kg (1 lb), castings up to 1m (36 m.) and 35 kg (80 Ib) have been produced. Products rangingfrom stainless steel ortitanium golf club heads to superalloy twbine blades have become quite routine. Figure 12-30 shows some typical investment castings. One should note that a high degree of shape complexity is a common characteristic af investment cast products. 7he high cost of dies to make the wax patterns has traditionally limited investment casting to large producuon quantities.R e w t advances in rapid prototyping, however, now enable the production of wax-like patterns directly from CAD data. The absence of part-specific tooling now enables the economical casting of one-of-a-kind or smallquantity products using the investment methds, The majority of investment castings now fall within the range of 100 to 10,000pieces per year.

COUNTER-GRAVITY INVESTMENT CASTING Counter-graviry tnvestment casting turns the pouring process upside down. in one variation of the process, aceramic shellmold is placed in an open-bottom chamber with the sprue end down.The open end of the sprue is lowered into a pool of molten metal, and the bottom of the &amber is set against a seal. A vacuum is then induced wlthin the chamber. As the air is withdrawn, the vacuum draws metal up through the central sprue and into the mold. The castings are allowed to solidify, the vacuum is released, and any unsoaidified metal flows backinto the melt.In another vanation, a Low-pressure lnert gas is used to push the moltzn metal upward into the mold. This approach is discussed in more detatl and is also illustrated in the section on low-pressure permanent-mold casting in Chapter 13.

Typical parts castinq. >stment

308

CHAPTER 12 Expendable-Mold Casting Processes

The counter-gravity processes have a number of distinct advantages. Becausf molten metal is withdrawn from below the surface of its ladle, it is generally free of and dross and has a very low level of inclusions. The vacuum or low-pressure fi allows the metal to flow with little turbulence, Iurther enhancing metal quality. reduction in metallic inclusions improves machinability and enables mecha~ properties to approach or equal those of wrought material. Since the gating system does not need to control turbulence, simpler gating sysl can be used, reducing the amount of metal that does not become product. In the cou gravity process, between 60% and 95% of the withdrawn metal becomes cast pro[ compared to a 15% to 50% level for gravity-poured castings.The pressure differe: enables metal to flow into thinner sections, and lower "pouring" temperatures ca used, resulting in improved grain structure and better surface finish.

EVAPORATIVE PATTERN (FULL-MOLD AND LOST-FOAM) CASTING

Several limitations are common to most of the casting processes that have been sentcd. Somc form of pattern is usually required, and this pattern may be costly tc sign and fabricate. Pattern costs may be hard to justify, especially when the numbt identical castings is rather small or the part is extremely complex. In addition, reuse, patterns must be withdrawn Crom the mold, and this withdrawal often requires SI form of design modification or compromise, division into multiple pieces, or spt molchng procedures. Investment casting overcomes the withdrawal limitations thra the use of patterns that can be removed by melting and vaporization. Unfortunatel) vestment casting has its own set of limitations,including a large number of individual erations and the need to remove the investment material from the finished casting In the evaporative pattern processes, the pattern is made of expanded polysq (EPS),or expanded polymethylmethacrylate (EPMMA).and remains in the mold.1 ing the pour, the heat of the molten metal melts and burns the polystyrene, and metal fills the space that was previously occupied by the pattern. When small quantities are required, patterns can be cut by hand or machinedfi pieces of foamed polystyrene (a material similar to that used in Styrofoam drid cups).This material is extremely light in weight and can be cut by a number of meal including oncs as simple as an electrically heated wire. Preformed material in the& of a pouring basin,sprue,runner segmcnts,and risers can be attached with hot-meltfi to form a complete gating and pattern assembly. Small products can be assembled! clusters or trees, similar to investment casting. When producing largcr quantities 01identical parts, a metal mold or die is ally used to mass-produce the evaporative patterns. Hard beads of polyst preexpanded and stabilized. The preexpanded beads are then injected metal die or mold, usually made from a1uminum.A steam cycle causes t expand, fill the die, and fuse, after which they are coole mold.The resulting pattern, a replica of the pr sists of about 2.5% polymer and 97.5% air. quite complex, and large quantities of patterns ca and rapidly produced. When size or complexity is great, or etry prevents easy removal, the pattern can be divided into ple segments, or slices, which are then assembled by hot-me1 The ideal glue should be strong, fast setting, and produce mum amount of gas when it decomposes or combusts. After a polystyrcne gating system is attached to the polystyrene patte severaloptions for the completion of themold.Tn the,fill-molrlprocess, ically in Figure 12-31,green sand or some type of chemically bonded (nocompacted around the pattern and gating system, taking care not to cr The mold is then poured like a conventional sand-mold casting. In the lost-foam process, depicted schematically in Figure 12-32,the polysty sembly is first dipped into a water-based ceramic that wets both external and in faces and forms a thin refractory coating.The coating must be thin enough and s

FIGURE 12-31 Schematic o f the full.mold process. (Left)An

uncoated expanded polystyrene pattern is surrounded by bonded Sand to Produce a mold. (Right) Hot metal progressively vaporizes the expanded polystyrene pattern and f i l l s the resulting cavity.

Polystyrene pattern

Dipped in refractory slurry

Surrounded with loose unhanded sand

1232 Schematic of

Cornpscted by vlbratton

Metal poured onto polystyrene pattern

Castrng removed and sand recla~med

permeable to permit the escape of the molten and gaseous pattern material, but n@d enough to prevent mold collapse dunng pounng. After the coating dries, the pattern assembly is suspended in a one-piece f l s k and stmounded by fine unbonded smd.Vihtion ensures that the sand compacts around the pattern and fills all cavities and passages. During the pour, molten metal melts, vaporizes, and replaces theexpanded polystyrene, while the coating isolatesthe metal from the loose, unbond~d sand. After the casting has cooled and solidified, the loose sand is then dumped froin the flask, Geetng the casting and attached gating system The backup sand can then be reused, provided the coating residue is removed and the organlc condensat% are periad~cally burned off. Figure 12-33shows the series of operations used in produwng a rather complex lost-roam casting. Thtr full-moldand lost-loam processes ran produce castings of any size in both ferrous and nonferrous metals. S~nce the pattern need not be withdrawn, no draft is required in the design. Complex patterns can be produced to make shapes that would ordinarily require multiple cares, loosc-piece patterns, or extensive finieh machining. Multrcompollent assemblies can often be replaced by a single casting. Bwause of the high p~ecis~on and smooth surface finish, machining and finishing operaaons can often be reduced ur totally elimmated. Fragile or complex-geometry cores are nu longer required, and the absence of parting lines ehminates the need to remove azsociated lines or fins on the metal casting. As the molten metal progresses through the patlern, it loses heat due to the melting and volatilmg of the foam.As a result, the material farthest irum the eate is the wolest,

a103

1ua$jzd-2~~p-puw-adoa

&!~%!uedmo~ kFq!sdello~ xoq-plea ssa~ord

sPJoM

If@

pur 30 L a ~ ne s q ~ a8pal~ouy w s q luama~ddns 1 l ! m laldeq3 dxa) asn-al8ms e ql* s2unsm asnpord leql sassaaod pajuas q!ssod 1le30 8qw).srapunw wl!nbal uoge3ndde uahg e .IOJ jsaq aql jo uoy3alas am pne 'suo!lel!ar!l pm 'sa2e)wnpe 'sagnrqede3 30 las anb!un i d T e a .sadeqs par!sap a n p o d o) UO!~~~!J!P!IOS xuaubasqns pue k~!p!n@p ! n ys axears 01 padolahap uaaq aheq sassasold 30 Jaqmnu nsse Kaqj xeql hadoad aqs~ra~ae~eqa a n aheq sp!nb!~

'8u!u!q3em jo s m r o ~ snoueh 10'2n!urm!~l 'Zu!pu!.~% suo~~erado Zgqqug [euog!ppv~smepnssnopen aqj asodxa 01uoge1nd!um 'speaq sse18 'ap~xomnu!mnle aug '(laals lo uo~!X[[ansn) ue3 mpam p a ~ l a d o ~ aU'Zu!lse3 d a q l j o sa3ejsns aqj lsuye8e sap!l~ed .slJnq au!~-8,u!~red pue d ox pasn s ! a x o j 1e8njulua3 10 118 passa~duro3 . 1 8 3 8u!ueap lsqq sa q3ns sassasold es Bu!laqpe ahouralol pasn aq 1 . s s a u y 3 alqeuosear ~ jo s8ugse3 sselq pue 'laals'uo~!1se301 palqqsar L[il@nsn 01 a 2 ~ m e p a[q!ssod 30 asne3aa qmnj h l e l o ~ 'sKu!~se3 pauo!mas-mql~o ly%!a~lqZ!1 urn1 saqstua 's8u!)se:, aql moij slauutu pue sale8 ayl ~p Zu!lelo~'1alamwp-a8~el'Zu!umj-MO[Se ap!su! sjuaj 3 moq pues aq) anomal s~ojemdas L1elo8 'sar~uanb 8ue1 e ur alqel!ehe aJe'slualuo3 p a p q x a aqa ro syseu uo alelado ue3 q3!qm 'sau!q3em k o x e ~ q ,~au!eluo3 !~ aql mo.13 (8u!xse3 UES 8u!p[om qjoq) y s e e ~ 30 sjua~uos a q u a am a3103 oi pasn aq ue3 sau!q:,eur und .sZu!~sesaql m o g saro3 aqj aaomal l o aleledas pue 'pues Zmp~om aql urolj e3 aqx aleledas '(s1an!eluo3 '.a,!) sqsea aql ruoq pues pue sp~oru aqj .a)sledas 01 s3qsea aql 8u!le~ sap ale suope~ado inoaynqs.1e!.taaeur p1om pue spIom ayl ~ O I J sanronu! dajs [eug aql'1ajdeq3 s!qj u! paluasald sassa3old 8 u g s a aql jo q3ea u~

~ N I H S I Nm I ~ v '~NINVT
.sa!r!~qedea sl! ~remurns L-ZLalqeA'asn pue Aluelndod u! klp!der u ~ o r seq 8 Kqsea 'suoseal laqlo pne asaql .1od.q8!q laqjel aq 01 spual (rejaw pa~nod KI) pla j [elam .pa~!nba~ lou axe =asp 's8u!lse3 Kuear q3eq lauueur Ieuo!pa.np e u y paa3o~d 01 spual uo!lesg!p!yos pue >ad3
Ie18 ale8
8,

I k
!

( u ! d ()()()1-()()1) suni3!w SZ-S.Z : t p u ? J a ~ o & ~

-[In1
'IlOJ

lseu dxa

dxa
ena

-333

LP unp

.pal!nba~ a u n :a3uontulp ~ gsq .ssa1 l o (.u![u! COO'") u 3 p 3 g 0 ~ ( ) : s a ~ u ~ r 7 ln?!dL~ /01, .l!ru!l ~addn ou q l ! (u! ~ 1.0) u w ; 'z su 11mnss v : q ' ~ g ssau.~~y~ I " '(S!JOllVJaAaS01 ql 1) 87 pUEsnOI&l [QIan9S 01 4 y 5 0 : S l l U + ! l allS -1aalsssalu!ss pue raddo3 qi!m pawmjlad osla :sLop [ayqu pue'Iaals'uor! 'wnu!wnly S p I J i u unwtuo3 . w h l l s MOI jo asnmaqpallolsrp l o pa8emvp hl!ssa are sula~led :sa!~!iuenb[laws ~ o q%!q j aq u a ,503 uIalled :suu!im!rcg '(s*u!I %!lnd ou) luasa~d s! qsea ou pun paqnbai s! ireip ou:lwa aq ueo slalaur ISOW rz!s pue adeqs uo si!w!l ou lsowlv :.s23nlunnpv puns aql qZno~rli pa~uan s! pue wglodea parnod s! lelaw u a 1 l o W . p ~ as001 ~ ~ Lq papunouns ST]!alaqw'yse~ e u! pauo!~!x,dpue'pa!~p'lo!l=lBm 3!woraa e q3n[~'u1alledaql oluo dnja~rp ~1 padd!p s! 11 'aoarh~sd~od rn qgns '3!1seld pame03 30 sa~a!d aid!ilnru l o a[Zu!s morj aperu s! as:^ puu's~aurmr 'studs e iiu!uje~uo~ u~a~v ed :s~amrd

lvepattern castmg

pressive strength

grccn-sand cores H process hand rammmg hardness hot-box method hot tears mvestment ca~ting joltlng loose-p~ece pattern lost-foam castmg lost-wax plocess match plate mold hardness muller no-bake sand

one-piece pattern pattern penetratlun pormeab~lity plaster mold

rcfrdctu~lmcss

rubber-mold castlng runner sand expansion defects sand slmnger shakeout Shaw process shell moldmg s~bca sand single-p~ece cores

skin-dried mold sodium silicate-COI molding split pattern sprue squeezing slack molding standard rammed specimen V-process vacuum molding vertically parted flaskless molding water glass

i l l l Review Questions
B a t are some ol the factors that Influence the select~on of

fic casting process as a means of making a product? t metals are frequently cast into products? h type of casting is the most common and most versatile? at is a casting pattern?

is the simplest and least expensive type of casting is a match plate; and how docs it aid molding? is a cope-and-drag pattern different from a match-plate m l Whcn might this he attractive?

t ways might a molding sand he a compromise material?

s a muller, and what function does it perform?

tare some of the properties or characteristics of foundry s that can be evaluated by standard tests? t is a standard rammed specimen for evaluating foundry

ngth of green sand? w does the size andshape of thesand grains relate tomoldcan sand expansion defects be minimized? at features can cause the penelration of molten metal b e en the grains o C the molding sand? at are hot tcars, and what can cause them to form? cribe the distribution of sand density after compaction by e number of mold sections required to produce a series of at is slack molding?

27. How might extremely large molds be made? 28. What is green sand? 29. What are some of the limitations or problems associated with green sand as a mold material? 30. What restricts the use of dry sand molding? 31. What are some of the advantages and limitations of the sodium silicate-C02process? 32. What is the primary feature of no-bake sands? 33. What material serves as the binder in the shell-molding process, and how is it cured? 34. Why do shell molds have excellent permeability and collapsibility? 35. What is the sand binder in theV-process?The Eff-set process? 36. What types of geometric features might require the nseof cores? 37. What is the primarv limitation of green-sand cores? 38. What is the sand binder in the core-oil process, and how is it cured? 39. What is the binder in the hot-box core-making process? 40. What is the primary attraction of the cold-box core-making process? The primary negative feature? 41. What is an attractive feature of shell-molded cores? 42. Why is it common for greater permeability, collapsibility, and refractoriness to be required of cores than for the basemolding sand? 43. What is the role of chaplets, and why is it important that they not completely melt during the pouring and solidification of a casting? 44. Why are plaster molds suitable only for the lower-meltingtemperature nonferrous metals and alloys? 45. What is the primary performance difference between plaster and ceramic molds? 46. For what inaterials might a graphite mold be required? 47. What materials are used to produce the expendable patterns for investment casting? 48. Describe the progressive construction of an investmentcasting mold. 49. Why are investment-casting molds generally preheated prior to pouring? 50. Why are investment castings sometimes called "lost-wax" castings? 51. What are some of the attractive ieatures of investment casting? 52. What are some of the advantages of counter-gravity investment casting over the conventional gravity pour approach?

312

CHAPTLR 12 Expendable-Mold Castlng Processes

55. What are some oi the attractlve features ofthe evapt patter11pracesscs7 56 What are some of the objectives of a shakeout operat 57 How mght castidgs be cleaned after shakeout?

53 What are somc of the Lxnehts of not hav~ng to remove the pattern from thelnold (as m Investment ~z~tmng,full-mold caqt ing, and lost-foam castlng)? 54. S ~ n c both e use expanded polystyrene a9 a pattern. what is the primary dfference between full-mold and loat foam castmg?

ill Problems
1. While cores increase the cost of castings, they also provide a numhw of distinct advantages, The mosL significant is the ability to produce complex internal passages. They can also enable the production of difiicult external features, such as unde~cuts, or allow the production of zero-draft walls. Corcs can reduce or eliminate additional machining. reduce the weight of a casting,and reduce o r eliminate the nced for multipiece assembly. Answer the following questions about cores. a. The cores themselves must be produced, and generally they have to he removed from core boxes or molds. What geometric limitations might [his impose? How might these limitations he overcome? b. Cores must b e positioned and supported within a mold.

Discuss some of the limitations associated with co tioning and orientation. Consider the weight of prevention of core fracture, minimization of core dei and possible buoyancy. c. Since cores are internal to the casting, adequate vt necessary to eliminate or minimize porosity proble~ cuss possible features to aid in venting. d. How might core bchavior vary with different rn being cast-stccl versus aluminum, fur example? e. Core removal is another design concern. Discuss h era1 different core-making processes might perf011 area of removal. What are some ways to assist orf; core removal?

Movable and Fixed law Pieces for a Heavy-Duty Bench Vise

he figure presents a cutaway sketch of the movable and fixed jaw pieces of a heavy-duty vise that might s e e use in vocational schools, factories, and machine s intended to have a rated maximum shops.The vise i clamping force of 15 tons.The slide of the moving jaw has been designed to be a 2-in. box channel.The jaw width is 5 in., the maximum jaw opening i s 6 in., and the depth of the throat i s 4 in.The designer has elected to use replaceable, serrated jaws and suggests that the material used for the receiving jaw pieces h a v e a yield strength in excess of 35 ksi, with at least 15% elongation in a uniaxial tensile test (to ensure that an overload or hammer impact would not produce brittle fracture). Replaceable jaws

1. Determ~ne some poss~ble comblnatlons o f materlal a process that could fabricate the deslred shapes wlth fhe requ~red propertles. Of the alternatives presented, wh~ch would you prefer and why? 2 Would the components require some form of subse. quent heat treatment? Consider the posslbil~t~es of stres rellef, homogenlzat~on, or the establishment of desired flnal properties. What would you recommend? 3 One of your colleagues has suggested that the slides beflnlshed w~th a coat of palnt Do you thlnk a surface treatment a necessary or deslrable for your selected m terlal and process7 If so, what would you recommend? not, defend your recommendat~on.

Note: shaded surfaces have been produced by cross-sectional cuts

LTIPLE-USE-MOLD CASTING PROCESSES

FNT-MOLD CASTING ressure and Vacuum anent-Mold Cast~ng
13.7 MEL~TNG cupolas Ind~rect Fuel-Fired Furnacea (or Crucible Furnaces) Direct Fuel-Elred Furnaces or Reverberatory Furnaces Arc Furnaces Induction Furnaces 13.8 PQUR~NG PRAI-IICI. 13.9 CLEANINO. FINISHING, AND H E ATREATING ~ OF CASTINGS Cleaning and Rnishing Heat Treatment and Inspection of Castmgs 13.10 A U I O M A ~ IN O FOUNDRY N OPERA~ONS 13.11 PROCESS SELECIIION Case Study BASEPLATE FOR A HOUSEHOLD STEAM IRON

TRIFUGAL CASTWG

uous CASTING

.1 INTRODUCTION
In each of the expendable-mold casting processes discussed in Chapter 12, a separate mold had to be created for each pour. Variations 1n mold consistency, mold strength, rnolsture content, pattern removal, and other faclorh contr~bute to dimens~onal and property variation krom casting t o casting. In addit~on, the need to create and then destroy a separate mold for each pour results in rather low production rates The multiple-use-mold casting processes overcome many of thebe limitations, but they,m turn, havc then own assets and lidhllities. Slnce tho molds are generally made from metal, many of the processes are restricted to casting the lowcr-meltmg-point nonferrous metals and alloys. Part s ~ z 1s e often llmited, and ihe dies or molds can be rather costly.

.2 PERMANENT-MOLD CASTING
In the permanent mold cnsnng process, also called gravity dle casting, a reusable mold 1s machined from gray cast iron, alloy cast iron, steel, bronze, graphite, or other matcrlal The molds are usually made in segments, wh~ch arc olten h~nged to permit rapid and accurate opening and closing.After prcheatmg, a refractory or mold coatmg is applied to the preheated mold,and the mold is clamped shut. Molten metal is then poured Into the pouring basin, and it flows through the feeding system into the mold cavity by simple gravity flow. After solid~ficarion, the mold is opened and the product iq removed. Since the heat from the previous cast 1s usually sufficlent to maintam mold temperature, the process can be immedialely repeated, with a smgle refractory coating servlng far several pouring cycles. Alummum-, magnesium-,zinc-,lead .and copper-based alloys are the metals most frequently cast,along with gray cast iron. If graphite is used as the mold material. Iron and steel castings can also be produced. Numerous advantages can be cited for the permanent-mold process. Near-net shapes can be produced that require llttle flnnh machin~ng The mold IS reusable, and a good surCace fin~sh is obtained if the mold 1 s in good condition. Dimens~ons are consis0.25 mm tent from part to part, and dimensional accuracy can oiten be held to wlth~n (0 010 m.). D~rectional solid~fication can be achieved through good deslgn or can be promoted by select~vely heating or chillmg various portlons of the mold or by varying the thickness of the mold wall. The result 1s usually a sound, defect-free castlng with good mechanical piopert~es.The fastcr cool~ng rates of the metal mold produce a finer grain structure, reduced porosity, and h~gher-strength products than would result from

313

a sand casting process Cores, both expendable sand or plaster or retractable meta be used to increase the complexity of the casting, and mult~ple cavities can ofte cluded In a single mold. When sand cores are used. the process is often called s manent mold castmg. On the negative side, the process is generally limited to the lower-melting-p loys, and high mold costs can make low production runs prohibitively expensi useful life of a mold is generally set by molten metal eroslon or thermal fatigue. making products of steel or cast iron, mold life can be extremely short. For the 1 temperature metals, one can usually expect somewhere between 10,000 and 120,O clesThe actual mold life will depend upon the followmg.

1. Alloy bezng cast. The hlgher the melting point, the shorter the mold life. 2. Mold mukrul. Gray cast iron has about the best resistance to thermal fatigue an
chines eaaly.Thus it is used most frequently for permanent molds. 3. Pouring temperature. H~gher pouring temperatures reduce mold hfe, increase s age problems, and ~uduce longer cycle times. 4. Mold temperature.If the temperature is too low, one can expect misruns and larg perature differences in the mold If the temperature is too hlgh, excessive cycle result and mold erosion 1s aggravated. 5. Mold configurut~on. Differenceb in section sizes of either the mold or the cas produce temperature differences withrn the mold and reduce its life. The permanent molds contan the mold cavity,pouring basm,sprue,runner$ gate$ possible core supports, alignment pmu, and some form of ejection syste molds are usually heated at the beginning of a run. and continuous operatio maintains the mold at a fairly uniform elevated temperature T h ~ m~nimizes s the of thermal fatigue,facilitates metal flow, and controls the cooling rate of the meta cast. Slnce the mold temperature rises when a casting is produced, it may be ne to provide a mold-cooling delay betore the cycle is repeated. Refractory was graphite coatings can be apphed to the mold walls to control or direct the coolin vent the casting from stxcking, and prolong the mold hfe by minimizing thermal and fatigue. When pouring cast iron, an acetylene torch is often used to apply a c of carbon black to the mold. Since the molds are not permeable, special prov~sion must be made for vent1 is usually accomplished through the shght cracks between mold halves as by ve vent holes that permlt the escape of trdpped air but not the passage of molten Smce gravlty is the only means of lndue~ng metal flow,risers must still be empl compensate for solidrficat~on shrinkage, and with the necessary sprues and yields are generally less than 60%. Mold complexity is often restricted because the rigid cavity offers no col ty to compensate for the sohd-state shrmkage of the casting. As a best alterna common practice to open the mold and remove the casting ~mmediately after s cation.This prevents the formation of hot tears that may form if the product is res dunng the shr~nkage that occurs during cooldown to room temperature. For permanent-mold casting, h~gh-volume production is usually required t fy the high cost of the metal molds Automated machlnes can be used to coat t pour the metal, and remove the casting. Figure 13-1 shows a variety of automo truck pistons that were manufactured by the permanent-mold process, which marized in Table 13-1.

SLUSH CASTING
Hollow castings can be produced by a variant of permanent-mold casting known castmng. Hot metal IS poured into the metal mold and is allowed to cool until a desired th~ckness has formed. The mold is then inverted and the remaining poured out. The resulting casting is a hollow shape w~th good surface detall able wall thickness. Common applications Include the cabtlng of ornamental obje as candlesticks, lamp bases, and statuary from the low-melting-temperature met

RE 13-1 Truck and car

IS are

mass-produced by iii~ons uslng permanentcastlng. (Courtesy of a1Motors Corporatron, t, MI )

Advanrages: Good Surface finish and dimensional accuracy; metnl mold gives rapid cooling and fine-grain
structure:multiplr-use molds (up IQ 120,000 uses): metal cores or collapsible sand cores c i l n be used. Limitations: High initial mold casl, shape, size, and complexity are limited; yicld rate rarely exceeds 6096, but rueners and risers can be directly recycled: mold lifcis vcry limitzd with high-melting-point metals such as steel.

Size limits: 100 grams to 75 kilograms (sevcml ounces to 150 pounds). Thickness limils: Minimum depends an malerial but generally greater than 3 mm (i in.); n~aximurn thickness about 50 mm (2.0 in.). Geometric irmits The need to extract Lhc part from a irgid mald may llmil certaln gcometrlc features Unrform semlon thickness is desuable. Typicaltolerunces: 0.4, mm f<rrthe first 2.5. cm (0.015 in. lor the first inch) and 0.02 rnm for esch additional centimeter (0.0U2 in.fot each additional incll);O25mm (0.01 in.) added if thc dime~lsion crosses a parting line. Drafi allowance: 2"-3*. Surface finish: 2.5 to 7.5 ym (100-250 pin.) rms.

LOW-PRESSURE AND VACUUM PERMANENT-MOLD CASTING

Gravity pouring is the oldest, simplest, and most tradlt~onal form of permanent-mold casting. In a vanation known as ttlt-pour permanent-mold tamng, the molten metal is placed in the pouring basin and the mold then rotates to Induce flow into the mold cavity. In this way, turbulence 1s minim~zed as the metal flowa through the gating system and into the mold. In low-pressure and vacuum permanent-mold castmg. the mold is turned ups~de down and positioned above a sealed, autlght chamber that contains a crucible of molten meta1.A small pressure d~fference then causes the molten metal to flow upward into the die cavity In the low-pressurepermanent-mold (LPPM) process, lllustraled in Figure 13-2, a low-pressure gas (3 to 15 pst) IS Introduced lnlv a sealed chamber, dr~ving molten metal up through a refractory fill tube and into the gatmg syatem or cavlty of a metal mold.This metal is exceptionally clean, since it flows from the center of the melt and IS fed directly Into the mold (a distance of about 10 cm, or 3 to 4 in.), never passing through the atmosphere. Product quality is further enhanced by the nonturbulent mold filling, wh~ch helps to mimmize gas porosity and dross formation. Through design and coolin& the products directionally solidify from the top down. The molten metal in the pressurized fill tube acts as a rrser to cont~nually feed the casting

7~

13

Multiple-Use-Mold Casting Processes

Pneumatic clamps

FIGURE 13-2 Schematic of

during solidification.When solidification is complete, the pressure is released and th used metal in the feed tube simply drops back into the crucible.The reuse of thi coupled with the absence of additional risers,leads to yields that are often greater Nearly all low-pressure permanent-mold castings are made from alu magnesium, hut some copper-based alloys can also he used. Mechanical properti typically about 5 % better than those ol'conventional permanent-mold castings times are somewhat longer, however, than those of conventional permanent mo Figure 13-3 depicts n similar variation of permanent-mold casting,where av is drawn on the die assembly and atmospheric pressure in the chamber forces upward. All of the benefits and features of the low-pressure process are set cluding the subsurface extraction of molten metal from the melt, the bottom feed mold, the minimal metal disturbance during pouring, the self-risering action, downward directional solidification. Thin-walled castings can be produced w i metal yield and excellent surface quality. Because of the vacuum, the cleanlines metal and thc dissolved gas contcnt are superior to that of thelow-pressure proc castings typically range from 0.2 to 5 kg (0.4 to 10 lb) and have mechanical prope arc even better than those of the low-pressure permanent-mold products

I n the die-casting process, or more specifically pressure die casting,molten m forced into metal molds under pressures of several thousand pounds per squar (tens of MPa) and held under high pressure during solidification. Because of th bination of metal molds or dies and high pressure, fiue sections and excellent det be achicved, together with long mold life. Most die castings are made from nonf metals and alloys. with special zinc-, copper-, magnesium-, and aluminum-based having been designed to produce excellent properties when die cast. Ferrous-me castings are possible but are generally considered to be uncommon. Productionra high, the products exhibit good strength, shapes can be quite intricate, and dime precision and surface qualities are excellent.There is almost a complete elimin subsequent machining. Most die castings can be classified as small- to mediu parts, but the size and weight of die castings are continually increasing. Parts c be made with weights up to 10 kg (20 lb) and dimensions as large as 600 mm (2

SECTION 13.3 Die Casting

31 7

Vacuum Retractabl upper die

Casting

Feed tube
~ o w edie i

Schematic

Molten pool

Die temperatures arc usually maintained at about 150" to 250C (300" to 500F) below the solidus temperature of the metal being cast in order lo promote rapid freezing. Since cast iron cannot withstand the high casting pressures, die-casting dies are usually made from hardened hot-work tool steels and are typically quite expensive. As shown in Figure 13-4,the dies may be relatively simple, containing only one or two mold cavities, or they may be complex, containing multiple cavities of the same or different products, or even be an assembly of multiple subcomponents.The rigid dies must scparate into at least two pieces to permit removal o l the casting. It is not uncommon, however, for complex die castings to require multiple-segment dies that open and close in several different directions. Die complexity is further increased as the various sections incorporate water-cooling passages, retructubfe cores, and moving pins to knock out or eject the finished casting. Die life is usually limited by wear (or erosion), which is strongly dependent on the temperature of the molten metal. Surlace cracking can also occur in response to thc large number of heating and cooling cycles that are experienced by the die surfaces. If the rate of temperature change is the dominant feature,the problem is called heat checking. If the number of cycles is the primary cause, the problem is called therrniil fatigue. In the basic die-casting process, water-cooled dies are first lubricated and clamped tightly together. Molten metal is then injected under high pressure. Since high injection pressures cause turbulence and air entrapment, the specified values of pressure

Multiple-cavity die

Combination die

FIGURE 1 3 4

Various

types of die-casting dies. (Courtesy of American Die Casting Institute, Inc,, Des Plaines, IL.)

318

CHAITER 13

Multiple-Use-MoldCasting Processes

K
FIGURE 13-5 Principal components of a hot-chamber die-casting machine. (Adapted from Metals Handbook, 9th ed., V o l 15, p. 287, ASM Internotional, Metals Park, OH.)
Molte metal

Gasloil accumulator

and the time and duration of appllcat~on vary cons~derably. The pressure nee be constant, and there has been a trend toward the use of larger gates and rnjection pressures, followed by the application of higher pressure after the mol completely filled and the metal has started to solidify. By reducung turbulence solidifying under high pressure, this cycle reduces both the porosity and incl wntent of the finished castung.After sohdification is complete, the pressure is rele the dies separate, and ejector pins extract the finished castmg along with its atta runners and sprues. There are two basic types of die-casting machines. Figure 13-5 schema ~llustrates the hot-chamber, or gooseneck, variety.A gooseneck chamber is partiall merged in a reservoir of molten metal. With the plunger raused, molten flows through an open port and fills the chamber.A mechanical plunger then forc metal up through the gooseneck, through the runners and gates, and mto the die, it rapidly sobdifies. Retraction of the plunger then allows the gooseneck to refill casting is being ejected, and the cycle repeats at speeds up to 100 shots per minut Hot-chamber die-casting machines offer fast cycling times (set by the abil the water-cooled dies to cool and solldify the metal) and the added advantage th molten metal is injected from the same chamber in which it is melted (i.e., there handling or transfer of molten metal). Unfortunately, the hot-chamber des~gn c be used for the higher-melting-point metalg and it is unattractive for aluminumsi molten aluminum tends to pick up some iron dunng the extended time of contac the casting equipment. Hot-chamber machmes, therefore, see prlmary use with tin-, and lead-based alloys Zinc die castings can also be made by a process known as heated-munzfold d znject~on die casting (also known as direct-injection die casting or runnerless die cas The molten zinc is forced through a heated manifold and then through heated nozzles directly into the d ~ cavity.This e approach totally eliminates the need for gates, and runners. Scrap is reduced, energy is conserved (less molten metal p and no need to provide excess heat to compensate for cooling m the gatmg system product quality is increased. Exlsting die-casting mach~nes can be converted t the addition of a heated manifold and modification of the various dies. Cold-chamber machines are usually employed for the die casting of materi are not suitable for the hot-chamber design.These include alloys of aluminum,m sium, and copper as well as high-aluminum zinc.As illustrated m Figure 13-6,met has been melted in a separate furnace is transported to the die-castmg machme, a measured quantity is fed into an unheated shot chamber (or injection cylinder) an sequently driven into the die by a hydraulic or mechanical plunger.The preswre maintarned or increased unul sohdification is complete. Since molten metal transferred to the chamber for each shot, the cold-chamber process has a longer ating cycle compared to hot-chamber machines. Nevertheless, productivity is still

SECTION 13.3 D i e Casting

asloil accumulator

URE 13-6

Principal ponents of a cold-chamber asting machine. (Adapted Metals Handbook, 9th ed., 5, p. 287, ASM International,

In all variations of the process, die-casting dies fill with metal so fast that there is little time for the air in the runner system and mold cavity to escape, and the metal molds offer no permeability. The air can become trapped and cause a variety of defects, including blowholes, porosity, and misrnns. To minimize these defects, it is crucial that the dies be properly vented, usually by wide, thin (0.13-mm or 0.005-in.) vents positioned along the parting line. Proper positioning is a must, since all of the air must escape before the molten metal contacts the vents.The long thin slots allow the escape of gas but promote rapid freezing of the metal and a plugging of the hole.The metal that solidifies in the vents must be trimmed off after the casting has been ejected. This can be done with special trimming dies that also serve to remove the sprues and runners. Risers are not used in the die-casting process since the high injection pressures ensure the continuous feed of molten metal from the gating system into the casting.The porosity that is often found in die castings is not shrinkage porosity; it is more likely to be the result of either entrapped air or the turbulent mode of die filling. This porosity tends to be confined to the interior of castings, and its formation can often be minimized by smooth metal flow, good venting, and proper application of pressure. The rapidly solidified surface is usually harder and stronger than the slower-cooled interior and is usually sound and suitable for plating or decorative applications. Sand cores cannot be used in die casting because the high pressures and flow rates cause the cores to either disintegrate or have excessive metal penetration. As a result, metal cores are required, and provisions must be made for their retraction,usually before the die is opened for removal of the casting. As with all mating segments and moving components, a close fit must be maintained to prevent the pressurized metal from flowing into the gap. Loose core pieces (also metal) can also be positioned into the die at the beginning of each cycle and then removed from the casting after its ejection. This procedure permits more complex shapes to be cast,such as holes with internal threads. but production rate is slowed and costs increase. Cast-in inserts can also be incorporated in the die-casting process. Examples include prethreaded bosses, electrical heating elements, threaded studs, and high-strength bearing surfaces. These high-temperature components are positioned in the die before the lower-melting-temperature metal is injected. Suitable recesses must be provided in the die for positioning and support, and the casting cycle tends to be slowed by the additional operations. Table 13-2 summarizes the key features of the die-casting process. Attractive aspects include smooth surfaces and excellent dimensional accuracy. For aluminum-, magnesium-, zinc-, and copper-based alloys, linear tolerances of 3 mmim (0.003 in./in.) are not uncommon.Thinner sections can be cast than with either sand or permanent-mold casting.The minimum section thickness and draft vary with the type of metal, with typical values as follows:
Metal
Aluminum alloys Brass and bronze Magnesium alloys Ziuc alloys

Minimum Section
0.89 mm (0.035 in.) 1.27 m m (0.050 in.) 1.27rnm (0.050 in.) 0.63 mm (0.025 in.)

Minimum Draft
1:lOO (0.010 in./in.) 1:80 (0.015 in./in.) 1:100 (0.010 in./in.) 1 :2W (0.005 in.iin.)

It~ple-Use-Mold Casting Processes

and take the fonn of ntovlng metal segments

Advantages Extremely smwth surfaces and amellent dimensional accuracy, rapui prnductron ~ete,p

tcnsticstrcngths as I ~ g h as 415 Mpa (60 ks~). Ltrnirairana High ~nitlal dle cosq hm~ted to high-flulaty nonferrous metals,. part size a itmrted, porosl may be a problem; some scrap m spruw. runhers, and flash, but this can be dlrectly recycled Common metnls Alloys of alum~num, ant, rnagnEnum, and lead; also poss>bleWlth alloys of copper an S t e limirs L e s than 30 grams (1 oz] up through about 7 kg (15 lh) most common. Thickness l~rr~ils As thin as 0 75 mm (0 03 i n ).but generally less than 13 mm i n) V p m l tolerancer Varles with metal be~ng -sf, typl~ally D lmm for the iim 2.5 cm (a 004 m for the mch) and 0.02 mm f ~each r additional cerztimerrr (0 062 m for each Addtianal inch) Drafr n1lavnne.w 103" S~lrfncefinzsh 1-2 5pm(40-100pin ) rms

(i

dnlling or tapping of holes Production rates are high, and a set of dies can pro many thousands o f castings wlthout sigmficant change ~ndimensions While die ca is most economical for large production volumes, quantities as low as 2WO can b tified if exrensive secondary machining or surface finishing can be eliminated.

tion, weathering, snd stress cracking when exposed to various reagents.

the key features af the four d o w a n t families of die-casting alloys, andTable 1 pares the mechanical properties of various die-cast alloys with the properhes

engtneering materials. ASTING AND

SEMISOLID CASTING

involve allQysof akurmnum, each d the processes bas been successfully applied to nesium, zinc, copper, and a limited number of ferrous alloys.

Magnesium

machinab~hiy electrical condunttlfy, and heat-transfer characranshcs Lowest denstty, faster pfoduat~on than alnmlnumsince hot-chamber cast, hghest strength to-werght ratio,good viirat~on dnmpmg, best machmabihty,can provide eleclromagnetb sh%elding

o f aluminum (left) and zlnc (r~ght) d ~ cast~ngs. e (Courtesy of Yoder Die Casting FIGURE 13-7 Var~ety Corporation, Dayton, OH)

. . .. . ~?~,of.~nt~ertlsr (De-Cast Metals vs. Other ;&iefdj)" : . .' .., . , .. . .. . Yield Strength Tensile Strength Elastic Modulus
Materral
Die cast alloys
?hn a m ~ n~ u m ~ .l n ,~ ~ ~~. .
~ ~

. ..

MPa
-

ksi
~

MPa
300

ks~
44

GPa

lo6 psi

170 ~

25 ~

380 aluminum AZ91D magnesium Zamak 3 zinc (ACi4DA) Zamak 5 zinc (AC4IA)

Other metals Steel sheet

HSLA cteel sheet Powdered lron

172-241 414 483

-

25-35 60
70

276 414 55 62 152 145

40 60 -

Plastics ABS

Polycarbonate
Nylon 6' PET"

8 9 22 21

7 1 10 14

10
10

15 20

L n the squeere ccastmg process, molten metal ts introduced into the die cavlty of a metal mold, using large gate areas and slow metal velocities to avoid turbulence. When the cavlty has filled, h ~ g pressure h (20 to 175 MPa, or 3000 to 25,000 psi) is then applied and maintained during the hubsequent solidificat~on. Parts luust be designed to directionally solidify Coward the gates, and the gates must he sufficiently large that they freeze after solidificatron in the cavity, thereby allowlng the pressurized runner to Ieed additional metal to compensate for shrinkage. Intricate shapes can be produced ar lower pressures than would normally be required for hot or cold forging Both retractable and disposable cores can be used to create holes and internal passages. Gas and shrinkage porosity are substantially reduced, and mechanical properties are enhanced.While the rqueeze casting process 1s most commonly applied to aluminum and magnesrum castings, it has also been adapted to the production of metal-matrix composites where the pressunzed metal is forced around or through foamed or fiber reinforcements that have been positioned in the mold.

For most alloy compositions, there is a range of temperatures where liquid solid coexist, and several techniques have been developed to produce shapes from ~rmisola'd material. In the rheocartlng proeess,molten metal is cooled to the semis state with constant stirrrng The stirring or shearing action breaks up the deudntes, ducing a slurry of rounded particles of solid in aliquid melt.Thii slurry, with about a solid content, can be readily shaped by high-pressure injection into metal dies B the slurry contains no superheat and is already partially solidified, it freezes qui In the fhirocasting variation, there is no handling of molten metal.The mater first subjected to special processing (stirring during solid~ficatlon as in rheocas to produce solld blocks or bars with a nondendr~tic structure. When reheated semisolid condition, the thtxotropic materral can be handled like a solid but flow liquid when ag~tated or squeezed. The solld material is then cut to prescr~bed reheated L o a semisolid state where the material 1s about 40% hquid and 60% so chanically transferred to the shot chamber of a cold-chamber die-casting machine, injected under pressure. In a variation of the process, solid metal granules or pellets fed Into a barrel chamber, where a rotating screw shears and advances the ma through heating zones that raise the temperature to the semisolid region. When c~ent volume of thixotropic material has accumulated at the end of the barrel, system drives it into the die or mold at veloc~tles of 1 to 2 5 mlsec (40-100 in. The injection system of this process is a combination of the screw feed used in p injection molding and the plunger used in conventional die castlng. In all of the semisolid casting processes, the absence of turbulent flow durin casting operation mirumues gas pickup and entrapment. Because the material is a1 partially solid, the lower injection temperatures and reduced solidiication time extend tool life The prior solidification coupled with further solidification under sure results m a significant reduction in solidification shrinkage and related por The minim~zatlon of porosity enables the use of hlgh-temperature heat treat such as theT6 solution Weatment and art~ficial aging of aluminum, to further en strength S ~ n c the e thixocastmg process does not use molten metal, both wrough cast alloys have been successfully shaped. Walls have been produced w ~ t h thtckne low as 0.2 mm (0 01 m.).

CASTING
The inertial forces of rotation or spinmng are used to distribute the molten meta the mold cavity or cavit~es 1n the centrifugal cajting processes, a catcgory that incl true centrifugal casting, semicentrhgal casting, and centrifuging. In true centrifugal mg, a dry-sand, graphite, or metal mold is rotated about either a horizontal or ve axis a t speeds of 300 to 3000 rpm. As the molten metal is introduced, it 1 s flung t surface of the mold, where it solidifies into some form of hollow product.The e profile 1s usually round (as with gun barrels, pipes, and tubes), but hexagons and symmetrical shapes are also possible. No core or mold surface is needed to shape the interior, which will always round profile because the molten metal 1s uniformly distributed by the centrifugal When rotatlon is about the horizontal axis, as illustrated in F~gure 13-8,the inner 7 is always cylindrical I f the mold is oriented vertically, as in Figure 13-9, grav~t forces cause the lnner surface to become parabolic, with the exact qhape be~ng tion of the speed of rotation.WaI1 thickness can be controlled by varying the amou metal that is introduced into the mold. During the rotation, the metal is forced agamst the outer walls of the mold considerable force, and solidification begins at the outer surface. Centrifugal force tinues to feed molten metal as solidificat~on progresses inward. Since the proces pensates for shrinkage, no risers are required. The final product has a strong, exterior with aU of the lighter impurities (includtng dross and pieces of the refra mold mating) coI1ecting on the Inner surface of the castmg This surface 1s often the fmal casting, but for some products, it may be removed by a light boring ope Products can have auts~dediameters ranging from 7.5 crn to 1.4 m 55 ~n) and wall th~ckness up to 25 cm (10 m . ) .Pipe (up to 12 m, or 40 ft, in len

S~CTI 13.5 ~ N Centrifugal Casting Sand lrning Orifice

URE 13-8 Schematic

esentation of a horizontal rifugal casting machlne rnesy o f Amencon Casr Iron AL ) Company, Blrm~ngham,

, '

'Bottom rollersr

iboioid A

\ iy

Parabolo~dB

13-9 Vert~cal

ai castmg, showing of rotational speed on e of the inner surface. old A results from fast ng, whereas slower nlng will produce iboloid B.

pressure vessels, cylinder liners, brake drums, the starting material for hearing rings, and all of the parls illustrated 111 E~gure 13-10 can be manufactuied by centr~lugal casting. The equipment 1s rather ~peciahzed and can be quite expenslve for large c a s t ~ n g s n pere manent molds can also be expenslve, but they offer a long servicc lice, especially when codted wlth some form of refractory dusl or wash. Since no sprues, gates, or risers are required, ylelds can be greater than 90%. Composite products can also be made by centrlfugal castlng of a second malerial on the Inside surface of an already-cast product. Table 13-5summarkes the features ol the centrifugal casting process 13-11) the centrifugal torce assfsts the flow of In semitentrrfu,qczlcastlng (F~gurc metal from a central reservoir to the extrem~ties of a rotating symmetrical mold.The roLational speeds are usually lower than for true centrifugal castmg, and the molds may be cither expendable or multiple-use Several molds may also be stacked on top of one another, so they can be fed by a common pourlug basln and sprue In general, the mold shape is more omplclx than for true centrifugal casting, and cores can be placed in the mold to further increase the complcx~ly of the product. The central reservoir acts as a riser and must be large enough to ensure that it will be the last material to freeze Since the tighter impurities concentrate in the center, however, the proce5s IS best used for caslmgs where the central reglon will ultimately be hollow. Conunon produclv mclude gear blanks, pulley sheaves, wheels, impellers. and electric motor rotors. Cenalfugmg, or ceretrrfu,qe tentribgo1 cartlag (Figure 13-12), uses centrifugal action to force metal from a central pouring reservoir or sprue, through spoke-type runners, into separate mold cavitles that are offset trom the axls ol rotation. Relatively low rotat~onal speeds are required to produce sound castlngs w~th thin walls and intricate shapes. Centrifugrng is often used to asstst 1 1 1the pouring o f mult~ple-product investment casting trees.

FIGURE 13-70 Electrical products (collector rings, slip rings, and rotor end rings) that have been centrifugally cast from alurnlnum and copper (Courtesy of The Electr~e Mate~~ala Company, North East, PA )

e-Use-Mold Casting,Processes

Prrtccsc Mgltenmetal is intmdncedinfo a rorating aand,metll,orgraphil~hold and held against the mold wail by ccnrnfugalforce until it is soiidified. Advunlnges; Can produce a wide rawe of Lylindrical part@, including ones of large size; good dimensionai accuracy. selmdnesi. and cleanlinessl Linlitarions: Shape is limited; spinning equipmeni e m be expasive. Common mrtsk: Iron; steet; stalnltss %.%&.andalloys @f alummum,copper. and nickel, Size ljmiia: Up t , 3 ~ m (10 fi) in diameter and 15m (50 ft) in length. T1rickne.r~ limils:Wall thickness 2.5 to 125 mm (0.1-5 in.). Typical t~dcromcek: O.D.to within 2 5 mm (U.l in): I.D. to ahaut 4 rnm 10.15 in.). Drafl allowance 10 mmlm (i in.Kt). Su&ce,%irh: 2.5-12.5 pm (10&5-500 pin.) mt%

Casting
ewoir for feeding

Revolv~ng table

w
FIGURE 13-12 Schematic of a centrifuging process. Metal is poured into the central pouring sprue and spun into the various mold cavities. (Courtesy of American Cart Iron Pipe Company, Birmingham, AL.)

of a semicentrlfugal castlng FIGURE 13-11 Schemat~c

process.

Centrifuging can also he used to drive pewter, zinc, or wax into spinning rubb molds to produce products with close tolerances, smooth surfaces. and excellent det These can bc finished products or the low-melting-point patterns that are subsequen assembled to form the "trees" [or investment casting.

CASTING
As discussed in Chapter 6 and depicted in Figure 6-5, continuous casting is usuallge ployed in the solidification of basic shapes that become the feedstock for deformati processes such as rolling and forging. By producing a special mold. continuous casti can also be used to produce long lengths of complex cross-section product. such as t one depicted in Figure 13-13.Since each product is simply a cutoCI section of the tinuous strand, a single mold is all that is required to produce a large number of pie Quality is high as well, since the metal can be protected from contamination dur~ melting and pouring, and only a minimum of handling is rcquired.

13-13 Gear produced nuous casting. -East material; (right) ~dlining. (Courtesy of , Tucson, AZ.)

. .

All casting processes begin with molten metal. Ideally, the molten metal should be available in an adequate amount, at the desired temperature, with the dcsired chemistry and minimum contamination.The melting furnace should be capable of holding material for an extended period of time without deterioration of quality, be economical to operate, and be capable of being operated without contributing to the pollution of the environment. Except for experimental or very small operations, virtually aU foundries use cupolas, air furnaces (also known as direct fuel-fired furnaces), electric-arc furnaces, electric resistance furnaces, or electric induction furnaces. In locations such as fully integrated steel mills, molten metal may be taken directly from a steelmaking furnace and poured into casting molds. This practice is usually reserved for exceptionally large castings. For small operations, gas-fired crucible furnaces are common, but these have rather limited capacities. Selection of the most appropriate melting method depends on such factors as (1)the temperature needed to melt and superheat the metal, (2) the alloy being melted and the form of available charge material, (3) the desired melting rate or the desired quantity of molten metal, (4) the desired quality of the metal, (5)the availability and cost of various fuels, (6) the variety of metals or alloys to be melted, (7) whether melting is to be batch or continuous, (8) the required level of emission control, and (9) the various capital and operating costs. The feedstock entering the melting furnace may take several forms. While prealloyed ingot may be purchased for remelt, it is not uncommon for the starting material to be a mix of commercially pure primary metal and commercial scrap, along with recycled gates, runners, sprues, and risers, as well as defective castings.The chemistry can be adjusted through alloy additions in the form of either pure materials or master alloys that are high in a particular element but are designed to have a lower melting point than the pure material and a density that allows for good mixing. Preheating the metal being charged is another common practice, and it can increase the melting rate of a furnace by as much as 30%.

CUPOLAS
A significant amount of gray, nodular, and white cast iron is still melted in cupolas, although many foundries have converted to electric induction furnaces. A cupola is a refractory-lined, vertical steel shell into which alternating layers of coke (carbon), iron (pig iron and/or scrap), Limestone or other flux, and possible alloy additions are charged and melted under forced air draft. The operation is similar to that of a blast furnace, with the molten metal collecting at the bottom of the cupola to be tapped off either continuously or at periodic intervals. Cupolas are simple and economical, can be obtained in a wide range of capacities, and can produce cast iron of excellent quality if the proper raw materials are used and good control is practiced. Control of temperature and chemistry can be somewhat

LU>LIIIg

as the

he con- 1
pxeces. during

I
1

difficult, however.The nature of the charged materials and the reactions that occur in the cupola can all affect the product chemistry. Moreover, by the time the final istry is determined through analysis of the tapped product, a substantial cha material 1s already work~ng its way through the furnace. Final chemistry adjustme therefore, are often performed in the ladle, using the various techmques of ladle me lurgy discussed in Chapter 6. Various methods can be used to increase the melting rate and improve thee omy of a cupola operatlon. In a hot-blast cupola, the stack gases are put through a exchanger to preheat the incoming air. Oxygen-enriched blasts can also be used t crease the temperature and accelerate the rate of melting. Plasma torches can be ployed to melt the iron scrap. With typical enhancements, the melting rate continuously operating cupola can be quite high, such that production of 120 ton hot metal per hour is not uncommon.

INDIRECT FUEL-FIRED FURNACES (OR CRUCIBLE FURNACES) Small batches of nonferrous metal are often melted in ~ndrrect fuel-fired furnaces are essentially crucibles or holding pots whose outer surface 1s heated by an exte flamc.The containment cruc~bles are generally made from clay and graphite, s~licon bide, cast iron, or steel Stirring action, temperature control, and chemistry control often poor, and furnace size and melting rate are limited. Nevertheless, these furn do offer low capital and operating cost. Better control of temperature and chemistry can be obtained, however,$ the cible furnaces are heated by electrical resistance heatmg. DIRECT FUEL-FIRED FURNACES OR REVERBERATORY FURNACES Direct fuel-firedfurnaces, also known aa reverberatory furnaces, are similar to open-hearth furnaces hut are less sophisticated. As illustrated in Figure 13-14, a fired flame passes directly over the pool of molten metal, with heat being transferre the metal through both radiant heating from the refractory roof and walls and co tive heatlng from the hot gases. Capacity is significantly greater than that of the nu furnace, but the operation is st111limited to the batch melting of nonferrous metals the holding of cast iron that has been previously melted III a cupola.Tbe rate of he and melt~ng and the temperature and composition of the molten metal are all e controlled. ARC FURNACES Arc furnaces are the preferred method of melting ln many foundrres because of th rapid meltlng rates, (2) ability to hold the molten metal for any desired period of and (3) greater ease of incorporatingpollution control equipment.The basic featur operating cycle of a direct-arcfurnuce can he described with the a ~ of d Figure 13-15 top of the wide, shallow unit is first lifted or swung aside to permit the introducho charge material.The top is then repomtioned, and the electrodes are lowered to cr an arc between the electrodes and the metal charge. The path of the heating curr usually through one electrode, across an arc to the metal charge, through the charge, and back through another arc to another electrode.

FIGURE 13-14 Cross section of a direct fuel-fired furnace Hot combustion gases pass across the surface of a molten metal pool.

SECTION 13.7

Melting

327

Door for charging

CURE 13-15 Schematic diagram of a three-phase

FIGURE 13-16 Electric-arc furnace, tilted for pounng. (Courtesy of Lectromelt Corporotfon, P~ttsburgh,PA)

Fluxing materials are usually added to create a protective slag over the pool of molten metal. Reactions between the slag and the metal serve to further remove impurities and areefficientbecause of thelarge interface area and the fact that theslag is as hot as the metal. Because themetal is covered and can be maintained at a given temperature for Long periods of time,arc furnaees can beused to produce high-quality metal of almost any desired composition.They are available in sizes up to about 200 tons (but capacities of 25 tons or less are most common), and up to 50 tons per hour can be melted conveniently in batch operations Arc furnaces are generally used with ferrous alloys, especially steel, and provide good mixing and homogeneity to the molten bath. Unfortunately, the noise and level of particle emissions can be rather high, and the consumption of electrodes, refractories,andpower results in high operating costs. Figure 13-16shows the pouring of an electric-arc furnace. Note the still-glowing electrodes at the top of the furnace.

INDUCTION FURNACES
Because of their very rapid melting rates and the relative ease of controlling pollution, electric inductionfurnaces have become another popular means of ~nelring metal.There are two basic types of induction furnaces. The high-frequency, or coreless units, shown schematically inFigure 13-17,consist of a crucible surrounded by a water-cooled coil of copper tubing. A high-frequency electrical current passes through the coil, creating an alternating magnetic field.The varying magnetic field induces secondary electrical eurrents in the metal being melted, which bring about a rapid rate of heating. Coreless induction furnaces are used for virtually all common alloys, with the maximum temperature being limited only by the rehactory and the ability to insulate against heat 1oss.Theyprovide good control of temperature and composition and are available in a range of capacities up to about 65 tons. Because there is no contamination from the heat source, they produce very pure metal. Operation is generally on a batch basis. Low-frequency or channel-type induction furnaces are also seeing increased use. As showninFigure 13-18,onlya small channelis surrounded by the primary (current-carrying or heating) coi1.A secondary coilis formed by a loop, or channel, of molten metal, and all the liquid metal is free to circulate through the loop and gain heat.To start,enough molten metal must be placed into the furnace to fill the secondary coil, with the remainder of the charge taking a variety of forms.The heating rate is high, and the temperature can be accurately controlled. As aresult, channel-type furnaces are often preferred as holding fure the molten metal is maintained at a constant temperature for an extended e. Capacities can be quite large,up to about 250 tons.

ss

induction furnace.

Pneumat~cally operated l ~ d

Molten

metal

Insulating

Channel wh~ch forms metal loop or secondary co~l

Plug

firebrick

FIGURE 13-18

Cross section showinq the principle of the low-frequency or channel-type induct~on furnace.

Some type of pouring device, or ladle, is usually required to transfer the meta melting furnace t o the molds.The primary considerations for this operation are (1 maintain the metal at the proper temperature for pouring and (2) to ensure that I high-quality metal is introduced into the molds. The specific type of pouring ladle is termined largely by the size and number of castings to be poured. In small foundrim handheld, shank-type ladle is used for manual pouring. In larger laundries, either tom-pour or teapot-type ladles are used, like the ones illustrated in Figure 11-6.Tl are often used in conjunction with a conveyor line that moves the molds past the pt ing station. Because metal is extracted from beneath the surface, slag and other in rities that float on top of the melt are not permitted to enter the mold. High-volume, mass-production foundries often use automatic pouring systt like the one shown in Figure 13-19. Molten metal is translerred k o m a main me1 furnace L o a holding furnace. A programmed amount of molten metal is further t r ferred into individual pouring ladles and is then poured into the corresponding rn,

FIGURE 13-19 Autornat~c

pourlng of molds on a conveyor lhne (Courtesy of Roberts Smto Corporat~on, Lansmng, MI)

as they traverse by the pouring statlon. Laser-based control umts position the pouring
ladle over the sprue and control the flow rate into the pouring cup.

CLEANING AND FINISHING After solidificat~on and removal from the mold, most castings requlre some additional cleamng and finishing. Specific operations may include all or several of the followmg:

1 . Removing cores 2. Removing gates and nsers 3. Removing fins, flash, and rough spots from the surface 4. Cleaning the surface 5. Repalring any defects
Cleanlng and fin~shing operations can be quite expensive, so cons~deration should be given to thelr minimization when designing the product and selecting the specific method of casting. In addition, consideration should also be dlrected toward the possibility of automating the cleaning and li~shing. Sand cores can usually be removed by mechan~cal shaking.At tlmes, however, they must be removed by chemically dissolving the core binder. On small cashngs, sprues, gates, and risers can sometimes be knocked off For larger castings, a cutting operation 1 s usually required. Most nonferrous metals and cast Irons can be cut with an abrasive cutoff wheel, power hacksaw, or band saw. Steel castings treqnently requlre an o x y acetylene torch.Plasma arc cutting can also be used. The specitlc method of cleanlng often depends on the size and complex~ty of the castmg. After the gates and rlsers have been removed,small castings are often tumbled in barrcls to remove fins, flash, and sand that may have adhered to the surface.Tumblmg may also be used to remove cores and, in some cases, gates andrlsers. Metal shot or abrasive material is often added to the barrel to aid in the cleaning. Conveyors can be used to pass larger castings through special cleaning chambers, where they are subjected to blasts of abras~ve or cleaning material. Extremely large castings usually requlre manual h i s h n g , using pneumatic chisels,portable gnndei.~, and manually dlrected blast hoses. While defect-free castings are always desued, flaws such as cracks, voids, and laps are not uncommon J n some cases, especially when the part i? large and the production quantity is small, it may be more attractive to repair the part rather than change the pattern, die, or process. If the material is weldable, repairs are often made by removing the defectwe reglon (usually by chipplng or grindlngl and f~lling the created void with deposited weld metal. Poros~ty that is at or connected to free surfaces can be filled with reainous material, such as polyester, by a process known as ampregnation. If the pores are Ellled wlth a lower-melting-pointmetal, the process becomes mn~5ltratron. [See Chapter 16 for a further discnss~on of these processes.)

HEAT TREATMENT AND INSPECTION OF CASTINGS Heat treatment is an attractivemeans of altering properties while retaining the shape of the product Steel castmgs are frequently given a full anneal to reduce the hardness and brittleness of rapidly cooled, thm sechons and to reduce the internal stresses that result from uneven mling.Nonferrous castmgs are often heat treated to provide chemical homogenization or stress relief as well as to pepare them for subsequent machinlug For hnaI properties, virtually all of the treatments discussed in Chapter 5 can be applied Ferrous-metalcast~ngs often undergo a quench-and-temper treatment, and many nonferrous castmgs are age hardened to Impart additional strength.The variety of heat treatments is largely responsible for the wide range of properties and character~tlcs available in cast metal products Virtually all of the nondestructive rnspectzon techniques can be applied to cast metal products. X-ray radiography,11qu1dpenetrant inspection, and magnehc particle mspectiou are extremely common.

UTION IN

FOUNDRY OPERATIONS

Many of the operations that are performed in a foundry are ideally suited for robotica tomation since they tend to be dirty, dangerous, or dnll. Robots can dry mold6 coat car vent molds, and clean or lubricate dies. Tley can tend stationary, cyclic e q ~ p m e n tsu , as die-cashng machines, and if the machines are properly grouped, one robot can oftense vice two or three machines In the finisHing room,robots can be equpped with pl ters or torches to remove sprues, gates, and runners.Theycan perform grinding an operations, as well as various functions involved in the heat treatment of castrnys In the investment-casting process, robots can be used to dip the wax pattern refractory slurry and produce the desired molds. In a sunilar manner, robots have used to dip the Styrofoam patterns of the full-mold and lost-foam processes in th refractory coating and hang them on conveyors to dry In a fully automated lost-fo operation, robots could be used to position the pattern, fill the flask w t h sand,pour metal, and use a torch to remove the sprne.

.ss SELECTION
As shown In the individual process summaries that have been included through0 Chapters 12 and 13, each of the casting processes has a charactenstic set of capabilih assets, and hmitations, The requirements of a particular product (such as size, comp ity, required dimensional precision, desired snrface fmish, total quantity to be made, desired rate of product~on) often limit the number of processes that should be wns ered as production candidates. Further selection is usually based on cost. Some aspects of product cost, such as the cost of the material and the energy qured to melt it, are somewhat mdependent of thespecific process.The cost of other f tures, such as patterns, molds, dies, melting and pouring equipment, scrap materi cleaning,mspection, and all related labor, can vary markedly and be quite dependent the procesr. For example, pattern and mold costs for sand casting are quite a hit 1 than the cost of die-casting dies. Die casting, on the other hand, offers high producb rates and a high degree of automation. When a small quantity of parts is dedred, theca of the die or tooling must be distributed over the total number of parts, and unit cost cost per casting) is high. When the total quanbty is large, the toolmg cost is distrihut over many parts, and the cost per piece decreases. Figure 13-20 shows the relationship between unit cost and production quantit a product that can be made by both sand and die casting. Sand casting is an expen mold process Since an individual mold is required for each pour, increasing quan does not lead to a significant drop in unit cost. Dle casting involves a multiple-use mo and the cost of the die can be distributed over the total number of parts. As shown the figure, sand casting is often less expensive for smaIl production runs, and process such as die casting are preferred for large quantities. One should note that whde dle-casting curve in Figure 13-20 is a smooth line,it is not uncommon for an actual cu to contaln abrupt d~scontmuities. If the lifetime of a set of tooling is 50,000 casts,

II
FIGURE 13-20 Typical unit cost of casttngs comparing sand castinq and die casting. Note how the large cost of a die-castlng die diminishes as it IS spread over a larger quantity of parts
+ .C

Sand cast\((

Die cast

quantity Product~on

Review Questions

331

cost perpart for 45,000 pieces, using one set of tooling, would actually be less than for 60,000 pieces, since the latter would require a second set of dies. In most cases, multiple processes are reasonable candidates for production, and the curves for all of the options should be included.The final selection is often based on a combination of economic, technical, and management considerations. Table 13-6 presents a comparison of casting processes, including green-sand casting, chemically bonded sand molds (shell, sodium silicate, and air-set), ceramic mold and investment casting, permanent-mold casting, and die casting. T h e processes are compared on the basis of wst for both small and large quantities, thinnest section, dimensional precision, surface finish, ease of casting a complex shape, ease of changing the design while in production, and range of castable materials.

Green-Sand Casting Lowest L o w

Chem~cally Bonded Sand (Shell, Sodrum S~hcate, Air-Set) Med~um h~gh Medium high

Ceram~c Mold and Investment Casting Medtum

Permanent-Mold Cast~ng
Migh

Die Casting
Hlghesl

35

0.001-0 015
Fair to good

Good

Very good

Good
Fair

Best

Far to good
Bert
Fair

Fair Unlimited

Poor

Poorest Low-melting-po~nt
metals

Low-meltlng-point metals

Dl Key Words
heat checking heat treatment heated-manifold directinjection die casting hot-chamber die-casting machine impregnation indirect fuel-fired furnace induction furnace infiltration inserts Inspection low pressure permanent-mold castrng permanent-mold casting pourlng ladle retractable cores reverberatory furnaces rheocasting sem~centrrfugal castmg semipermanent mold castmg semisohd castlng slush casting squeeze casting thermalfatlgue tluxocasting th~xatrop~c matenal true centrifugal castlng vacuum permanent-mold castlng venting

Fll Review Questions

at are some of themajor d~sadvantages of theexpendabled casting processes? at are some posstble Itmitations of mulbple-use molds? at are some common moLd matwlats for permanent-mold mg" What aresome of the metals morecommonly cast? 4. Describe some of the process advantages of permanent-mold castmg. 5. Why m~ght low produn~on runs he unattraetlve for permanent-mold casting? 6. What features affectthe he of a permanent mold?

7. How is venting provided in the permanent-mold process?

8. Why are permanent-mold castings generally removed from the mold immediately after solidification has been completed? 9. What types of products would be possible candidatesfor manufacture by slush casting? 10. How does low-pressure permanent-mold casting differ from the traditional gravity-pour process? 11. What are some of the attractive features of the low-pressure permanent-mold process? 12. What are some additional advantages of vacuum permanentmold casting over the low-pressure process? 13. Conlrast the feeding pressures on the molten metal in lowpressure permanent molding and die casting. 14. Contrast the materials used to make dies for gravity-pour permanent-mold casting and die casting.Why is there a notable difference? 15. By what mechanisms do die-casting dies typically fail? 16. Why might it be advantageous to vary the pressure on the molten metal during the die-casting cycle? 17. For what types of materials would a hot-chamber die-casting machine he appropriate? 18. What malals are routinely cast with cold-chamber die-casting machines? 19. How does the air in the mold cavity escape in the die-casting process? 20. Are risers employed in die casting'? Can sand cores be used? 21. What are some of the sttractive features of die casting compared to alternative casting methods? 22. Whcn might low quantities be justified for the die-casting process? 23. Describe the squeeze casting process. 24. What is a thixotropic material? How does it provide an attractive alternative to squeeze casting o r rheocasting? 25. What are some of the attractivc features of semisolid casting? 26. Contrast the structure and properties of the outer and inner

surfaces of a centrifugal casting. 27. What are the key differences between true centrifugal ca in& semicentrifugal casting, and centrifuging? 2%. How can continuous casting - be used in the direct producti of products? 29 What are some of the factors that ~nfluence the selectlono operabun? furnace type or melt procedure m a cast~ng 30. What are some of the possible feedstock mater~als thatmay put ~nfoundry melt furnaces? 31. What types of metals are commonly melted m cupolas7 32. What are some of the ways that the melting rate of a cup can he increased? 33. What are some of the pros and cons of ~ndirect fuel-fir furnaces? 34. What are some of the attractive features of arc furnace foundry applications? 35. Why are channel induction furnaces attractive for met holding applications where molten metal must be held a specified temperalure for long periods of time? 36. What are the primary functions of a pouring operation? 37. What are some of the typical cleaningand finishing operati that are performed on castings? 38. What are some common ways to remove cores from cas To remove runners, gates, and risers? 39. What are some of the alternative methods of cleaning finishing castings? 40. How might defective castings be repaired to permit succe use in their intended applications? ,41. What are some of the ways that industrial robots can be played in metal-casting operations? 42. Describe some of the features that affect the cost of a producl.Why might the cost vary significantly with the q tity to be produced? 43. What are some of the key factors that should be consid when selectmg a casting process?

I
asqlate for

caze study

333

a Household
2. What material or materials would appear to be strong candidates' 3 What are- some posslble means of producing the deslred &ape? Wh~ch w u M you pretafl Coold the heat~ng element assembly be ~ncorporated during manufacture, or does it have to be added as a secnndery operation? What are the major advanwges of the method you propose? 4. Could all of the design featmes (holes, webs, and recesses) be incorporated in the rnitlai manutactur~ng aperation, or would secondary processing be requrred? if secondary processing IS required, for what features, and h w would you recommendthat they be produced? 5. Some commenrai nons have baseplates for wh~ch the bottom surfaces have been finlshed by asimpie buff and pollsh, wh~ie others have aTeflon coatlng or have been anodized. If your deslre 1 s to produce a high-quality product, what form of surfaeefrnishtng would you recotnmend?

he ~ t depccted m in the flgur high-quality household steal at up to 1200 watts and is d dy steam and burst of stear to the des~gn IS an integral ele seshoe" that must be therm. e but remain electr~cally tnsi n takes the form of a reslsta ded by ceramic insuiatlon, ; )The steam emerges thrm holes in the base, each abc re are about a dozen larger out Ti8 Inch rn diameter, that e vanous components , Discuss the various features possess in order to funtilon I Gonslder strength, impact re< twlty, conoslon resrstance, w

14.1 INTRODUCTION 14 2 FABRIC~TION OF PLAFTICS Casting Blow Moldmg Compression Molding or Hot-Compression Molding Transfer Mold~ng Cold Molding Injection Molding Rcaction Injection Moldiig Extrusion Thermoformig Rotat~onal Moldmg Foam Molding

Other Plastic-Forming Processes Mach~ning o f Plast~cs Finishing and Assembly Operahons Designing for Fabricallon Inserts Design Factors Related to Finishing or RUBBER AND 14.3 PROCESSING ELASTOMERS OF CERAMICS 14 4 PROCESSING Fabncat~on Techn~ques for Glasses Fabrication of Crystalline Cerarmcs

Producing Strength in Particulate Ceramics. Machining of Ceramics Joining of Ceramies Design .of Ceramic Co 14.5 FABRICATTON OF COM MATERIALS Fabrication of Particulate Composites Fabrication of Laminar Composites Fabrication of Fiber Composites Case Study: FABRICAT B LAVATOW WASH

In Chapters 6,7, and 8,plartlcr; cemmlcr, and composctes were shown to be different from metals in both structure and propertles. It is reasonable to fore, that the principles of material selection a i d product design, as well as the fab tion processes, will also he somewhat different. In addition, there will also be s similarities.The specific material will still be selected for its ability to provide the req properties and the fabrication processes for their ability to produce the desired in an economical and practical manner. In terms of differences,plastics, ceramics, and composites tend to be used clos 10 their design limits, and many of the fabrication processes convert the raw material 10 a finished product in a single operalion. Large, complex shapes can often be forme 21s s: a single unit, eliminating the need for multipart assembly operations. Materials i classes can often provide integral and variable color*and the processes used to Ufacture the shape can frequently produce the desired finish and precision. As a r II. finishing operations are often unnecessary-an attractivc feature since, for ma ill these materials, altering the final dimensions or surface would be both difficult and >I. 1y.The joining and fastening operations used with these materials also tend to be diff 111 from those uscd with metals. As with metals, the properties of these matenals are affectedby the processes to produce the shape.The fabrication of an acceptable product, therefore, involve selection ot both (1) an appropriate material and (2) a cornpanton method of pro ing,such that the recult~ng combination prov~des the desired shape, propertles. prec and finish.

The manufacture of a successful plastic product requires satlsfyrng the various chanlcal and physical property requirements through the use of the most econo resin or compound that will perform sat~sfactorily, and coupling it with a manufac process that is compatible w~th both the part deslgn and the selected material.

Chapter 8 presented material about the wide variety of plastics or polymers that are currently used as engineering materials. As we move our attention to the fabrication of parts and shapes, we find that there are also a variety of processes from which to choose. Determination f ' the preferred method depends on the desired size, shape, and quantity, as well as whether the polymer is a thermoplastic, thermoset,or elastomer.Thermoplastic polymers can be heated to produce either a soft, formable solid or a 1iquid.The material can then be cast, injected into a mold, or forced into or through dies to produce a desired shape.Thermosetting polymers have far fewer options, because once the polymerization has occurred, the framework structure is established and no further deformation can occur. Thus the polymerization reaction must take place during the shape-forming operation. Elastomers are sufficiently unique that they will be treated in a separate section of this chapter. Casting, blow molding, compression molding, transfer molding, cold molding, injection molding reaction injection molding, e.rtrusion, thermoforming rotational molding, andfoam molding are all processes that are used to shape polymers. Each has its distinct set of advantages and limitations that relate to part design, compatible materials, and production cost. To make optimum selections, we must be familiar with the shape capabilities of a process as well as how the process affects the properties of the material.

CASTING
Casting is the simplest of the shape-forming processes because no fillers are used and no pressure is required. While not all plastics can be cast, there are a number of castable thermoplastics, including acrylics, nylons, urethanes, and PVC plastisols. The thermoplastic polymer is simply melted, and the liquid is poured into a container having the shape of the desired part. Several variations of the process have been developed. Small products can be cast directly into shaped molds. Plate glass can be used as a mold to cast individual pieces of thick plastic sheet. Continuous sheets and films can be produced by injecting the liquid polymer between two moving belts of highly polished stainless steel, the width and thickness being set by resilient gasket strips on either end of the gap. Thin sheets can be made by ejecting molter1 liquid from a gap-slot die onto a temperature-controlled chill roll.The molten plastic can also be spun against a rotating mold wall (centrifugal casting) to produce hollow or tubular shapes. Some thermosets (such as phenolics, polyesters, epoxies, silicones, and urethanes) can also be cast, as well as any resin that will polymerize at low temperatures and atmospheric pressure. Because of the need for curing, the casting of thermoset resins usually involves additional processing, often some form of heating while in the mold. Figure 14-1 depicts a process where a steel pattern is dipped into molten lead, withdrawn, and allowed to coo1.A thin lead sheath is produced when the pattern is removed, and this becomes the mold for the plastic resin. Curing occurs, either at room temperature or by heating for long times at temperatures in the range of 65" to 95C (150' to 200F). After curing, the product is removed, and the lead sheaths can be reused. Since cast plastics contain no fillers, they have a distinctly lustrous appearance, and a wide range of transparent and translucent colors are available. Since the product is shaped as a liquid,fiber or particulate reinforcement can be easily incorporated.The process is relatively inexpensive because of the comparative lack of costly dies, equipment, and controls. Typical products include sheets, plates, film&rods, and tubes, as well

Steel

mandrel

a/

Molten lead

Lead shell

castlna o f plastic parts using a lead>heli mold

as small objects,such as jewelry, ornamental shapes, gears,and 1 e n s e s . W e dlmenslo i x precis~on can be quite high, quality problems can occur because of inadequate m alr entrapment, gas evolution, and shrinkage.

BLOW MOLDING A variety of blew molding processes have been developed, the mo5t common used to convert thermoplastic polyethylene, polyvinyl chloride (PVC), polyprop and PEEK resins Into bottles and other hollow-shape c0ntainers.A solid-bottom,ho tube preform, known as apartson, 1s made from heated plastic by either extrusio injection molding. The heated preform is then positioned between the halves of a mold, the mold closes, and the preform is expanded agalnst the mold by air or gasp sure.The mold is then cooled, the halves separated, and the product is removed An is then trimmed for direct recycling. Figure 14-2 deplcts a form of this process the starting material 1s a s~mple tube and the solid bottom is created by the pinc action of die closure. Blow molding has recently expanded to Include the enginee thermoplastlcs and has been used to produce products as d~verse as antomotrve tanks, seat back$ ductwork, and bumper beams. Variations of blow molding have been des~gned to prov~de both axla1 and expansion of the plastic (for enhanced strength) as well as to produce multilayered ucts. In one process, a sheet of heated plastlc is placed between upper and lower ties, the lower one having the shape of the product. Both cavities are then press to 2 to 4 MPa (300 to 600 psi) with a nonreactive gas such as argon. When the pr in the lower segment 1 s then vented, the gas in the upper segment "blows" the mat into the lower die cavity. Because the thermoplastlcs must be cooled before removal from the mold molds for blow moldlng must contain the desired cavlty as well as a cooling sys venting system, and other design features The mold material must provlde therm ductlvlty and durability whlle berng inexpensive and compatible with the resin processed. Beryllium copper, alummum, tool steels, and stainlesb steels are all po mold materials COMPRESSION MOLDING OR HOT-COMPRESSION MOLDING In compression molding, illustrated schematically in Figure 14-3,sohd granules o formed tablets of unpolymerized plastic are introduced into an open, heated c A heated plunger then descends to close the canty and apply pressure.As the m a1 melts and becomes flu~d, it 1s driven into all portions of the cavlty The heat and sure are maintamed until the mater~al has "set" (t.e., cured or polymerized).lle 1s then opened and the part is removed. A wide variety of heating systems and materials are used, and multiple cavities can be placed with~n a mold to produce than one part m a s~ngle pressing. The process is simple and used pslmarily w i thermosett~ng polymers, although recent developments permt the shaplng of th plastics and compos~tes Cycle times are set by the rate of heat transfer and the re or curing rate of the polymer.They typically range from under 1minute to asm 20 mlnutes or more.

FIGURE 14-2 Steps in blow molding plastic parts (1) a tube of heated plastic 1 s placed in the open mold; (2) the mold closes over the tube, simultaneously sealing the bottom, (3) alr expands the tube against the sides of the mold; and (4) after suff~clent cooling, the mold opens to release the product

FIGURE 14-3 The hotcompression molding process: (1) solid granules or a preform pellet is placed in a heated die; (2) a heated punch descends and applies pressure; and (3) after curing (thermosets) or cooling (thermoplastics), the mold is opened and the part is removed.

Heated punch
Flash

Flash

cleara
Preheated pellet Heated die

Ejector

The tool and machnery cost3 for compression molding are often lower than for competmg processes, and the dimensional preci~ion and surface finish are high, thereby reduclng or eliminating secondary operations Compression moldlng is most econom~cal when it is applied to small prodnct~on runc of parts requiring Close tolerances, h~gh Impact strength, and low mold shnnkage. It is a poor cho~ce when the part contains thick sections (the cure tlmes become quite long) or when large quantitie~ are desired. Most products have relatively slmple shapes because the flow of materlal is rather limited. Typical compress~on-moldedparts include gaskets, seals, extenor automotive panels, alrcraft fairings, and a wlde variety of lnterior panels More recently, compression molding has been used to form fiber-reinforced plastics, both thermoplastics and thermosets, Into parts with properties that nval the englneering metals. In the thermoset famlly, polyesters, epoxies, and phenolics can be used as the base of fiber-containing sheet-molding compound, bulk-moldmg compound, or sprayed-up reinforcement mats. These are introduced Into the mold and shaped and cured in the normal manner. Cycle bmes range from about 1to 5 minutes per part, and typical products Include wash basins, bathtubs, equlptnent hous~ngs, and various electrical components. If the startlng material 1s a fiber-contaming thermoplastic, precut blanks are first heated in an Infrared oven to produce a soft, pliable matenal The blanks are then transferred to the press, where they are shaped and cooled in spec~ally deslgned dies Compared to the thermosets, cycle times are reduced and Lhe scrap is often recyclable. In addition, the products can be joined or assembled using the thermal "welding" processes appl~rd to plastics. Compression moldrng equipment is usually rather simple, typically consisting of a hydraulic or pneumatic press with parallel platens that apply the heat and pressure Pressing areas range from 15 cm2(6 in.') to as much as 2.5 m2 (8 ft2),and the force capacities range from 6 to 9000 metric tons.The molds aIe usually made of tool steel and are polished or chrome plated to improve materlal flow and product quahty. Mold temperatures typically run between 150' and 200C (300' and 400F) but can go as high as 650C (1200F) They are heated by a variety of means, including electrrc heaters,steam, oil, and gas.

TRANSFER MOLDING
Transfer moldrng 1s sometimes used to reduce the turbulence and uneven flow that can result from the high pressures of hot-compression molding As shown in F~gure 14-4, the unpolymenzed raw material is now placed m a plunger cavity, where it m healed until molten.The plunger then descends forcing the molten plastic through channels or runners into adjo~ning die cavitles.Temperature and pressure are maintained until the thermosetting resin has completely cured.To shorten the cycle and extend the hfetime of the cavity, plunger, runner, and gates, the charge material may be preheated before being placed in the plunger cavlty. Because the material enters the d1e cavities as a liqu~d, there is little pressure until the cavity is completely filled. Thm sections, excellent detail, and good tolerances and finish are all characteristics of the process. In add~tion, inserts can be incorporated into the products of transfer molding.They are s~mply po~ttioned wrthin the cavity and maintained In place as the liqmd resln is introduced around them.

14-4 Diagram of the

olding process. Molten d material is first in the upper heated plunger then drives the into an adjacent die.

relatively complex shapes. It combines elements of both compression moldmg

chanical or electrical properties and reduce shrinkage or warping. The maln

parts that requlre good resistance to chemical attack

COLD MOLDING
and more econom~cal, the resulting products generally lack good surface fin~sh a mensional precision.

INJECTIONMOLDING

hydro-mechanical.

a finished product needtng no further work before assembly or use.

FIGURE 14-5 Schematic

diagram of the injection molding process. A moving plunger advances material through a heating region (in this case, through a heated manifold and over a heated torpedo) and further through runners into a mold where the molten thermoplastic cools and solidifies.

cola the maeerial in the runner solidX1es with each cycle and needs to be ejected and reprocessed or disposed of. With hot runners, the thermoplastic material is maintained in a liquid state until it reaches the gate.The material in the runners can be used in the mbsequent shot, thereby reducing shot ~ i z and e cycle time,srnce less material must be heated. Quality i s improved,since all material enters the mold at the same temperature, recy~led sprues and runners are not ~ncorporated into the charge, and there is less turbulence since pressurized matenal is not ~njecied into empty runners. Hot runners do add an additional degree of complex~ty to the design, operation, and control of the system,so the additional cost must be weighed against the cited benefits. Injection molding can also be applied to the thermosetting materials, but the process must be modified to provide the temperature, pressure, and tune required for curing.The injection chamber is now cold, and the mold is heated.The time in the heated mold must be sufficient to complete the curing process, and the relatively long cycle times are the majm deterrent to the injection molding of the thermosets.

REACTION INJECTIONMOLDING
F~gure 14-6 depicts the reactron rnject~on molding process, in which two or more liquid reactants are metered into a unit where they are intimately mixed by theimpingement of liquid streams that have been pressurized to a value between 13 and 20 @a (2000 and 3000 psi). The combined material flows through a pressure-reducing chamber and exits the mixhead directly into a mo1d.h exothermic chemical reaction takes place between the two components, resulting in thermoset polymerization. Since no heating is required,the producbou rates are set primarily by the curing time of the polymer,wh~ch is often less than 1minute. Molds are made from steel, aluminum, or nickel shell, with selection being made on rhe basis of number of parts to be made and the desired quality. The molds are generally clamped in low-tonnage pressea. At present, the domlnant materials for reaction injectmn molding are polyurethanes,polyamldes, and composites containing short fibers or flakes. P~operties can span a w~de range, depeudii on the combination and percentage of base chemicals and the additives that are used.Different formulations can result in elastomericor flexible, structural foam (foam core with a hard, solid outer skin), solid (no foam core), or composite produots.Part Size can range frofn to 50 kg (1to 100 ib),shapes can be quite complex (with variable wall thickness), and surface f~nish is excellent. Automotive. applicationsinclude steering wheels, airbag covers instrument panels, door panels, armrests, headliners,and center consoles, as wdl as body panels,bumpers,and wheel covers. Rigid polyurethanes are alsa used in such products as computer housings, household refrigerators, watkr skis, hot-water heatei-s, and picnic coolers, From a manufacturingperspechve, reactron Injection molding has a number of attractive features. The low processing temperatures and low injection pressures make the process attractive for molding large parts, and the large slze canoften enable parts consolidation. Thermoset parts ran generally be fabricated with less energy than the

Reactant tank

f reacrants are tolnbineo in the AGURE 14-6 The reaction 'nlection molding process. (Left) Measured amounts o molo. (Right)Alter s~ff:c;ent ct.r'ng, the mold is opcned and the component mixing head antcl :nlected into the spl~t is ejected.

Split mold

injection molding of thermoplastics, with similar cycle times and a similar d automation. 711e metering and mixing equipmenf and related controls tend to sophisticated and costly, but the lower temperatures and pressures enable the cheaper molds, which can be quite large.

EXTRUSION Long plastic prMlucts with unilorm cross sections can be readily produced by extrusion process depicted in Figure 14-7. Thermoplastic pelIets or powders are through a hopper into the barrel chamber of a screw extruder. A rotating screw pro the material through a p~eheating section, where it is heated, homogenized, and pressed, and then forces it through a heated die and onto a conveyor belt.To pres its newly imparted shape, the material is cooled and hardened by jets of air or spray water. It continues to cool as it passes along the belt and is then either cut into len or coiled, depending on whether the material is rigid or flexible and on the desires o customer.The process is continuous and provides a cheap and rapid method of m ing. Common production shapes include a wide variety of constant cross-section pro such as window and trim molding, as well as tubes,pipes, and even coated wires an bles. Thermoplastic foam shapes can also be produced. If an emerging tube is exp ed by air pressure, allowed to cool, and then rolled, the product can be a double I of sheet or film. THERMOFORMINC In the thermoforming process, thermoplastic sheet material is first heated to a war temperature.The starting material can be either discrete sheets or a continuousr material. If continuous material is used,it is usually heated by passing through an or other heating device.The material emerges over a male or female mold and is fo by the application ol vacuum, presure, or another mechanical tool. Cooling occurs contact with the mold, and the product hardens in its new shape. After sufficient ing, the part can be removed from the mold and trimmed, and the unused strip rn a1 is diverted for recycling. Figure 14-8 shows the process using a female mold cavity and discrete sheets o terial. Here the material is placed directly over the die or pattern and is heated in Pressure andor vacuum is then applied, causing the material to draw into the c The female die imparts both the dimensions and finish or texture to the exteriors The sheet material can also be stretched over male form blocks, and here the tooling trols dimensions and finish on the interior surface. Mating male and female di also be used. An entire qiclerequires only a few minutes. While the starting material is a uniform-thickness sheet, the thickness of the ucts will vary as various regions undergo stretching.Typical products tend to be shaped, thin-walled parts, such as plastic luggage, plastic trays, panels for light fix or even pages of Braille text lor the blind.
'

ROTATIONAL MOLDING Rotationul molding can be used to produce hollow, seamless products of a wide v of sizes and shapes,including storage tank$ bins and refuse containers, doll parts balls, helmets,and even boat hulls. The process begins with a closed mold or cavit

Hopper

FIGURE 1 4 7 A screw extruder produc~ng thermoulast~c ~roduct. Some

Heat lamps on

Hear lamps off

At Y pe of where heets are shaped atlon of heat and

\ Vacuum pump off

Vacuum pump on

has been filled with a premeasured amount of thermoplastic powder or liquid. The molds are either preheated or placed in a healed oven and arc then rotated simultaneously about two perpendicular axes. Other designs rotate the mold about one axis while tilting or rocking about another. In either case, the resin mclts and is distributed in the form of a uniform-thickness coating over all of the surfaces of the mold. The mold is then transferred L o a cooling chamber, where the motion is continued and air or water is used to slowly drop the temperature. After the material has solidified, the mold is opened and the uniform-thickness, hollow product is removed. All o l the starting material is used in the product; no scrap is generated. The lightweight rotational molds are frequently made from cast aluminum, but sheet mctal is often used for lal.ger parts; electroformed or vaporformed nickel is used when fine detail is to he reproduced.

FOAM MOLDING
Foamedplnstlc products have become an impostant and widely used form of polymer. In foam molding, a foaming agent i s mixed with the plastic resin and releases gas or volatilizes when the material is heated during molding.Thc materials expand to 2 to 50 times their o 640 giL (2 to 40 lb/ft3). original size, resulting in products with densities ranging from 32 L Opm-cellfoams have interconnected pores that permit the permeability of gas or liquid. Closed-cell f o a m have the property of being gas- or liquid-tight. Both rigid and flcxible h a m s have been produced using both thermoplastic and thermosetting materials. The rigid type is useful for structural applications (including housings for computers and business machines), packaging, and shipping containers; as patterns for the full-mold and lost-foam casting processes (see Chapter 12); and for injection into the interiors of thin-skinned metal components, such as aircraft fins and stabilizers. Flexible foams are used primarily lor cushioning. Variations of the conventiollal molding processes can be used to produce a variety of unusual products. By introducing foaming material into the interior of a mold that has partly filled, parts can be produced with a solid outer skin and a rigid foam core. Figure 14-9 shows the cross section of a plastic gear with this type of dual structure.

FIGURE 14-9 A plastic gear a sol~d outer sk~n and a w~th r~g~ foam d core. (From Amer~can ) Mach~nlst

OTHER PLASTIC-FORMING PROCESSES In the calendering process, a mass of dough-like thermoplast~c is forced between and two or more counter-rotating rolls to produce thin sheets or films of polymer, whi thencooled to induce hardenmg. Product thicknesses generally range between 0. 1.0 mm (O.Ql to in.) but can be reduced further to as low as 0.05 mm (0.002 m.) b sequent stretchmg. Embossed deslgns can be ~neorporated into the rolls to pr products with textures or patterns. Conventional drawing can be used to producefibers, and rollzng can be perfo to change the shape of thermoplastic extrusions. In addition to changing the p dimensions, these processes can also serve to induce crystalhzation, or produce ferred orientation to the thermoplast~c polymer chams Filaments, fibers, and yarns can be produced by rpmnmg, a modified form trusion. Molten thermoplastic polymer 1 s forced through a die containing many holes.Wbere multistrand yarns or cables are desired, the dles can rotate or spin to duce the twists and wraps. The var~ous plastic-forming processes are often combined in elther sequen mtegrated forms to produce spec~fic products For example, the closed-bottom p for blow molding can be formed as a separate injection moldlng or as an integrate trusion operation that is then followed by blow molding.Another example 1s the ufacture of thin plastic bags, such as those that are used as htchen or bathroom tr hners. Polymer granules flow through a hopper and enter the barrel of a screw e er. As the screw drives the material forward. ~t is melted and driven through an ended metal dir: that forms a thm-walled plastic tube.Aii flows through the center die, causing the diameter of the tube to expand substantially as it emerges fro constraint. Air jets around the cncumference of the expanded tube then coo plastic matenal, after wh~ch it is passed through flattening rolls. The flattened then periodically seam welded and perforated, and wound on a roll for easy dispe MACHINING OF PLASTICS Plastlcs can be milled, sawed, drilled, and threaded much like metals, but theirpr ties are so variable that it is impossible to give descriptions that would be correctf It may be more important to consider some of the general characteristrcs of plasti affectthen machinability Since plastics tend to be poor thermal conductors,most heat generated during chrp formation remams near the cut interface and is no ducted Into the material or carried away in the chips Thermoplastics tend to soft swell, and they occasionally bind or clog the cutting tool. Considerable elasticflexi also occur, and this couples with material softening to reduce the precision of fi mensions. Because the thermosetting polymers have higher rigidity and reduce enlng, they generally machine to more precise dimensions. The h ~ g h temperatures that develop at the point of cuttmg also cause th that are machiningplast~c to run very hot, and they may fall more rapidly than wh ting metal. Carbide tools may be preferred over high-speed tool steels if the cuts moderate durat~on or if high-speed curting is performed. Coolants can often advantageously if they do not discolor the plast~c or induce gumming. Water,so and water, and weak solut~ons of sod~um silicate have been used effect~vely. The tools that are used to mach~ne plastin should be kept sharp at all tunes is best done by means of straight-flute drills or by "dubbing" the cuttlng edge of a twist drill to produce a zeto rake angle.These configurations are shown m Figur
FIGURE 14-10 Straight-flute

drill (left) and "dubbed" dr~ll used for drllling plastics. (r~ght)

SE(:TI~N 14.2

Fabrication of Plastics

343

Rotary files,saws, and milling cutters should be run at high speeds to improve cooling but with the feed carefully adjusted to avoid clogging the cutter. Laser machining may be an attractive alternative to mechanical cutting. Because it vaporizes the material instead of forming chips; precise cuts can be achieved. Minute holes can be drilled, such as those in the nozzles of aerosol cans. Abrasive materials, such as filled and laminated plastics, can be machined in a manner that also eliminates the fine machining dust that is often considered to be a health hazard.

FINISHING AND ASSEMBLY OPERATIONS Polymeric materials frequently offer the possibility of integral color, and the as-formed surface is often adequate for final use. Some of the finishing processes that can be applied to plastics include printing, hot stamping, vacuum metallizing, electroplating, and painting. Chapter 36 presents the processes of surface finishing and surface engineering. Thermoplastic polymers can often be joined by heating the relevant surfaces or regions.The joining heat can be provided by a stream of hot gases, applied through a tool like a soldering iron, or generated by ultrasonic vibrations.The welding techniques that are applied to plastics are presented in Chapter 34. Adhesive bonding, another popular means of joining plastic, is presented in Chapter 35. Because of the low modulus of elasticity, plastics can also be easily flexcd, and snap-fits are another popular means of assembling plastic components. Because of the softness of some polymeric materials, self-tapping screws can also be used. DESIGNING FOR FABRICATION The primary objective of any manulacturing activity is the production of satisfactory components or products, and this involves the selection of an appropriate material or materials. When polymcrs are selected as the material of construction, it is usually as a result of one or more of their somewhat unique properties, which include light weight, corrosion resistance, good thermal and electrical insulation, ease of fabrication, and the possibility of integral color. While these properties are indeed attractive, one should also be aware of the more common limitations, such as softening or burning at elevated temperatures, poor dimensional stability, and the deterioration of properties with age. The basic properties and characteristics of polymeric materials have been described in Chapter 8. One should note, however, that property evaluation tests are conducted under specific test conditions. While a standard tensile test may show a polymer to have a moderately high strength value, a reduction in loading rate by two or three orders of magnitude may reduce this strength by as much as 80%. Conversely, an increase in loading rate can double or triple tensile strength. Polymers are often speed-sensitive materials.The can also be extremely sensitive to changes in tcmperature. Strength values can vary by a factor o l 1 0 over a temperature range of as little as 200F (100C). Materials should be selected with full consideration given to the specific conditions of temperature, loads and load rates, and operating environments that will be encountered. A second area of manufacturing concern is selecting the process or processes to be used in producing the shape and establishing the desired properties. Each of the wide variety of fabrication processes has distinct advantages and limitations, and efforts should be made to utilize their unique features. Once a process has been selected, the production of quality products further requires an awareness o l all of the various aspects of that process. For example, consider a molding process in which a liquid or semifluid polymer is introduced into a mold cavity and allowed to harden. The proper amount of material must be ii~troduced and caused to flow in such a way as to completely fill the cavity. Air that originally occupied the cavity needs to be vented and removed. Shrinkage will occur during solidification andlor cooling and may not occur in a uniform manner. Heat transfer must be provided to control the cooling andlor solidification. Finally, a means must be provided for part removal or ejection from the mold. Surface finish and appearance, the resultant engineering properties, and the ultimate cost of production are all dependent on good design and proper execution of thc molding process. Product properties can be significantly affected by such factors as melt temperature. direction of flow, pressure during molding, thermal degradation, and cooling rate.

In all molded products,it is important to provide adequate fillets between adj sectlons to ensure smooth flow of the plastic into all sections of the mold and toe corners.These fillets also make the mold nate stress concentrations at sharp ~nterior expensive to produce and reduce the danger of mold fracture durmg use. Even th tenor edges should be rounded where possible. A rad~us of 0.25 to 0.40 mm (0.0 0.015 tn.) is scarcely nohceable but will do much to prevent an edge from c h ~ Sharp corners should also be avoided in products that w~ll be used tor electrical cations, slnce they tend to increase voltage gradients, which can lead to product fa Wall or section th~ckness is also very important,since the hardening or curlng of a polymer 1s detemlned by the thickest sectlon. If possible, sectrons should be nearly umform in fhlckness, stnce nonunifom~ty can lead to serious warpage an mensional control problems As a general rule, one should use the minimum thlc that will provide satisfactory end-use performance The speclficvalue will be deter primarily by the slze of the part and, to some extent, the procers and the type of p being used. Recommended minimum thicknesses for molded plastics are as follo Small parts Average-sized parts Large parts 125 mm (0.050 in.) 2 15 mm (0.085 in.) 3 20 mm (0 125 in.)

Thick corners should also be avo~ded because they can lead to gas pocke dercur~ng, or cracking. When extra strength is needed in a corner. ~tcan usually b vided by incorporating rlbs into the destgn. Econom~cal production n also fac~hlated by appropriate dimensional toler A minimum tolerance ot 0.08 mm (0.003 m.) should be allowed in dtrections that ar allel to the parting line of a mold or contamed w~thin a mold segment. In direction cross a partlng surface, a minimum tolerance of 0 25 mm (0.010 in.) IS desirable. cases, increasing these values by about 50% can s~multaneously reduce manufac difficulty a? well as cost. Since most molds are reusable, careful attention should also be given tot moval of the part. R~gid metal molds should be des~gned so that they can be opened and c1osed.A small amount of unidirectional taper should be provlded t ltate part withdrawal. Undercuts should be avolded whenever posstble,since th prevent part removal unless additional mold sectlons are used.These must mov to the costs of mold producti pendently 01the major segments of the mold,add~ng maintenance, and slowing the rate ot product~on

INSERTS
Metal mserts, usually of brass or steel, are often incorporated into plastic p provide enhanced performance or unlque features. S~nce molded threads are produce,machtned threads require add~tional processing, and both types tend to deform, thrcaded metal inserts ale frequently used when assemblies requrre consi strength or when frequent disassembly and reassembly are antic~pated. Flgur depicts one form of threaded insert, along with other types that provide pins or ho alignment or mountmg. The successful use of inserts requlres careful attention to design since generally held in place by only a mechanical bond that must reslst both rotati pullout. Knurling or grooving IS often required to provlde su~table sites for gr A medium or coarse knurl is usually adequate to resist torsional loads and modera forces. C~rcumferential grooves are excellent for axial loads but offer little resis torsional rotation. Axial grooves resist rotation but do httle to prevent pullou

I,
FIGURE 14-11 Typlcai

Inserts used to provide cav~ties, holes, and al~g plns in plastic parts.

lash fi h e th

SECTIDN 14.2 Fabr~cat~on of Plastics

4-12 Various ways of metal inserts in plasttc o rbght): bending,

345

~otching, swaging, r head, and grwves iers. Knurling a

means of nnchori~ig includc hzndins. \plittin<.notching. and swaging. Ilci~ded pans with nulicir<ul:~r huads u13! be 115dd as f o r m d . ('~mhinatic>nc u t nutcl~~ @rcJu\rt.s,a~ld ~s. shuul-

14-12 depicts some common means of znsest attachment. d a s are also common. F~gure If an insert is to act as a boss for mounting or serve as an electrical terminal, it should protrude slightly above the surface of the plastic.This permits a firm connection to be made without creating an axial load that would tend to pull theinsert from its surround~ngs. If the insert serves to hold two matingparts together, it should be flush witb the surface. In this way, the parts can be held together snugly w~ihout danger of loosenmg the insert. In a U cases, the wall thickness of the surrounding plastic must be sufficient tu support any load that may be transmitted through the insert For small inserts, the wall thickness should be at least hall the diameter of the Insert. For insertslarger than 13 mm (4 in.) in dimeter, the wall thickness should he at least 6.5 mm ({ in.).

DESlGN FACTORS RELATED TO FINISHING Because plastics are frequently used where consumer acceptance is or great importance, special attenhon should be given to finish and appearance In many cases, plastic parts can be des~gned to require very little finishmg or decoratwe treatment. For small parts, suitable abiasives flus and rough spots can often be removed by a barrel tumbling t ~ i t h or polishing agents. Smoofh~ng and polishing occur in the s a ~ operation. e By etching the surfaces nf a mold, dewrations or letters can be produced that protrude approximately 0.01 mm (0.004 in.) above the sudace of the plastic.When higher relief 1s required, the mold can be engraved, but t h s adds significantly to mold cost. Whenever possible, depressed letters or designs should be avoided fiese features, when transferred to the mold become raised above the surrounding surface. Mold making amount of intricate machlnmg as the surrounding materithen requires a aons~derable al must be cut away from the dwign or letters When recessed features are absolutely required, manufacturing cost can be reduced if they can be incorporatedlnto a small area that is raised above the primary surface. When designing plastic parts, a prime objective is often the elimination of seeondary machining, especially on surfaces that would be exposed to the customer. Even when fillers are used cas they are in most plastics), the surfaces of molded parts have a thin film of pure resin.TE.8 film provides the high luster that is characteristic of polymeric products. Machining cuts through the surface, exposmg the underlying filler.The result IS a poor appearance, as well as a slte for.the absorption of moisture One location that frequently requires machining is the parting line that is produced where the mold segments come together. Since perfect mating is difficult to achieve, a small fin, or "flash," is usually p r o d u d around the part penmeter, as iUmhated ~nFigure 14-13, When the flash is trimmed off, the resultmg line of exposed Mler may be objectionable. By locating the parting line along a sharp corner, it is easier to maintain satisfactory mating of the mold sections, and the eetrposed filler that is created by flash removal will be confined to a corner, where it is less noticeable.
Polymer-conta~nlng
filler
Llne of

xposed
14-13 Trimming the
i

a plastlc part ruptures ayer of pure resln along xg hne and creates a posed filler.

i~ne

4 i 3 !
Flash
Thm pure layer resln of

ffller

346

Cwm~n 14 Fabrica tlon of Plastics, Ceramics, and Composites

and should be avolded whene figure 14-14, can be used to to reveal flow marks from t handling or service. Extern

FIGURE 14-14 Stiffness

can be Imparted to large surfacesof plastic parts through the use of r~bb~ng or dorn~ng

consideration. Dunng th these plns can be subjec at one end, the length s duced filling pressures, t excessive problems. Holes that are to countersunk.This not only assists In starting the tap or screw but also reduces c at the outer edge of the hole. If the threaded hole is less than 6.5 mrn In ) in ter, it is best to cut the threads after mold~ng, using some form of thread tap. For eters greater than 6.5 mm, the threads can be molded or an Insert should be used. threads are molded, however, special provisions must be made to remove the part the mold. Since the additional operations extend the moLding time and reduce pr t~vity, they are generally considered to be uneconomical

(i

)F

RUBBER AND ELASTOMERS

Rubber and elastomeric products can be produced by a varlety of fabrication px es. Relatively thin parts wlth uniform wall khickness, such as boots. gloves, and fa are often made by som (usually based on natu dry. W ~ t h each dip, a c dips being used to produce a final deslred thickness After vulcamzat~on,usually in the products are stripped from the molds The dipping process can be accel charge 1 s introduced to the latex par charge, either through an applied voltage or by a coagulant coating that release tive ions when dipped into the soluti site charges causes the elastomericpart~cles to be depos~ted on the form at a fas and in thicker layers than the basic process. With electrostatic depos~tion, man ucts can be made m a single immersion. When the parts are thicker or complex-shaped solids, the first step is poundmg of elasto ments.This IS usually done in some a homogeneous mass. Adaptations frequently used to produce the d molding are used, along with speci also be directly cast to shape. Rubber compounds can be made into sheets uslng calenders, llke that s Figure 14-15.The sheet coming from the calender is oflen rolled w~th a fabric prevent the material from sticking. Three- or four-roll calenders can also be place a rubber or elastomer covering over cord or woven fabric. In the three-ro etry, only one side of the fabric i shown m hgure 14-16, enables Products such as Inner t produced by the extruszon process.The compounded elastomer is forced throu by a screw device similar to that dexribed for plast~cs.

Rubber

Rubber sheet

3 roll calender

Uncoated fabric

R
4-roll calender

Rubber

Coated

Rubber or artificial elastomers can be bonded f o megal, such as bsass or steel, using a variety of polymeric adhesives, Only moderate pressures and temperatures are required to obtain excellent adhesion.

CERGMICS
The fabrication processes applied to ceramic materials generally fall into two distinct classes, based on the properties of the material, Classes can be manufactured into useful artlcles by Eirst heating the material to produce a molten or viscous state, shaping the material by means of viscosls f k w , and then cooling the material to produce a =lid product. Crystalline ceramics havs a characteristically brittle behavior and are normally manufactured into useful components by pressing moist aggregates or powder into a shape, followed by drying, and then bonding by one of a vmiety of mechanisms, which include chemical reaction, vitrificakon (cementing with a liquefied material), and sintenizg (solid-state diffusim).

FABRlCATlQN TECHNIQUES FOR GLASSES

Glass is generally shaped at elevated temperatures, where the viscosity can be controlled. A numhw of the processes begin with material m the liquid or molten condition. Sheet and plate glass is formed by processes such as extsuding through a narrow slit, rolling through water-cooled roll% or floating on a bath of molten tin. Glass shapes can be produced by pouring the molten material directIy into a mold. The cooling rate is then controlled, usually as slow as passible, to minim= res~dual stresses and the tendency for cracking. Constant-cross-section shapes can be made by extrusion, and glass fibers can be produced by forcing liquid &ass through multiple openings in an extrusion die.

348

CHAMER 14

Ceramics, and Composites Fabric ation of Plast~cs, Other glass-forming processes begin with viscous maases and use matlug and female dle members to press the material into the desired shape, as illustra 14-17. Bottles, containers, and shapes like incandescent hght bulbs are ma F~gure a process similar to the blow molding of plastics. Cupahaped pleces of VISCOUS mat are expanded agalnst the outside of heated dles, as illustrated in Figure 14-18 Special heat treatments can also be applied to glass material By applying fa coohug to the exposed surfaces, a residual stress pattern of surface compression be induced. The surface layers cool and contract, and the softer intenor flo conform. This 1s followed by the coohng of the intenor, whlch tries to contract restra~ned by the already cold surface, creatlng the surface compression.The res product, called tempered glass, 1s stronger and more fracture resistant, since cr tend to 1111tlateon free surfaces. When unfavorable residual stresses are present might lead to crackmg, annealtng treatments can be used to reduce then magnltu Glaly-ceramlcs form a unlque class of materials that are part crystalline and glass They are fabricaled Into shape a5 a glass and are then subjected to a special treatment (devlrr~fication)that controls thc nucleation and growth of the c component. Because of the dual structure, the final properties Include good and toughness. along w ~ t h low thermal expansion. Typical products include co (such as the wh~te CorningWare products), ceramlc stove tops, and materials us electrical and computer components

FIGURE 14-17 Viscous glass can be easily shaped by

FABRICATION OF CRYSTALLINE CERAMICS

Crystall~ne ceramics are hard, briltle materials wth hlgh melbng polnts As a resul cannot be formed by techniques requiring either plastlc~ty(1.2, forming metho mclt~ng(i.e , casting methods). Instead, theae materials are generally processed solld statc by techniques that utllke partlcles or aggregates and resemble those u powder metallurgy The partlcles can also be blended with additives that plasticity or flow and enable the forming or casting processes to be used. Dry powders can be compacted and converted Into useful shapes by pressi either environmental or elevated temperatures. Dry preouzg wlth ngid tooling, presstng, and hot-c,ostutrc pressing (HIP) with flexible molds are common tec and exh~hit teatnres and hmitations sim~lar to those discussed ln Chapter 18. Clay productr are based on special types of ceramics blended with wate various additive5 to produce a material that can be shaped by most of the tradi forming methods. Plavtl~ forming can also bc applled to other ceramics if the ce particles are comblned wlth additives that impart plast~citywhen subject pressure and heat. Wetpressing can be used to produce shapes at loaer pressure dry presslng. Extnuion can be used to produce products \nth constant cross sect1 Inlect~onmolding was discussed earlier in t h s chapter as a means of fa plastics, and melal injection molding (MIM) is presented in Chapter 19 as a producing small, complex-shaped metal parls. A torm of injection moldlug can used to form complex, three-d~mensionalshapes from ceramic materials. powder is rmxed wlth polymer material, and heated material (125'-I 250"-300F) is then Injected Into an aluminum &e under pressures on the or 30-100 MPa (5-15 ks~). The mixture cools, and after about 30 seconds, it i enough Lo perm& ejection from the d ~ e Additwe . materials are then remo

mat~ng male and female d ~ e members.

Parison
\

, Air

FIGURE 14-18 Th~n-walled glass shapes can be produced of presslng by a comblnat~on and blow mold~ng.

thermal, solvent, catalyt~c, or wick~ngmethods. and the remaining ceramic is fused together by a firing operation. As with metal injection molding, the die forms a part that is considerably oversized, and controlled shrinkage dunng fir~ng produces the final d~mensions The major dimensions of most parts are less than 10 cm (4 in.); the m.), and tolerances are on the order of 1% or wall thickness 1s less than 6 mm 0 1mm (0.005 in.), whichever is greater. Most parts are made from the oxide ceramics, such as alumina or zirconla, but the process has also been used with s~licon carbide and silicon nitride. Several castlng processes can be used to produce ceramic shapes, beginning wlth a pourable slurry that strengthens by partial removal of the liquid or the gellation, polymerizat~on, or crystallization of a matrul phase. In the slip castzrlg process, ceram~c powder is mixed with a liquid to form a slurry,which is then cast mto a mold contaln~ng very fine pores Capillary action pulls the liqu~d from the slurry, allowing the ceramic particles to arrange into a "green" body with suffinent strength for sub~equent handling. Pressure applied to the slurry, vacuum applied to the mold. or centrifugal pressure can all a d in liquid removal Hollow shapes can be produced by pounng out the remaining slurry once a desired thickness of sohd has formed on the mold walls Shp casting has been used to produce a vanety of porcelain products, including bathroom Gxtures, f ~ nchlna e and dmerware, and ceramic productr for the chemical industry. In the tape tastzng process, a controlled film of slurry is formed on a substrat~ Evaporation of the liquid during controlled drying produces a thin. flexible, rubbery tape or sheet that has smooth surfaces and uniform thickness. These products are widely used in the multilayer construction of electron~c circu~ts and capacitors. In other casting-type processes, slurries contain~ng bondmg agents can be used to produce cast-in-place products, such as furnace linings or dental fillings. When mixed with a sticky binder, the materlal can be blown through a plpe to apply ceramic coatings or build up refractory Limngs The nu~nerous variations of sol-gel procebsmg can be used to produce ceramic films and coatings, fibers, and bulk shapes These processeq begn with a solubon or colloidal dispersion ( ~ o l )which , undergoes a inolecular polymerization to produce a gel, whlch IS then dried. This approach offers higher purity and homogene~ty, lower firing temperatures, and finer grain size at the expensc of higher raw-material cost, large volume shnnkages during processing, and longer processing times. Table 14-1 summarizes some of the primary processes used to fabricate shapes Erom crystalline ceramics.

(a

PRODUCING STRENGTH IN PARTICULATE CERAMICS Each of the processes just described can bc used to produce useful shapes from ceramic materials but useful strength generally requires a subsequent heating operation, known a$ &in8 or ~ m t e r i n g Slurry-type . materials must first be dr~ed in a manner that is designed to control dimens~onalchanges and minlmize stresses, distortion, and cracking.The mater~al is then heated to temperatures between 0.5 and 0.8 tlmes the absolute melting point. where diffusion processes act to fuse the

Starting material
D r y powder
Dry powdw
Slurry Ceramic-plaslic blend Ccctmic-binder blend Clay, uraler, and additives

Advantages
Low cost: can h e auromaied Uniform density: variable cmss seainn%; can he automated Large sires; complex shapes; low Luoiing cost Complex crass sections; fast: can be aulomaled; high volume Low cost;variahle shapes (SU&as long lengths) Easily shaped by forming methods; wide range of size and shape

Limitations
Limited cross sections; density gradients Long cyclc times; small numher of products per cycle Long cycle limes: lahor-intensive Binder must bc rcmovcd: high tool cost Blnder must he removed partrclec orcented by flow Requlrcs controlled drylng

Fabrication of Plastics, Cerarn~cs,and Composites particles together and lmpart the desired mechanical and physical properbes. temperature and time are selected to control the resulting grain size, pore size slnterin pore shape. In some firing operations, surface melting (lzqu~d-phase component reactions (reactzon sintenng) can produce a substantial amount of material (vitnfiultion).The liquid then flows to produce a glassy bond hetwee ceramic particles and either solidifies as a glass or crystallues. Cementfition is an alternative method of producing strength that doe requue elevated temperature. A liquid binder mater~al1s used to coat the ce particles, and a subsequent chemical reaction converts the liquid to a solid,for strong, ngid bonds. Prototypes or small production quantities of ceramic products have been by the laser siatenng of ceramic powders. Successive layers of material are together by the laser sintering (or laser melting) of thin layers of heat-fuslble po For ceramic parts, the powder particles are actually coated w ~ t ha ve thermoplastic polymer binder. The laser then acts on the polymer coating to p the bond.After the laser bonding, the parts then undergo conventional deb'mdt sintering to about 55 to 65% of theoretical density. Isostatic pressing pnor to sint can ralse the final density to 90 to 99% of ideal,

MACHlNlNG OF CERAMICS Most ceramic materials are brittle, and the techniques used to cut metal generally produce uncontrolIed or catastrophic cracks. In addition, ceramic typically hard materials, Since ceramics are often used a9 abrasives or coatin cutting tools, the tools needed to cut them have to be even harder. D~rectproduction to the desired final shape is clearly the most attr alternative, but there are times when a material removal operation is necess inachinlng can be performed before or after the final firing. Before firing, the tends to be rather weak and fragile. While fracture is always a concern, a significant consideration might be the dimensional changes that will occur subsequent firmg. Shrinkage may be as much as 3046, so it may be difficult to a or maintain close final tolerances. For this reason, machming before firing, kno green rnachmmg, is usually rough machining designed to reduce the amo finishing that will be required after firing. When machining is perfoxmed after firing, the processes are generally o might conslder to be nonconventional. Grindmg, lapping, and pohshing with di abrasives, drilling with diamond-tipped tooling. cutting with diamond saws, nltr machining, laser and electron-beam machining, water-jet machining, and ch etchimg have all been ubed. When mechanical forces are applied, material sup qulte critical (since ceramic materials are almost always brittle). Because hardness of the ceramic, the tools must be quite rigid. Selection and use of co are also importmt issues. Materials producers have developed "machinable" ceramics that themselves to precision shaping by more traditional machining operations. It be noted, however, that these are indeed special matenals and not characteri ceramlcs as a whole. JOININGOF CERAMICS When we conslder joining operations, the unique properties of ceramics on introduce fabrrcation limitations. Brittle ceramics cannot be joined by fusion w deformation bonding, and threaded assemblies should be avoided whenever Therefore, most joining uhlizes some form of adhesive bonding, brazing, bondmg, or special cements Even with these methods, the stresses that devel surfaces can lead to premature failure. As a result, most ceramic products are to be monolith~c (single-piece) structures rather than multipart assemblies.

DESIGN OF CERAMIC COMPONENTS Since ceramics are brittle materials, special care should be taken to minimize bending and tensile loading as well as design stress raisers. Sharp corners and edges should be avoided where possible. Outside corners should be chamfered to reduce the possibility of edge chips. Inside corners should have fillets of sufficient radius to minimize crack initiation. Undercuts are difficult to produce and should be avoided. Specifications should generally use the largest possible tolerances, since these can often bc met with products in the as-fired condition. Extremely precise dimensions usually require hand grinding, and costs can escalate significantly. In addition, consideration should be given to surface-finish requirements, since grinding, polishing, and lapping operations can increase production cost substantially.

As shown in Chapter 8, composite materials can be designed to offer a number of attractive properties. In some market areas, such as aerospace and sporting goods, their acceptance and growth have been phenomenal. Use can only occur, however, if the material can be produced in useful shapes at an acceptable cost and rate of production. Many of the manufacturing processes designed for composites are slow, and some require considerable amounts of hand labor.There is often a large degree of variability between nominally identical products, and inspeclion and quality conlrol methods are not as well developed as for other materials. While these limitations may be acceptable for certain applications, they often restrict the use of composites for high-volume, mass-produced items. Faster production speeds, increased use of automation, reduced variability, and integrated quality control continue to be important issues in the expanded use of composite materials. In Chapter 8, composite materials were classified by their basic geometry as particulate, laminar, and fiber-reinforced. Since the fabrication processes are often unique to a specific type of composite, they will also be grouped in the same manner.

FABRICATION OF PARTICULATE COMPOSITES Particulate composites usually consist of discrete particles dispersed in a ductile, fractureresistant polymer or metal matrix. Their fabrication, however, rarely requires processes unique to composite materials Instead, the particles are simply dispersed in the matrix by introduction into a liquid melt or slurry, or by blending the various components as solids, using powder metallurgy methods. Subsequent processing generally follows the conventional methods of casting or forming, or utilizes the various techniques of powder metallurgy. These processes have been presented elsewhere in the text and will not be repeated here. Reinforcement particles have been successfully blended into the highly viscous slurries of rheocast material, the semisolid mixtures that are viscous when agitated but retain their shape when static. Particle reinforcements have also been produced by spray forming multicomponent feeds. FABRICATION OF LAMINAR COMPOSITES Laminar composites include coatings and protective surfaces, claddings, bimetallics, laminates, and a host of other materials.Their production generally involves processes designed to form a high-quality bond between distinct layers of different materials. When the layers are metallic, as in claddings and bimetallics, the con~posites can be produced by hot or cold roll bonding. Sheets of the various materials are passed simultaneously through the rolls of a conventional rolling mill. If the amount of deformation is great enough, surface oxides and contaminants are broken up and dispersed, metal-to-metal contact is established, and the two surfaces become joined by a solid-state b0nd.U.S. coinage is a common example of a roll-bonded material.

the foam, corrugated, or honeycomb filler.

FABRICATION OF FIBER-REINFORCED COMPOSITES In the fiber-reinforced composites, the matrix and fiber reinforcement prov system that offers properties not attainable by the individual components alone. The fiber reinforcement produces a significant increase in streng stiffness, while the matrix functions as a binder, transfers the stresses, and p protection against abrasion and environmental effects.

FIGURE 14-19 Fabrication of a honeycomb sandwich structure using adhesive bonding to jo facing sheets to the lightweight honeycomb filler. (Courtesy of A S M International. Metals Pork, 0

SECTION 14.5 Fabrication of Composite Mater~als

353

Some of the fiber-reinforced processes are identical to those previously described for unreinforced plastics: compression, transfer and injection molding, exlrusion, rotational molding, and thermoforming. Others are standard processes with simple modification$ such as reinforced reaction injection molding and resin transfer molding. Still others are specific to fiber-reinforced composites, such as hand lay-up, spray-up, vacuum-bag, pressure-bag, and autoclave molding; filament winding; and pultrusion.
Production of Reinforcing Fibers. A number of processes have been developed to produce the various types of reinforcement fibers used in composites. Metallic fibers, glass fibers, and many polymeric fibers (including the popular Kevlar) are produced by variations of conventional wire drawing and extrusion. Boron, carbon, and ceramic fibers such as silicon carbide are too brittle to be produced by the deformation methods. Boron fibers are produced by chemical vapor deposition around a tungsten filament. Carbon (graphite) fibers can be made by carbonizing (decomposing) an organicmaterial that is more easily formed to the fiber shape. The individual fine filaments are often bundled into yarns (twisted assemblies of filaments), tows (untwisted assemblies of fibers), and rovings (unlwisted assemblies of yams or tows). Fibers can also be chopped into short lengths, usually 12 mm in.) or less, for incorporation into the various sheet-or bulk-molding compounds. In these materials the fibers usually assume a random orientation.

(1

Processes Designed to Combine Fibers and a Matrix. A variety of processes have been developed to combine the fiber and the matrix into a unified material suitable for further processing. If the matrix material can be liquefied and the temperature is not harmful to the fibers, casting-type processes can be an attractive means of coating the reinforcement. The pouring of concrete around a steel reinforcing rod is a crude example of this method. In the case of the high-tech, fiber-reinforced plastics and metals, the liquid can be introduced between the fibers by means of capillary action, vacuum infiltration, or pressure casting. In a modification of centrifugal casting, resin is introduced into the center of a rotating mold and is then uniformly forced against and into the reinforcing material. Yet another alternative is to draw the fibers through a bath of molten material and combine them into aligned bundles before the liquid solidifies. Prepregs, or pre-impregnated reinforcements. are sheets of unidirectional fibers or woven iabric that have been infiltrated with a matrix material. Mats are sheets of nonwoven, randomly oriented fibers in a matrix. When the matrix is a polymeric material, the resin in the prepreg or mat is usually only partially cured. Later fabrication then involves the stacking of layers and the application of heat and pressure to further cure the resin and bond the layers into a continuous solid matrix. Prepreg layers can be stacked in various orientations to provide various directional properties. Individual filaments can be coated with a matrix material by drawing through a molten bath, plasma spraying, vapor deposition, electrodeposition, or other techniques. The coated fibers can then be used, either individually or in various assemblies. They can also he wound around a mandrel with a specified spacing and then cut to produce tapes that contain continuous, unidirectionally aligned filaments. These tapes are generally 1fiber diameter in thickness and can be up to 1.2 m (48 in.) wide. When the temperalures of the molten matrix become objectionable or potentially damaging to the fiber, matrix-fiber bonding can often be achieved through diffusion or deformation bonding (hot pressing or rolling). A common arrangement is to position aligned or woven fibers between sheets of foil material. Loosely woven fibers can also be infiltrated with a particulate matrix, which is then compacted at high pressures and sintered to form a continuous solid. Sheer-molding compounds (SMC) are composed of chopped fibers (usually glass in lengths of 25 to 50 mm or 1 to 2 in.) and partially cured thermoset resin, along with fillers, pigments, catalysts, thickeners, and other additives, in sheets approximately 2.5 to 5 mm (0.1 to 0.2 in.) thick. With strengths in the range of 35 to 70 MPa (5 to 10 ksi)

and the ab111ty to be press formed in heated dies, these materials offer a feas alternative lo sheet metal in applications where light weight, corrosion resistance, integral color are attractwe Ieatures. Atter Initial compounding and a few days of curing, sheet-molding cotnpo generally take on the consistency of leather, making them easy to handle and a When they are placed in a heated mold, the v~scosityis quickly ieduced and

orient the reinforcing fibers, making the final orientation nonrandom and difficu predict and control. Bulk-molding compaunds (BMC) are fiber-reinforced, thermoset, mol materials, where short fibers (6 to 12 mm or to in.) are d~stributedm ran

used to shape the matrix material (usually metals or polymers). Others are uni the fanlily of fiber-reinforced composrtes. The donunant techniques wlll be dis individually in the sections that follow

PULTRPJ~ION Pultrusion is a continuous process that is used to produce long len of relatively s~mple ~hapes wlth uniform cross section, such as round, rectan tubular, plate, sheet, and structural products. As shown in Figure 14-20, bund continnous reinforcing fibers are drawn through a bath of thermoset polymer r

Cross sections can be as much as 1.5m (60 in.) wide and 0.3 in (12 m.) thick.

helical patterns, or a combination of these, designed to take advantage of their

part and mandrel are placed in an oven for curing, after wh~ch the product 1s s
Re~nforclng

f~hers

FIGURE 14-20 Schemat~c d~agram of the pultrus~on

process. The heated dies cure the thermoset restn

, large tank

ment
of

Roti r Inc.,

from the form.The matrix, often an epoxy-type polymer, binds the structure together and transmits the stresses to the fibers. Figure 14-21 shows a large tank being produced by filament winding. Products such as pressure tanks and rocket motor casings can be made in vi~tuauyany size, some as large as 4.5 m (15 ft) in diameter and M m 165 ft) long. Smaller parts include helicopter rotor blades, baseball bats, and light poles. Moderate production quantities are feasible, and because the process can be highly mechanized,uniform quality can be maintained. A new form block is all that is required to produce a new size or design. Because the toollng is so mexpensive, the process offers tremendous potential for cost savings and flexibility. With advancements in computer software, equipment, and control, parts no longer need t o be axisymmetric. Filament-wound products can now be made with changing surfaces, nonsymmetric cross sections, and compound curvatures.

LAMINATION AND LAMINATION-TYPE PROCESSES In the lamination process, prepregs, mars, or tapes are stacked to produce a desired thickness and cured under pressure and heat. The resulting products possess unusually high strength as a result of the integral fiber reinforcement. Becaue the surface is a thin layer of pure resin, laminates usually possess a smooth, attractwe appearance. If the resin is transparent, the fiber material is visible and can impart a variety of decorative effects. Other decorative laminates use a separate patterned face sheet [hat is bonded to the laminate structure. Laminsred materials can be produced as sheets, tubes, and rods. Fiat sheets can he made using the method illustrated in Figure 14-22. Prepreg sheets or reinforccme~tsheets saturated m resin are stacked and then compressed under pressures on the order of 7 MPa (1000 psi). Figure 14-23 depicts the technique used to produce cnds or tubes. For tubing, the impregnated stock is wound around a mandrel of the desired internal diameter. Solid rods are made by using a small-diameter mandrel, which i s removed prior to curing, or by wrapping the material tightly about ~tself. Sheet laminating can also be a continuous process in which multiple reinforcement sheets are passed through a resin bath and &en through squeeze rolls
Paper or cloth sheet

Drying oven

Cutter

R E 14-22 Method o f

I W

Resin dip

cing multiple sheets inated plastic material.

356

CHAPTER 14

Fabrication of Plastics, Ceramics, and Composites

FIGURE 14-23 Method o f producing lam~nated plastic operation, tub~ngIn thef~nal the rolled tubes are cured by be~ng held in heated tool~ng

hardwood, or even particle board. The layers of prepreg or resin-dipped fabric stacked in various orientations until the desired thickness is obtained. Carc must

flexible hag, and the contained air is evacuated. Air pressure then elimin entrapped air. expels excess resin, and holds the laminate against the mold while resin is cured. While curing may occur at room temperature. moderately elevat temperatures may also he used. In pressure-hag molding, a flexible membran' positioned over the female mold cavity and is pressurized lo force the individual p together and drive out entrapped air and excess resin. Pressures usually range fr

deflectors for tractor-trailers. and body panels for trucks.

preformed mat is placed in the press, and heat and pressure are applied. Temperat

labor-intensive lay-up processes and the faster compression molding or inje molding processes, which generally require more expensive tooling. Continuous mat or woven material (usually employing glass fiber) is positioned dry in the ho half of a matching mold, which is then closed and clamped. A low-viscosity cataly resin is then injected into the mold. where it displaces the air, permeates

14-24 Aerodynamic nd smooth surfaces

ize the hood and

's AernMax truck.

,Metals Park, O H )
reinforcement, and subsequently
CUES

at low temperatures. Because of the low

presses can be used to clamp the mold segments, and inflatable bags can be used to produce simple holes or hollows, in much the same way that cnres are used in conxcentional casting. The resulling products can have excellent surfaces on both sides, since both moId surfaces can beprecoaled wlth a pigmented gel. Large parts can often be made a s a smgle unit with a relatively low capital investment Cycle time$ range from a few mmutes to a few hours, dependn~g on the part size and the resin system being used. The aerodynamic hood and fender assembly for Ford Motor Company's AeroMax heavy-duty truck (Figure 14-24) is an example of a large resin-transfer molding. When the quality demands are not as great, thereinforcement-to-resin ratio is not exceptionally high, and only one surface needs to be finished to high quality, the pressing operations can often be elimmated. In a process known as hand lay-up or open-moldprocessirzg, depicted in figure 14-25,sucoessive layers of pliable resin-coated cloth are simply placed in an open mold or draped over a fomSqueegees or rollers are used to manually ensure good contact and remove any entrapped air, and the assembly ts then allowed to cure, generally at room temperature. If prepreg layers ate not used, a layer of mat, cloth, or woven roving can be put in place and a layer of resin brushed, sprayed, or poured on.This process can then be repeated to build the desired thickness. While the hand lay-up process is slow and labor intensive and has part-to-part and operator-to-operator variab~lity, the toohng costs are sufficiently low that single items or small quantities became economcally feasible. Molds or forms can be made from wood, plaster. plastics, aluminum, or steel, so design changes and the associated tool modifications are rather inexpensive, and manufacturing lead time can be quite short. In add~tion, large parts can be produced as single units, significantly reducing

glass layers

Resin and

Mold

teleas

Mold
FIGURE 14-25 Schematic of the hand lay-up lam~nation

process.

the amount of assembly, and various types of re~nforcementcan be incorporated a single product, expanding design options High-quahty ~urfaces can be product apply~ng a pigmented gel coat to the mold before the lay-up.

SPRAY MOLDINGWhen continuous or woven fibers are not requlred to produc desired properties, sheet-type parts can be produced by mixing chopped fibers,fi and catalyzed resin and spraying the combrnat~on into or onto a mold form, as st m Figure 14-26. Rollers or squeegees can be used to remove entrapped au and I the resin into the reinforcement. Room-temperature curing n usually preferred elevated temperatures are sometimes used to accelerate the cure. As with the lay-up process, a 1 1 Initial gel w a t can be used to produce a smooth, pigmented sur SHEET STAMP~G Thermoplastic sheets that have been reinforced with nonwl fiber can often be heated and press formed in a manner slmilar to convent~onalr metal forming. Precut blanks are heated and placed between the halves of a mat metal mold that is mounted in a vertical press. Ribs, bosses, and contours c a formed In parts with essentially uniform thickness Cycle times range from 1 50 seconds for mo5t parts.
The injection molding of fiber-reinforced plastlcs is a prt that wmpetes with metal die castlngs and offers camparable properhe cons~derablyreduced weight. In its s~mplest form, chopped or continuous fiben placed in a mold cavity that 1s then closed and injected with resin. An imprf are premxed; method utilizes chopped fibers, up to 6 mm in ) in length, wh~ch the heated thermoplast~c(often nylon) prlor to mjection. Another variation u{ feedstock oI discrete pellets that have been manufactured by slicing continuousi pultruded rods The benefits of addlng fiber remforcement (compared to conven? plastic molding) include rncreased rigidity and impact strength, reduced possib'ili brittlc failure during impact, better d~mensional stability at elevated temperature! in humid environments, improved abrasion resistance, and better surface fimsh d! the reduced dimendonal w~ltract~on and absence of related sink marks. The m procrss is qulte rapid, and the final parts can be both precise and complex.
~JECTION MOLDING

(a

BRAIDING, THREE-DIMENSIONAL KNITTING, AND THREE-DI~NSIONAL WEAVING primary causes of failure in lamination-type composites arc interlammar crackin delamination (layer separat~on) upon impact. To overcome these prsblems, the strengfh reinforcing fibers can also be mterwoven Into three-dimensional prefor processes that include weavmg, brardmg, and stitching through the stacked two-dlmenslonal preforms. Resin 1s then Injected into the assembly, a resultant product is cured for use. Complex shapes can be produced, with th orientations selectcd for optimum properties. Computers can be used to desig control the weaving, making the process less expenslve than many of the more mtensive techniques.
Gel coat Chopperispray gun

1
I

Resin

FIGURE 1426 Schematic dlagram of the spray formlng of

chopped-flber-reinforced
fibers

polymeric composite.

SGCTI~N 14.5 Fabrication of Composite Materials

359

Fabrication of Fiber-Reinforced Metal-Matrix Composites. Continuous-fiber metal-matrix composites can be produced by variations of l'iiament winding, extrusion, and pultrusion. Fiber-reinforced sheets can b e produced by electroplating, plasma spray deposition coating, or vapor deposition of metal onto a fabric or mesh. These sheets are then shaped and bonded, often by some form of hot pressing. Diffusion bonding of foil-fabric sandwiches, roll bonding, and coextrusion are other means of producing fiber-reinforced metal products. Various casting processes have been adapted to place liquid metal around the fibers by means of capilla~y action, gravity, pressure (die casting and squeeze casting), or vacuum (countergravity casting). Products that incorporate discontinuous fibers can also he produced by powder metallurgy or spray-forming techniques and further fabricated by hot pressing, superplastic forming, forging, or some types of casting. In general, efforts are madc to reduce or eliminate the need for finish machining, which would often require the use of diamond or carbide tools, or methods such as electrodischarge machining (EDM). A critical concern with metal-matrix composites is the possibility of reactions between the reinforcement and the matrix during processing at the high temperatures required to melt and form metals as well as the temperaturcs of subsequent service. These concerns often limit the kinds of materials that can be combined. Barrier coatings have been employed to isolate reactive fibers. In terms of properties, graphite-reinforced aluminum has been shown to be twice as stiff as steel and one-third to one-fourth the weight, with practically zero thermal expansion. Aluminum reinforced with silicon carbide exhihits increased strength (tension, compression, and shear at both room and elevated temperature) as well as increased hardness, fatigue strength, and elastic modulus. Thermal creep and thennal expansion are both reduced, but ductility, thermal conductivity, and electrical conductivity are also decreased. Magnesium, copper, and titanium alloys, as well as the superalloys, have also been used as the matrix in fiber-reinforced metal-matrix composites. Fabrlcation of Fiber-Reinforced Ceramic-Matrix Composites. Unlike polymericor metal-matrix composites, where failures originate in or along the reinforcement fibers, ceramic-matrix composites often fail due to Raws in the matrix. If the reinforcement is bonded strongly to the matrix, a matrix crack might propagate right through the fibers. To impart toughness to the assembly, it is often desirable to promote a weak bond between the fiber and matrix. Cracks are redirected along the Fiber-matrix interface rather than through the fiber and the remaining matrix. The matrix materials and reinforcement fibers for ceramic-ceramic composites have been discussed in Chapter S, along with some of the unique property combinations that can be achieved. Fabrication techniques are often quite different from the other composite families. One approach is to pass the fibers or mats through a slurry mixture that contains the matrix material. The impregnated material is then dried, assembled, and fired. Other tecbniqucs include the chemical vapor deposition or chemical vapor infiltration of a coated fiber base, where the coating serves to weaken an otherwise strong bond. Silicon nitride matrices can be formed by reaction bonding.The reinforcing fibers are dispersed in silicon powder, which is then reacted with nitrogen. Hot-pressing techniques can also be used with the various ceramic matrices. When the matrix is a glass, the heated material behavesmuch like a polymer, and the processing methods are often similar to those used for polymer-matrix composites. Secondary Processing and Finishing of Fiber-Reinforced Composites. The various fiber-reinforced composites can often be processed further with conventional equipment (sawed, drilled, routed, tapped, threaded, turned, milled, sanded, and sheared), but special considerations should be exercised because composites are not uniform materials. Cutting some materials may be like cutting multilayer cloth, and precautions should be used to prevent the formation of splinters and cracks as well as frayed or delaminated edges. Sharp tools, high speeds, and low feeds are generally required. Cutting debris should be removed quickly to prevent the cutters from becoming clogged.

360

CHAPTER 14

Fabrication of Plastics, Ceramics, and Composites

In addition, many of the reinforcing fibers are extremely abrasive dull most conventional cutting tools. Diamond or p o may be required to achieve realistic tool life. Abrasive slurrys can conjunction with rigid tooling to assure the production of smooth surfaces. Lasers water jets are alternative cutting tools. Lasers, however, can burn o r c material o r produce undesirable heat-affected zones. Water jets can cr problems with some plastic resins, and pressurized water can cause dela t h e low heat and light cutting force are attractive characteristics. Elastic defl are minimized during the cut. Parts can often be held in place by simpl and water jets also minimize the generation of dust, which may b e toxic. When fiber-reinforced materials must be joined, t continuity of the fibers in the joint area. Thermoplastics can be softe by applying pressure with heated tools, combining pressure and ultr o r using pressure and induction heating. Thermoset materials generally requ use of mechanical joints o r adhesives, with each method having its charac advantages and limitations. Metal-matrix composites are often brazed.

II Key Words
adhesive bonding annealing auloclave blow molding braiding bulk molding compourlu compression molding calendering casting cementation ceramics ceramic-matrix composites clay products cold molding composites compression molding crystalline ceramics devitrification dipping dry pressing elastomer explosive bonding extrusion fibers filament winding firing foam molding glass glass ceramic hand lay-up hot-isostatic pressing injection molding inserts isostatic pressing lamination laser sintering matrix mats metal-matrix composites open-mold processing parison plastics prepregs pressure-bag molding pultrusion reaction injection molding resin-lransfer molding roll bonding rotational molding rovings sandwich structures sheet-moldmg compound sintering slip casting sol-gel processing spray molding tempered glass thermoforming~ tows transfer molding vacuum-bag molding viscous flow vitrification wet processing

IReview Questions
1. Why are the fabrication processes applied to plastics, ceramics, and composites often different from those applied to metals? What are some of the key differences? 2. How does the fabrication of a thermoplastic polymer differ from the processing of a thermosetting polymer? 3. What are some of the ways that plastic sheet, plate, and tubing can be cast? 4. Why do cast plastic resins typically have a lustrous appearance? 5. What types of polymers are most commonly blow molded? 6. Why do blow molding molds typically contain a cooling system? 7. For what types of parts and production volumes would compression molding be an appropriate process? 8. What are typical mold temperatures for compression molding? What is the most common mold material? 9. What are some of the attractive features of the transfer molding process?
10. Cold molding is faster and types of molding. What limits its use? 11. What is the most widely used process for Lhe fabric

12. In what ways is injection molding of plastic similar to casting of metal? 13. What is the benefit of a hot-runner distribution sy plastic injection molding? 14. thermoplastics? 15. How are the individual components mixed in the injection molding process? 16. What are some of the attractive consequences of temperatures and low pressures of the reaction molding process?

Problems What are some of the typical production shapes that are produced by the extrusion of plastics? For what types of materials and products might thermoforming beconsidered attractive? 'What types of products are produced by rotational molding? 42. What are glass-ceramics? How are they produced?

361

43. What are some of the techniques that can be used to impart

some degree of plasticity to crystalline ceramic materials? 44. Describe the differences between the injection mold~ng of plastics and the injection molding of ceramics.

What are some typical applications for rigid-type foamed What type of products are produced by the spinningprocess'l What are some of the general properties of plastics that affecttheir machinability? What are some of the attractive features of laser machining ,What property of plastics is responsible for making snap-fit assembly a popular alternative for plastic products'? . What are somc of the attractive properties of plastics that favor their selection? What are some o l the common What are some of the design concerns when specifying and setting up a plastic molding process? Why should adequate fillets be included between adjacent sections of a mold? Whal is a major benefit of rounding exterior corners? Why is it most desirable lo have uniform wall thickness in

. What are some of the ways in which metal inserts are held in

the processing of crystalline ceramic products? 47. How does cementation difier from sintering? 48. What are the bcnefits and limitations of machining ceramic materials before firing versus alier firing? 49. What are some of the nonconventional methods used to machine ceramics? 50. Why are joining operations usually avoided when fabricating products from ceramic materials? 51. Discuss some of the design guidelines that relate to the production of parts from ceramic material. 52. Why are the processes used to fabricate particulate composites essentially the same as those used for conventional material? 53. What are some of the processes that can bc used to produce a high-quality bond between the layers of a laminar composite? 54. List several fabrication processes for iiber-reinforced products that are essentially the same as for unreiuforced plastics. List several that are unique to reinforced materials. 55. What types of materials are used as reinforcing fibers in fiber-reinforced composites? 56. What are some of the forms in which rcinforcemcnt fibers appear in composite materials? 57. What is a prepreg? 58. What are sheet-molding compounds (SMCs)? Bulk-molding compounds (BMCs)?

winding? Bat feature lcss noticeable? h a t is the benefit of countersinking holes that are to be %readed or used for self-tapping screws? b a t types of products can be produced from elastomeric &aterials using the dipping process? used to shape products from laminated sheets of woven fibers? 62. What are the benefits of using an autoclave instead of roomtemperature and low-pressure curing? 63. What form of reinforcing fibers can be incorporated in the spray-molding process? Infection molding? reinforcement? 65. Describe some of the ways in which a metal matrix can be introduced into a fiber-reinforced composite.

at method is generally used to bond elastomers to other

composites. fiber-reinforced composites?

I I Problems
fabricated from composite materials, such as skis,
ntify composite materials that are currently being used d the companion shape-producing fabrication methods. ratchet wrench and a pipe wrench. These components are traditionally forged from ferrous alloy or made from a cast steel or cast iron. For various reasons, alternative materials may be desired.The ratchet wrench is quite long,and reduced weight may be a reasonable desire. Both of these tools could

he used in areas, such as a gas leak, where a nonsparking safety tool would be required. Current specifications for the ratchet handle call for a yield strength in excess of 50 ksi and a minimum o f 2% elongation in all directions to ensure prcvcntion of brittle fracture.The pipe wrench most likely has similar requirements.

a. Could a plastic or composite material be used tom quality product with these additional properties? (N Metal jaw inserts can be used in the pipe wrench, en the other components to be considered as separate pi b. If so, how would you propose to manufacture th handles?

JNDAMENTALS OF
RMlNG PROCESSES. EPENDENT VARIABLES
EPENDENT VARIABLES

METAL FORMING

FPENDENT-DEPENDEM

15.6 GENERAL PARAME~ERS Warm Formrng 15.7 FRICTION AND LUBRLCATLON Isothermal Forming UNDER METALWO~ING Case Study: REPAIRS TO A CONDI~ONS DAMAGED PROPELLER 15.8 TEMPERATURE CONCERNS Hot Working Cold Work~ng

INTRODUCTION
Chapters 11 through 14 have already presented a vanety of methods for producing a des~red shape from an engineering material. Each of those methods had its characterrsUc qct of capabilitlea, advantages, and limitatsons. If we are to select the best method to make a given product, however, we must have a reasonable understanding of the entue spectrum of available techniques for shape product~on and the~r related features. The next several chapters will further our study of shape production methods by considering the family of deformntronprocerses.These processes have been deslgned to explo~l a remarkable property of some engineering materials (most notably metals) w~thout deterioration of their properties. known as plnstrclty, the ability to flow as sol~ds Smce all processing 1s done in the sohd state, there is no need to handle molten matesla1 or deal with the complexitie~ oi solidification Slnce Lhe material is simply moved (or rearranged) to produce the shape. as opposed to cutting away unwanted reglons, the amount of waste can be substantially reduced. Unfortunately, the forces required are often h~gh. Machinery and toohng can be quste expensive, and large productson quant~ties may be necessary to justrfy the approach The overall usefulness of metals is largely due to the ease of fabrication into useful shapes. Nearly all metal products undergo metal deformation at some stage of their manufacture. By rolling, cast ingots. strands, and slabs are reduced in size and converted Into basic forms such as sheets, rods, and plates.These forms can then undergo further deformation to produce wire or ihe myriad of fni~shed products formed by processes such as forging, extrusion,sheet metal formmg, and others The deformation may be hu1k~'Iow m three d~mens~ons, s~mple shearing, simple 01 compound bendtng, or complex combinations of these. The stresses producing thcse deformalions can be tension, compression, shear, or any of the other varletses included in Table 1 5 . 1 . Table 15-2 dep~cts a wide variety of specific processes and identifies the primary state of stress responsible for the deformation. For most of these processes. a wide range of speeds, temperatures, tolerances, surface finishes, and amounts of deformation are poss~ble.

Slrnple untaxlal tens~on

Biaxiai tsnslon

Ti~axialtension Blax~al tension, compression

Biaxial tenslon and compress~on

B~ax~ai cornpresslon

B ~ a x ~eompression, al

tenslon

Tl~axial oompress~on

Pureshear

Simple shear with B~axial shear triaxlal compression triaxial compr

Forming procesres tend to be complex syrtems consisting of ~ndependent va dependent variables, and ~ndependent-dependentintenelations.Independent va are those aspects of the process over which the engineer or operator has direct and they are generally selected or specified when setting up a process. Consld of the independent variables in a typical forming process:
1. Starting material. When specifying the starting material, we may define not chem~stry of that material but also its condition. In so doing, we define th properties and characteristics. These may be chosen enlirely for ease of fahr o r they may be restricted by the desire to achieve the required final properties completion of the deformation process

siderations ofteri influence thii decision. 3. Tool or diegeomerry.Th~s 1s an area of major signiftcance and has many aspe as the diameter and profile of a rolling mill roll, the bend radius in a sheetoperation, the die angle tn wire drawing o r extrusion, and the cavity deta forging. Since the tooling will induce and control themetal flow as the maten from starting shape to finished prodnct,success or failure of a process often d on tool geometry. 4. Lubrication. It is not uncommon for friction between the tool and the work account for more than 50% of the power supplied to a deformation process. cants can also act as coolants,themal harriers, corrosion inhlbitoss,and pa pounds. Hence, theu selection is an important aspect in the succesa of operation.Spcification tneludes type of lubricant, amount to be applied, an of application.

5. Stsrtrng mperatllve. Since material properties can vary greatly with temp
temperature selection and control are often key to the success or failure ofa forming operation. Specification of starting temperatures may include the te tures of bath the workpiece and the tooling. 6. Speed of operation, Most deformation processing equipment can be opera range of speeds. Since speed ran directly influence the forces required fo tion (see Figure 2-32), the lubricant effectiveness, and the time avaiiabl transfer, its selection jffects far more than the production rate.

7. Amount of deformation. While someprocesses control this variable throu slgn of tooling, others,such as rolhng, may pernut its adjustment at the dn the operator.

Process
Roil~ng

Schematic Diagram

State of Stress in Main Part D u n n g Eormngn

7

Forging

Shear \plnnfng

Tubs splnnmg

Deep drawang

In flange o t blank, 5 In wall of cup, I

Wlre nnd tube drawins

Strmght bcnd~ng

At bend, 2 and 7

Contoured flanging

la) Convex

At outer flnnge, 6 At bend, 2 and 7

(a) Concave

'Numbersewrespond

LO those

m parentheses i n Table 15 1.

@ -

At outer. flange, 1 At bend. 2 and 7

After speclficatlon of the independent vanables, the process m turn determines th ture and values of a second set of features. Known as dependent vanables, the'

L Force or power [email protected] certain amount of force or power is required to vert a selected material Gom a starting shape to a final shape, with a specified 1 cant, tooling geometry, speed, and starting temperature A change in any o

the mdependent variables and then experience the consequences of that select It 1s extremely important, however, that we be able to pred~ct the forces or ers that w ~ lbe l required for any form~ng operation. Without a reasonable estimr forces or oower,we would be unable to specifv the equipment for the process,^

2. Materralproperties

of the product. W h l e we can easily specify the properti the starting material, the combined effects of deformation and the tempera experienced durmg forming will certa~nly change them. The starting properti the materlal may be of interest to the manufacturer, but the customer is far I concerned with receiving the des 11 is important to know, therefore, how the inliial properties will be alteiedb shape-produnng process. 3. Exct (or fmul) temperature. Deformation generates heat within the material workpieces cool when ~ncontact with colder toollng. Lubricants can break do7 decompose when overheated or may react with the workplece The properties engmeering matenal can be altered by both the mechan~cal and thermal aspei a deformation process.Therefore, if we are to control a process and produce qt products, it is Important to know and control the temperature of the material thrc out the deformation. (Nute:The fact that temperature may vary from loeatii location w~thm the product further adds to the complexity of this variable.) 4. Surface finlsh and precision. The surface finish and dimens~onal precislonq resultant product depend on the speafic details of the formtng process 5. Nature of the materialflow. In deformation processes, dies and tool~ng generally' forces or pressures and control the movement of the external surfaces of the i piece. While the objective of an operation 13the production of adesired shape,tl ternal flow of mater~al may actually be of equal importance. As w~ll be showd In t h ~ s chapter, producr propestles can he significantly affected by the details oj tertal flow, and that flow depends on all the deta~ls of a process. Customer satis? requlres not only the production of a desired geometric shape but also that the6 possess the nght set of compan~on propert~es,w~thout any surface or tnternaldi

-DEPENDENT RELATIONSHIPS
Fleure 15-1serves to illustrate the major problem facing metalform~ngpersonneLQ

which control cr entrrrly indrrect. Unfortunately, it is the dependent variables want to control, but their values are determined by the process, as complex of the Independent variable selection. If we want to change a must determme which indeoendent vartable (or combination of independent V i vartable-dependent vanable interrelanum. Understanding the links between Independent and dependent variables i , the most important area of knowledge tor a person In metalforming. Unfortunate1

SECTION 1 5 . 5 Independent variables Srarftng material Starting geometry Taol geometry Lubricatton Starting tempersture speedof &formatlon Amount o f def6rrnatlon

Process Modeling

367

Links
-Experrence-Enperlment-Modeling-

Dependent variables Force or power

requirements

Schernat~c )f a metalforrntng ~ndependent dent variables, means of linking

Product properttes Exit temperature Surface f ~ n r s h

D'menslonai

Metenal flow details

knowledge is often difficult to obtain. Metalfonning processes are complex systems composed of the material being deformed, the tooling performing the deformation, lubrication at surfaces and interfaces, and various other process parameters such as temperature and speed. The number of different forming processes (and variations thereof) is quite large. In addition, different materials often behave differently in the same procer$ and there are multitudes of avalable lubricants.Some processes are sufficiently complex that they may have 15 or more interacting independent varrables. of independent and dependent We cangain information on the interdependenc~es variables in three distinct ways:

1 . Experience. Unfortunately, this generally requires long-time exposure to a process and is often limited to the specific materials, equipment, and products encountered during past contact. Youngcr employee% may not have the experience necessary to solve production problems. Moreaver, a single change in an area such as material, temperature, speed, or lubti~ant may make the bulk of past experience irrelevant. 2, Experitnent. While possibly the least likely to be in error, direct experiment can be both time consuming and costly: Size and speed of deformation are often reduced when conducting laboratory studies. Unfortunately, lubricant performance and heat transfer behave differently at afferent speeds and sizes, and their effects are generally altered.The most valid experiment, therefore, 1s one conducted under full-size and ful-speed production condition-generally too costly to consider to any great degree. While laboratory experiments can provide valuable msight, caution should be exercised when extrapolating Pab-scale results t o more realistic production conditions. 3. Process enodeling. Hwe one approaches the proqess through high-speed computing and one or more mathematical models. Numerical values are selected for thevarious independent variables, and the models are used to compute predictions for the dependent outcomes. Most techniques rely on the applied theory of plasticity with various simplifyingassmptions Alternatives w. aryfrom crude, first-order approximations to sophisticated, computer-based methods, such as finite element analysis. Various models may incorporate strain hardenm~,thermal softening, heat transfer, and other phenomena. Solutions may be algebraic relations that describe the process and reveal trends and relations between the variables or simply numerical valuw based on the specific input features.

Metalforming simulations using the finite element modeling method became common m the 1980s but generally required high-power minicomputers or engineering workstations. By the mid-1990s, the rapid increase in computing power made it posSible t o model complex processes on desktop personal computers With the continued expansion of computing power and speed, process simulations are now quick, inexpensive, and quite accurate. As a result, modeling is being used in all areas of manufacturing, including part design, manufacturing process design, heat-treatment and surface-treatment optimization, and others. Models can predict how a material will respond to a rolling process, fill a forging die, flow through an extrusion die, or solidify in a casting.

Entire heat treatments can be simulated, including cool~ngrates in vanous quen Models can even predict the strain distribution, residual stresses, microstructur final properties at all locations within a product. Advanced simulat~on techniques can provide a clear and thorough unde ing of a process, ehrmnating costly tnal-and-error development cycles. Product des manufacturing methods can be optimized for quality andreliabitity,while reducln duchon costs and mtnimizing lead times. When coupled with appropriate sensor same models can be used to determine the type of adjustments needed to provld hne process control. Process models can also serve as laboratory tools to explor ideas or new products. New employees can become famil~ar with what works an doesn't in a quick and inexpensive manner. It is important to note, however, that the accuracy of any than that of the input variables For example, when modelin the mechantcal properties of the defonnittg matenal (i.e., must be known for the specific condition3 of temperatu formation), and straln rate (speed of deformation) belng considered.The m descriptions of material behavior as a function of the process conditions a censtitutlve relations. T h e development of such relationships is not a ever, becaube the same material may respond d~fferently to the s crostructure is different A 1040 steel that has been annealed ( not have the same properties as a quenched-and-tempered (tempered martens of the same cl~emistry. Microstructure and its effects on properties are difficult to in quantitative terms that can be input to a model. Another rather elusive varlable is the friction betw Studies have shown that fnctron depends on contact prem ish, lubncant, speed, and the mechanical properties of the know that these parameters often vary from location to loca time during a process, but many models tend to describe friction with a slngle of constant magnitude.Any variations with time and location are simply ignored of mathemancal simplicity or because of a lack of any better informatton. At first glance, problems such as those just discu barrier to the use of mathematical models. It should be n difficulties apply to the person trying to document, characteri sults of experience or expertments. Process modeling often r otherwise go unnoticed and can be qute useful when attempting to prevent ore defects, optinme performance, or extend a process into a prevrously unknown

,METERS
Whlle much metalformlng knowledge is speclfic to a given process there are features that are common to allprocesses, and these will be It is extremely important to characterize the material being defo strength or resstance to deformation at the relevant condrtions of temp deformat~on, and amount of prior strainmg?What are the formability li of anticipated fracture? What 1s the effect of temperature or vanattons what exTent does the material strain-harden?What are the recsystalluat material react with various envtronments or lubricants? These and must be answered to assess the sultabiity of a material to a given deformationp the properties of engineering materials vary widely, the details will not be pres trme.The reader 1s referred to the various chapters on engmeermg more m-depth references nted m Chapter 9 and the reference appendix. Another general parameter is thespeed of deformat~onand the vario effects. Some rate-sensitive materials may shatter or cra plast~cally when subjected to slow-speed loadings. Other mat when deformed at higher speeds. For these speed-seasztiv needed to produce the same result if we wish to do it fa~ter, an required. Mechanical data obtained from slow straln rates in tensile tests m ly useless if the deformahon process operates at a significantly greater

deformation. Speed sensitivity is also greatest when the materral is at elevated temperature, a condition that is frequently encountered in metalformmg operations. The selectlon of hammer or press tor the hot forging of a small product may well depend on the speed sensitivity of the material being forged In addition to the changes in mechanical properties, Caster deformation speeds tend to promote improved lubricant efficiency Faster speeds also reduce the time for heat transfer and cooling. During hot working, workpieces stay hotter and less heat is transferred to the tools Other general parameters mclude frzckon and lubncat~on and temperuture Both of these are of sufficient importance that they will be discussed in somedetail.

JG

LUBRICATION UNDER CONDITIONS

High forces or high pressures are applied through tools to lnduce the deformation of a matenal. Because of the relative motlon between the workpiece and the tool, an important consideration in metal deformation processes 1s the frict~o~l thal ex~sts at this interface. For some processes,more than 50% of the input energy is spent m overwmmg frichon.Changes in lubrication can alter the mode of material flow during fosming, create or eliminate defects,alter the surfacefinish and dimensionalprecision of The product, and modify product properties Production rates, tool deslgn, tool wear. and pt aces?optImlzation all depend on the ahihty to deternnne and control fr~ction between the tool and workpiece. In most cases, we want to economically reduce the effects of friction. However, some deformation processes, such as rolhng, can only operate when sufficrent friction 1s present. Regardless of the procesq Glction effects are hard to measure. As previously noted, the spec~fic friction condrtions depend on a number of variables, including contact area, contact pressure, surface finish, speed, lubricant, and temperature. Because of the many vanahles, the effects of friction are extremely d~fficult to scale down for laboratory testing, or extrapolate from laboratory tests to production conditions. It should be noted that friction under metalworking conditions is significantly different from the friction encountered in most mechanrcal devlcrs The frictlon cond~tions of gears, bearings, journals, and similar components generally involve (1) two surfaces of similar material and similar strength, (2) experiencrng elastic loads such that neither body undergoes permanent change in shape, (3) \sith wear-in cycles that produce surface compattb~lity, and(4) low to moderate operating temperatures. Metalformmg operatrons, on the other hand, involve a hard, nondeforming tool Interacting with a soft workp~ece at pressures sufficient to cause plastic flow in the weaker material. Only a s~ngle pass is involved as the tool and workpiece interact, the workpiece is often at elevated temperature, and the contact area ia frequently changing as the workpiece deforms. Figure 15-2 shows a typical relationship between frictional resistance and contact pressure.For hght,elastic loads,h c t ~ o is n directly proportional to the applied psessnre,mth

u
&

2 0
U

Slope = +

-2 The effea of rsure on the fr~ctlonal etween two surfaces.

Contact pressure. P

the proportionality constant, y,being known as the coefficrent offrrctron or, more sp lcally, the Coulomb coefficient of friction At high pressures,friction becomes ~ndepen of contact pressure and is more closely related to the strength of the weaker maten An understanding of these results can be obtained from modern fri theory, whoseprimary premise is that "flat surfaces are not flat" but have somed of roughness. When two irregular surfaces interact, sufficient contact is esta to support the applied load. At the lightest of loads, only three points of contact necesyary to support a plane. As thc load 1 s increased, the contacting points defor the contact area increases, initially m a Linear fashion As the load continues to inc more area comes into contact. Finally, at some high value of load, there n full co between the surfaces. Additional loads can nolonger bring additional areas Into co and friction can now be described by a constant, independent of pressure. Fricuon 1s the resistance to slidlng along an mterface. From a mechanistic point, this resistance can be attributed to (I)abrasion, the force necessary to plo peaks of a harder material through a softer one, andlor (2) udheszon,the Corce nec to rip apart microscopic weldments that form between the two materials Since the ment tears generally occur in the weaker of the two materials,it n reasonable to as that the resistance attributed to both features would be proportional to the s the weaker material and also to the actual area of metal-to-metal contact curve depicted in Figure 15-2 could also be v~ewed as a plot of actual contact area interface versus contact prcssurc. Unfortunately, Flgure 15-2. and the associated ry, applies only to unluhricated metal-to-metal contact.711e addition of a lubricant,a as any variation in its type or amount, can significantly alter the frictional respon Surface deterioration or wear is another phenomenon that is directly related tion Since the workpiece only interacts with the tooling dur~ng a s~ngle forming tion,any wear experienced by the workpiece is usually not objectionable Infact,a fresh-metal surface produced by wear w: often viewed as desirable Manufac proceses retain most or all of the anginal dull finish may be accused of substandard products. Wear on the tooling, however, is quite the reverse.Toolin pensive and it is expected to shape many workp~eces Tooling wear will general1 in change of workpiece dimensions Tolerance control will be lost, and at som the tools will have to be replaced. Other consequences of tool wear include Incr frictional resistance (increased required power and decreased process efficiency) surface finish on the product, and loss o i production during tool changes. Luhricatron is a key to success in many metalforming operations While lubr are generally selected for their abllity to reduce friction and suppress tool we ouclary comlderationb may include the ability to act as a thermal barrier, keep' in the workpiece and away from the tooling, the abllity to act as a coolant, re heat from the tools; and the ability to retard corrosion if left on the formed Other intluencing factors include ease of apphcation and retnoval; lack of toxici and flammabihty, reactivity or lack of reactivity with material surfaces: adaptabdit a useful range of pressure, temperature, and velocity: surface wetting characte cost; availability; and the ability to flow or thin and still function as a lubricant. cant selection is further comphcated by the fact that lubricant performance may with any change in the interface conditions.The exact response is often depend such factors as the finish of both surface$ the area of contact, the applied load, the the temperature, and the amount ot lubncant The ability to select an appropriate lubricant can be a critical factor in det ing whether a process IS successful or unsuccessful, efficient or inefficient. For ex if a lubncant layer can prevent mechanical contact between the tool and the wor (full-fluid or solid layer separation), the forces and power required may decre much as 30 to 40%, and tool wear becomes almost nonexistent. Considerab therefore, has been directed to the study of friction and lubricahon, a subject trrbology, as it applles to both general metalworking conditions and specifi forrmng processes A substantial information base has been developed that ca optimizing the use o E lubncants in metalworking.

SECCI~ 15.8 N Temperature Concerns

371

8 TEMPERATURE CONCERNS
In metalworking operations, workpiece temperature can be one of the most important process variables. The role of temperature m altering the properties of a material has been discussed in Chapter 2 In general, an increase in temperature brings about a decrease in strength, an increase in duct~lity, and a decrease in the rate of strain hardening--all effects that would tend to promote ease of deformation. Forming processes tend to be classified as hot workmg, cold working, or warm worklng based on both the temperature and the matenal being formed. In hot workmg, the deformation i5 performed under conditions of temperature and strain ratewhere recrystallization occurs simultaneously with the deformation. To achieve this, the temperature of deformation 18 usually in excess of 0.6 times themelung point of the material on an absolute temperature scale (Kelvin or Rankine). Cold working is deformation under condition5 where the recovery processes are not actlve. Here the workmg temperatures are usually less than 0.3 tlmes the workpiece melting temperature. Warm working is deformation under the conditions of transition (l.e., a working temperature between 0.3 and 0.6 tlmes the melting point).

HOT WORKING Hat workrag IS defined as the plastic deformation of metals at a temperature above the recrystalllzation temperature. It is important to note, however, that the recrystallization temperature varies greatly with different materials. Tin is near hot-working condit~ons at room temperature, steels require temperatures near 2000F, and tungsten does not enter the hot-working regime untll about 400O0F.Thus the term hot work~ng does not necessar~lycorrelate with high or elevated temperature, although such is usually the case. As shown in F~gures 2-30 and 2-31, elevated temperatures bnng about a decrease in the yield strength of a metal and an increase in ductility.At the temperatures of hot working, recrystallization eliminates the effects of strain hardening, so there is no significant increase m yield strength or hardness, or corresponding decrease in ductility. The true stress-true strain curve is essentially flat once we exceed the y~eld point, and deformation can be used to drastically alter the shape of a metal without fear of fracture and without the requirement of excessively high forces In addition, the elevated tempe~atures promote diffusion that can remove or reduce chemical inhomogene~ties,porcs can be welded shut or reduced in size during the deformation, and the metallurg~cal structure can often be altered through recrystallization to Improve the final properties. An added benefit is observed for steels, where hot working lnvolves the deformation of the weak, ducule, face-centered-cubic austenite structure, which then cools and transforms to the stronger body-centered-cubic ferrite or much stronger nonequilihr~um structures, such as martenslte. From a negative perspective, the high temperatures of hot work~ng may promote undesirable reactions between the metal and its surroundmgs. Tolerances are poorer due to thermal contractions, and warping or &stortion can occur due to nonuniform cooling. The metallurgical structure may also be nonuniform, since the final grain size depends on the amount of deformation, the temperature of the last deformahon1 recrystalliiation, the cooling history after the deformation, and other factor$ all of which may vafy throughout a workpiece. While recrystallization sets the minimum temperature for hot work~ng,the upper limit for hot working is usually determined by factors such as excess oxtdat~on, gram growth, or undesirable phase transformations. To keep the forming forces as low as possible and enable hot deformation to be performed for a reasonable amount of time, the starting temperature of the workplece is usually set at or near the highest temperature for hot working.
Structure and Properfy Modification by Hot Working. When metals solldlfy into the large sections that are typical of ingots or continuously cast slabs or strands, coarse structures tend to form with a certaln amount of chemical segregation The size of the

372

CHAPT~R 15 Fundamentals of Metal Forming

FIGURE 15-3 Cross sectlon of a 4-in.-diameter cast copper bar polished and etched to show the as-castgrain structure.

grainy is usually not undorm. and undesirable gram shapes can be quite commo as the columnar grans that have been revealed m Figure 15-3. Small gas caw bhiinkage porosity can also form during solidificat~on. If a ca7t mctal is rchcatcd without prior deformation, it w~ll simply expe grain growth and the accompanying deterioration in eng~neering propert1es.H er, if the metal experiences a sufficrent amount of deformation, the d~stortedst will be rapidly replaced by new strain-free grains. This recr~stallrzat~on nt lowed by either (1) grain growth, (2) addit~onal deformation and recry~tallizati (3) a drop in temperature that will terminate diffusion and "freeze m" the rec lized btruclure The structure in the h a 1 product is that formed by the last rec li~ation and the thermal history that follows. By replacing the initial structure , grains, it is poss~hle to produce new one consisting of f ~ n espherical-shaped crease not only in strength but also in ductil~ty and toughness-a somewhat uni enhancement of properties Engineering properties can also be improved through the reorienta ~nclus~ons or m p u n t y particles that are present in the metal. With normal and cooling, many impuril~es tend L o locate along grain boundary interfaces. are unfavorably or~ented or intersect surfaces, they can ~ n i t ~ aa te crack o propagation through a metal. When a metal is plastically deformed, the I tend to flow along with the base metal or fracture into rows of fragments st? that are aligned in the direction of working These nonmetallic impurities do crystallize w ~ t h the base metal but retain their distorted shape and orientati 15-4, and final product exhibits aflow structure, like the one shown in F~gure ties tend to e x h ~ bdirectional ~t vanatlon.Through proper design ol' the defor impurit~es can often be reoriented into a "crack-arrestor" configuration where peipendicular to the dlrectlon of crack propagation. The outer lobe of the in Figure 15-4, for example, has excellent fracture resistance since all flow parallel to the external surfaces. The i m p u r ~ t ~ e appear s as crack ~n~tiators propagators only at the top and bottom of the inner lobe, which hopefully stress or noncritical locat~ons. F~gure 15-5 ychematically compares a machined thread and a rolled thre threaded fastener. If the axial defects in the starting wire or rod are reoriented to allel to the thread profile, the rolled thread offers improved strength and fracture res

Flow structure of a hot-forged gear blank. Note how flow is parallel to all critical surfaces. (Courtesy of Bethlehem Steel Corporation, Bethlehem, PA.)
FIGURE 15-4

-.-

process Temperature Vuriations. The success or farlure of a hot defo~mation often depends on the abiity to control the temperatures withrn the workpiece. Over 90% of the energy imparted to a deforming workpiece will be converted into heat. If the deformation process is sufficiently rapid, the temperature of the workpiece may actually increase. More common, however, is the cooling of the workpiece in rts lower-temperature environment Heat is lost through the workpiece surfaces, wrth the majority of the loss occurring where the workpiece is in direct contact w ~ t h lowertemperature tooling. Nonumform temperatures are produced. and flow of the hotter, weaker, intenor may well result in cracking of the colder, less ductlle, surfaces. Thin sections cool faster than thick sections, and this may further complicdte the flow behav~or. To m i n m z e problems,it is desirable to keep the workpiece temperatures as uniform as poswble. Heated dies can reduce the rate of heat transfer, but die life tends to be comprom~sed. For example, dies are frequently heated to 325O to 450C (600' to 850F) when used in the hot forming of steel. Tolerances could be improved and contact times could be increased if the tool temperatures could be raised to 550" to 650C (1000" to 1200F),but tool Life drops so rapidly that these conditions become qulte unattractive. A final concern is the cool-down kom the temperatures of hot working Nonuniform coollngcan introducesign~ficant amounts of rescdualJtress in hot-worked products. Associated with these stresses may be warping or distortion, and possible cracking.

COLD WORKING
The plastrc deformatron of metals below the recrystalhzation temperature is known ah cold workmg Here, the deformatron 1s usually performed at room temperature, but mildly elevated temperatures may be used to provide increased ductility and reduced strength. From a manufacturing viewpomt, cold workrng has a number of distinct advantages, and the various cold-working processes have become quite prominent. Recent advances have expanded their capabilities. and a trend toward Increased cold working appears likely to contmue. When compared to hot working, the advantages of cold worklng Include the following:

1 . No heat~ng is required.
2. Better surface finish 1s obtained 3. Superior dimensional control is achieved since the tooling sets drmensions at room temperature. As a result, little, if any, secondary machining i~ requ~red. 4. Products possess better reproducibility and interchangeability 5. Strength,fat~gue, and wear properties are all improved through atraln hardenmg. 6. Directronal properties can be imparted. 7. Contamination problems are mlnlmized Some disadvantages a~sociated with cold-workrng processes mclude the following:

1. Higher forces are required to initiate and complete the deformation. 2. Heavier and more powerful equipment and stronger tooling are required. 3. Less ductility is available 4. Metal surfaces must be clean and scale-free 5. Intermedtate anneals may be required to cornpensale for the loss of d n c t ~ t y that
accompanies strain hardening. 6. The imparted directional properties may be detrimental. 7. Undesirable residual stresses may be produced. The strength levels induced by strain hardening are often cornp-drable to those produced by the strengthening heat treatments Even when the preciston and surface

finish of cold worhng are not required, it may be cheaper to produce a product working a less expensive alloy (achtevmg the strength by strain hardening) than treating parts that have been hot formed from a heat-treatable alloy. In addition, and more duct~le metals andan improved understanding of plast~c flow have done to reduce the difficulties often experienced during cold for most cold-working processes eliminate or minimize the production of wa and the need for subsequent machining-a significant feature w ~ t h today's e conservation and materials recycling. Because the cold-formingprocessesrequire pow tools or dtes, they are best su~ted for large-volume produc quantity of products can justify thecost of the equpment and has been devoted to developing and improving cold-forming machmery along w~th ods to enable these processes to be economically attractwe for modest product~ou lilies. By groupingproducts made from the same startlng mater~al and using qu~cktooling, cold-formlng processes can often be adapted to s manufacture.
Metal Properties and Cold Working. The suitab~hty of a metal for cold wo determined primarily by its tensile pmperhes, and these are a direct consequen metallurgical structure Cold working then alters that structure, thereby alterln slle properties of the resulting product. It 1s important for both the lnconnng going properties to be conridered when selecting metals that are to be prmes cold working. Figure 15-6 presents the true stress-true strain curves for both a low- and a carbon steel. Focusing on the low-carbon material, we note that plastic deformatio not occur unt~l the strain exceeds the strain associated with the elastic hmit,poin the stress-strain curve Plastic deformation then value x4,where the metal ruptures. From the viewpoint of c significant: (1) the magn~tude of the yleld-point stress, whtch d e t e n n e s the quired to inihate permanent deformation, and (2) the x, to x4,wbich ind~cates the amount of plastic defonnati ach~eved without fracture. If a considerable amoun rial like the low-carbon steel is more desirable than t duct~lity would be available and less force would b deformation.The curve on the nght, however, has (see Chapter 2 for discussion) If strain hardening i material would have a greater increase in strength fo In addition, the material on the right would be more and might be easier to machine (see Chapter 20). Spnngback is another cold-working phenomenon t aid of a stress-strain diagram. When a metal is deforme part of the result~ng deformation is elastic. For example, x, In Figure 15-6 and the load is removed, it will return to its original size and

E s +
2
GI

+ :
FIGURE 15-6 Use of true stress-true rtraln d~agrams to assess the suitability of two metals for cold worklng.

,/-y-:
a --.------.-+---..----..-,

Increase in tensile strength due to work hardening. The high-carbon steel will als have more springback

:1
I I I

I /

,
0 XI

8 I

X2 x3

x*

True straln Low-carbon steel

SECTION 15.8 Temperature Concerns

375

because all of the deformation is elastic. If, on the other hand, the metal is stretched by an amount x,,corresponding to point b on the stress-strain curve, the total strain is made up of two parts, a portion that is elastic and another that 1s plastlc.When the deforming load is removed, the stress relaxation will follow line bx2,and the final straln w1I1 only be x , . The decrease in strain, x, - x,, is known as elastic ~prtngback. In cold-working processes, sprlngback can be extremely important. If a desired size IS to be achieved,the deformation must be extended beyond that point by an amount equal to the springback. Slnce different materials have different elastlc moduli, the amount of springback from a given load will change from one material to another A subsmution m matenal, therefore, may well require adjustments in the forming process. Fortunately, sprlngback is a predictable phenomenon, and most difficulties can be prevented by proper desrgn procedures. Initial and Final Properties in a Cold-Working Process. The quallry of the startIng matenal IS often key to the success or failure ot a cold-working operation.To ohtam a good surface finish and maintain dimensional precision, the starting material must be clean and free of oxlde or scale that might cause abrasion and damage to the dies or rolls. Scale can be removed by plckhng, a process in whlch the metal 1s dlpped in acid and then washed. In addition, sheet metal and plate are somehmes given a light cold rolling prior to the major deformation.The rollrng operatlon not only assures uniform starting thickness but also produces a smooth starting surface. The light cold-rollingpass can also serve to remove theyleld-pomtphenomenon and the associated problems of nonuniform deformation and surface ~rregularit~es in the product. Figure 15-7 presents an expansion of theleft-hand regron of Figure2-6 or Figure 15-6, a stress-strain curve that is typical of many low-carbon steels.After loadmg to the upper y~eld pomt, the material exhlbits a ylrld-pornt runout wherein the material can strain up to several percent with no addit~onal force being required. Consider a plece of sheet metal that 1s to be formed mto an automotive body panel. lf a segment of that panel were to receive a total stretchless than the magn~tudeof the yreld-point runout,it would be induced by a stress equal to the yield-pomt streqs. Since the stress is constant in the runout reBon, the material is free to not deform at all, to deform the entire amount of the y~eld-point runout, or toselect some point in between. It is not uncommon for some regions to deform the entire amount and thim correspondingly,whlle adjacent regons resist deformation and retain the orlginal thickness The resulting ndges and valleys,shown in Figure IS-7,are referred to as Luders bands or stretcher stralns and are very difficult L o remove or conceal. By first cold rolling the material to a strain near or past the yleld-point runout, all subsequent forming occurs in a re@onwhere a well-defined strain corresponds to each value of stress If the body panel were shaped from pre-rolled material, the deformation and thmning would be un~form throughout the piece

50 40 V1

Upper yield point Lower y~eld polnt

t;
20 -

Yleid-point elongat~on or yield-point runoui

10
"

~~

4
X

5
10')

Strain (%orin.iln.

function of the amount of cold work (expressed in percent).

FIGURE 15-8 Mechanical properties of pure copper as a

Amount o f cold work (%) Figure IS-8shows how the mechanical properties of pure copper are affc cold working. Individual tensile tests were conducted on specimens that had enced progressively greater amounts of cold wot-k. As the graph shows, yield strer tensile strength increase with increased deformation. Hardness is not presentei graph but generally follows tensile strength. Since the ducttl~ty decreases,thE an cold working is generally limited by the onset of fracture, Reducnon m area wor a decline s~milar to elongation, as would ~lectricalconductivity and wrrosionru In order to maximize the amount of starting ductility,an asanealrng heat tri is often applied to ametal prior to cold working. If the required amount of defo exceeds the fracture limit, however, iniermediate anneal, may be performed tc ductility (set the amount of cold work hack to zero), thereby enabling further without the risk of fracture. If the desired final properties coincide with a given of cold work, the last anneal can he judleiously positioned in the deformation this way, the desired shape can be produced dong with the mechanical proper l l anne: acwmpany the amount of cold work imparted following that mneal. In a erations, care should be exercised to control the gram size of the resulting a Grain sizes that are too largeor too smallcan both be detrimental. Cold working, like hot worlung, also produces an amsotropic structure-on properties vary with direction. Here, the anisofropy is related to the distortec associa stmcture and is not sunply a h c t i o n of the nonmetallicinclusions~lso cold working is the generation of residualstresses While anisotropy and residu: es can be benef~cial, they can also he qutte harmful.Since they occur as a cons, of cold working, then effect on performance should always be consdered.

WARM MRMINC
Defonnatlm produced at temperaturesintermediateto hot rind cold forrmngi as warm formtng. Compared to cold forming, warm forming offers the advan reduced loads on the tooling and eguipment,increased material ductibty, and a reduction in the number of anneals due to a reduction in he amount of stmn ha The use of hlgher forming temperatures can often expand the range of matel geometries that can be formed by a given process or piece of equipment. Higt steels may h e formed without a spheroidization treatment. Compared to hot forming, the lower temperatures of warm working less scaling and decarburizatlon, and enable production of products with bette sional precisian and smoother surfaces. Finish machining is reduced and less is converted into scrap. Because of the finer structures and the presence o f sol:

ince. nent rtion $tore -king ount le. In that g OP:rial.

hardearng, the as-farmed properties may be a d e q u a t ~ for many applications, enabhng the eliminattonof final heat-treatment aperattons.The warm regime generally requim less energy than hot warkmg due t a the decreased energy in heating the workpiece (lower temprature), e n e m saved through h~gher precision (less mater~al being heated), and the possible elimtnazion of postforming heat treatments. Although the tools must ezert 25 to 60% higher forces, they last longer smce there is less thermal shock and thermal fatigue. When energy was cheap, metalfodng was usually conducted in either the hot- or cold-work- regimes, and warn workmg was largeiy tgnored Even today, material hehavior is Less well characterzed for the warm-working temperatures (the warm-working temperatures for steel are betwee11 550" and 800C or 3 000" and 1500F) Lubricants have not been a3 blly deweloped fathe warm-warktug temperatures and pressures, and dli: destgn technology k not as well established. Nevertheless, the pressures of energy and mat e n d oonservatioh conpled with the uther cited benefits, strongly favor the cantlnned developrsent of warm working. Cold forming is still the preferred method for fabricating small components, but wann fdrming is cansldereZl to be attractive for lager parts (up to about 10 lb) and steels with more than 0.35% carbon andlor high alloy content. Hot worktng and warn forming me usually applied to bulk formtng processes, llke sheet material, the surface-to-volume raho is sufficiently large forgrng and extrusion FOX that the workpiece can rapidly lose its heat As the major auto manufacturers seek to increase fuel effi~~enay, there has been signif~cant interest in aluminum sheet as a replacement tor steel. Unfortunately,the formab~lity of high-strength aluminum is much lower fhan thal of low-carbon steels of s~mllar converted ritrength. If the steel is ~lmply to aluminum, and Lhe design and tvoling rematn unchanged, fracture often occurs in the more heavily worked regions, If the material, dle, and blank holder are all heated to 2W0 to 3WT (400" to 57ToF),however,alummum sheet shows a signtficant mGrease in fonhabihty, and satisfactory par% can generally be produced.

ISOTHERMAL FORMING
F~gure 15-9 shows the reIatiomhip between y~eld strength (or forging pressure) and temperature for several enp;meering rnetalu.The 1020 and 4340 steels show a moderate Increase m strength w ~ t h decreasing temperature. In contrast, the strength of the titanium alkoy ( ~ p m cxcle8) and the A-286 nrckel-based superalloy (sohd circles) shows a much stronger vauattlun Wtthm the range of typtcal hot-workmg temperatures, coohng of as httle as 100C (200F) could result rn a doubling in strength. Dunng hot fanning, cooling surfaces surround a hotter intermr. Any vatlation in strength can result in n o n m form deformation and cracking of the less ductile surface. To successfully deform temperature-sensittve mater~als, deformatton may have to he performed under iso&emal (constant-temperature) condittous The dies or tooling

378

C H A ~ E 15 R

Fundamentals of Metal Forming must b e heated t o t h e s a m e temperature as the workpiece, sacr1ficmg d i e life f, uct quallty. D e f o r m a t ~ o n speeds must b e slowed s o that any h e a t generated b mation c a n be removed m a m a n n e r t h a t would malntain a u n ~ f o r m and c temperature. I n e r t atmospheres m a y b e required because of t h c long times at I temperature. Although such methods a r e tndeed costly, they are often the on1 of producing satisfactory products f r a m certainmaterials Because of t h e unlfo peratures a n d slow d e f o r m a t ~ o n speeds, isothermally formed components gl exhiblt close tolerances, low residual stresseq a n d fairly u n ~ f o r m metal flow

I I Key Words
abraslon adheslon anlsotropy annealing be~id~ng bulk flow coefflclent of fr~ctlon cold work~ng constitutive relatlon detormdt~on procesqes dependent variable dimensional precision elastic springback flow structure friction hot working independent variahle intermediate anneal isothermal forming lubncat~on Luders bands orlented structure piastlc~ty recrystalllzatlon resrdual stresses shear~ng speed sensitrv~ty spnngback strain hardening stretcher strains stringers surface finish tooling trihology warm working wear yield-point runout

II Review Questions
20. What type of informalion about the material heingd 1. What is plasticity? may be particularly significant to a metallormingeq 2. What are somc of the general assets of the metal deformation 21. How might a material's performance vary with chaq processes? Some general liabilities'! speed of deformation? 3. Why might large production quantities be necessary to justi22. Why is Giction such an important parameter in metal fy metal deformation as a means of manufacture? operations? 4. What is an independent variable in a metalloning process? 5. What is the significance of tool and die geometry in designing 23. Why are friction effects in metalworking difficult to sc, for laboratory testing or scale up from laboratory cr a succesful metalforming process'! to production conditions? 6 . Why is lubrication often a major concern in metalfonning? 7. What are some of the possible roles of a lubricant in addition 24. What are several ways in which the friction conditior metalworking differ from the friction conditions faun! to reducing friction? mechanical equipment? 8. What are some of the secondary effects that may occur when 25. According to modern friction theory,rictional resist the speed of a metalforming process is varied? be attributed'to what two physical phenomena? 9. What is a dependent variahle in a melalformingprocess? 10. Why is it important to be able to predict the Iorces or powers 26. Discuss the significance of wear in metallorming: (a) L ~ workpiece E and (b) wear on the tooling. required to perform specific lormingprocesses? 11. Why is it imporpant to know and control the thermal history 27. Lubricants are oMen sclected far properties in a4 their ability to reduce friction. What are some of thi of a mctal as it undergoes deformation? tional properties? 12. Why is,it often difficult to determine the specific relationships 28. What are some of the benefits that can be obtained hetween independent and dependent variables? separating a tool and workpiece by an intervening[? 13. What are the three distinct ways of determining thc interrebricant? lation of independent and dependent variables? 14. What features limit the value of laboratory experiments in 29. If the temperature of a malerial is increased, whatcb properties might occur that would promote Uleeasg modeling mctalforming processes? 1 mation? 15. What features have contributed to the expanded use of process 30. Define the various regimes of cold working,warm wdl modeling? hot working in terms of :hemelting point of thematei 16. What are some of thc uses or applications of process models? 17. What is a constitutive relation for an engineering material? formed. s k 18. What features may limit the accuracy of a mathematical 31. What is an acceptable definition of hot working? 1 temperature involved? ,, model? 19. What simplifying assumptions are often made regarding fric- 32. What are some,of the attractive manufacturing andi gical features of hot-working processes? tion between the tool and workpiece? 33. What are some of the negative aspects of hot worq

ow can hot working be used to improve the grain structure

e deformed grains recrystallize during hot working, how he process impart an oriented or flow structure (and dioually dependent properties)? yare heated dies or 1001s often employed in hot-working at generally restricts the upper temperature to which dies at is the primary cause of residual stresses in hot-worked mpared to hot working, what are some of the advantages of d-working processes? at are some of the disadvantages of cold-fonning processes? ow could cold working be used to reduce the cost of a modte to high strength product? w can the tensile test properties o f a metal be used to asits suitability for cold forming'!

43. Why is elastic springback an important consideration in coldforming processes? 44. What are L u d e ~ bands s or stretcher strains, and what causes them to form? How can they be eliminated? 45. What engineering properties are likely to decline during the cold working of a metal? 46. How can the selective placement of the final intermediate anneal be used to establish desired final propertiesin a cold-formed product? 47. Is the anisotropy induced by cold working an asset or a liability? What about the residual stresses? 48. What are some of the advantages of warm forming compared to cold forming? Compared to hot forming? 49. What material feature is considered to be the driving force For isothermal forming? 50. Why is isothermal forming considerably more expensive than conventional hot forming?

Problems
per is being reduced from a hot-rolled 318-in.-diameter to a final diameter of 0.100 in. by wire drawing through a ries of dies. The final wire should have a yield strength in ess of 50,000 psi and an elongation greater than 10%. Use ure 15-8 to determine a desirable amount of final w l d rk.Compute the placement of the last intermediate anneal that the final producl has both the desired size and the List and discuss the various economic factors that should be nsidered when evaluating a possible switchfrom wld formepeat pnrt a for a possible conversion from hot forming dvertisement for automobile spark plugs hascited thssuority of rolled threads over machined threads. Figue 15-5 ws such a comparison for hot forming, wherc the deforocess reorients flaws and defects without signifinging the structure and properties of the metal.The p u g threads, however, were formed by cold rolling. Du same benefits apply? Discuss the assets and liabilities of cold rolling of threads compared to thread formation by ter modeling of metal deformation processes is a powand extremely useful tool.At the same time, there are ral areas of limitation that can significaiitly compromise yen invalidate the final results. Consider each of the following areas of limitation, investigating what is currently being used o r what current aptions are available: a. A mathematical description of material behavior (a constitutive equation). In a l m ~ sall t cases, some simplification of actual flow behavior is assumed. For accurate modeling, flow behavior should be known and mathematically characterized as a function of strain,strain rate, and temperature. b. Interfacial friction between the tooling and the workpiece. How is this being modeled'? Does it consider the effects of surface finish, sliding velocity, interface temperature, and numerous other factors. As the process commenc~s,lubricants may thin or be wiped from surfaces, forces and pressures change, temperatures change, and surface roughness or texture is modified. Does the model reflect any of these changes? Some models assign a single value to friction over Lhe entire contact surface. This value may also remain constant throughout the entire operation. c. Assignment of boundary conditions. Often the mathematical solutions must conform to assigned features, such as defined motions or stresses at specified surEaces.The boundary conditions have a profou~td effect on the results that are calculated.Poor choices or choices made so as to facilitate easy analysis can often produce misleading or erroneous results.

380

CHAMER 15 Fundamentals of Metal Forming

Repairs to a Damaged Propeller

he propeller of a moderately large pleasure boar has been cast from a nickel-aluminum-bronzealloy that contalns 82% Gu, 9% Al, 4% NI, 4% Fe, and 1% Mn. It is approximately 13 lnches in diameter with three 10-pilch blades and has been designed for both fresh-water and 1 One of the blades has struck a rock and Is badly benf A replacement propeller is qulte expensive and cannot be obta~ned for several weeks An attractive alternative, therefore, may be to repair the existing plece. Would you recommend such a repair, and how would you prooeed? Can it simply be hammered back into shape? Would you recommend any additional processing, either before or after the I the rationale for your recommendat~or 2 Asecond propeller, identrcal to the one also been damaged by an impact.This the damage 1 s in the form of a crack at the blades, as shown in the figure. Sint not penetrate into the hub, it is proposc f welding or bi made using someform o Would you recommend such a repair? you suggest the repair be made? Expk for your and outline 1 should befollowed. Would there be an) quallty of petiofmancewtth the repalre,

16.5 FOROTNC

'

16.7 WIRE,ROB~AND TUBE DRAWING 16.8 C a w FORMING, COLD FORGING, AND IMPACT E~RUSLON
OTHER SQUEEZING PROCESSES Roll Extrusion Sizing Riveting Staking Coining Huhhing SURFACEIMPROVEMENT BY D E F O R M APROCESS~G ~N Case Skudy: HANDLE AM, BODY QF A LARGE RATCHET WRENCH

Rolling and Cold Roiling inqous (or Tandem)

1b.m ~esign'o~lm~ressioh-Die Por&gs m d ~ s s o & a t e d ' ~ ~ ) l i n g Upset F~rgiag ~utomalioHot r;crfging Wl'Forging Swaging Net-Shape and Rear-NetShape Forging 16.11
1 6 . 6 Ex~mfa~, Extrusion Mdfiods Metal Row in Extrusion ExmsFon oFHollow Shapes Hydrostatic Exmsian
Continuous Extrusion

n e s Control and Rolling rmomechanicd Processing d Controlled Rolling

6.1 INTRODUCTION
The shaping of metal by deformation is as old as recorded history.The Bible, in the fourth chapter of Genesis, introduces Tuhal-cain and cites his ability as a worker of metal. While we have no description of his equipmcnt, it is well established that metal forging was practiced long hcfore written records. Processes such as rolling and wire drawing were common in the Middle Ages and probahly date hack much further. In North America, by 1680 the Saugus Iron Works near Boston had an operating drop forge, rolling mill. ind slitting mill. Although the basic concepts oI many forming processes have remained largely unchanged throughout history, the details and equipment have evolved considerably. Manual processes were converted to machine processes during the Industrial Revolution.The macl~inery then became bigger, faster, and more powerful. Water wheel power was replaced by steam and then by electricity. More recently, computer-controlled, automated operations have become the norm.

.2 CLASSIFICATION OF DEFORMATION PROCESSES

A wide variety of processes have been developed to mechanically shape material, and a number of classification methods have been proposed. One approach divides the processes into primary and secondary. Primary processes reduce a cast material into intermediate shapes, such as slabs, plates, or billets. Secondary processes further convert these shapes into finished or semifinished products. Unfortunately.some processes clearly fit both categories, depending on the particular product heing made. In Chapter 15, we discussed the temperature of deformation and presented the various regimes based on the temperature of the workpiece.These included cold working, warm working, hot forming,and isothermal deformation.This classification has also become somewhat blurred, especially with the increased emphasis on energy conservation. Processes that were traditionally performed hot are now heing performed cold,

381

and cold-formlng processes can often be enhanced by some degree of heating. working has experienced considerable growth. Chaprers 16 and 17 utilize a dlvision that focuses on the size and shape of the plece and how that size and shape is changed. Bulk deformation processes are where the thicknesses or cross sectlons are reduced or shapes are sigmficantly ch Since the volume of the matenal remains constant, changes in one dimension requir portionate changes in others. Thus the enveloping surface area changes sign3 usually increasmg as the product lengthens or the shape becomes more complex.Th forming operations can be performed in all of the temperature regimes. Common pr es include: rolling; forging: extrusion; cold formmg; and wire, rod, and tube dra ' In contrast, sheet-forming operutions involve the deformation of amater~al the thickness and surface area remam relatively constant Common processes i shearing or blankmg, bending, and deep drawing.Because of the large surface-to-v ratlo, sheet mater~al tends to lose heat rapidly, and most sheet-forming oper are performed cold. Even this divlslon n not w~thout confus~on, however. Coining,for example, with sheet material but alters the thickness 1n a complex manner that n essentl deformation The bulk deformation processes w~ll be presented ~nChapter 17. forming processes can be found in Chapter 18.

11 16.3 BULKDEFORMATION PROCESSES

The bnlk deformation processes that nnll be presented in this chapter include:
1. Rolhng 2. Forging 3. Extrusion 4. Wire, rod, and tube drawing 5. Cold formmg, cold forging, and impact extrusion 6. P~ercmg 7. Other squeezing processes

These processes can be further dividedin several ways One grouping separa processes by focusing on the slze and shape of the deforming region. In continuo processes, such as forging, the size and shape are continually changing, and proces sis must reflect t h ~ change s In processes such as rolling or wire drawing, materi through the deforming region, but the size and shape of that region remaln unc Some form of steady-state analysis can often be applied. In all of the bnlk forming processes, the pnmary deformation stress 1s camp T h ~ may s be applied &ectly by tools or d ~ e that s squeeze the workpiece or m as in wlre drawing, where the workpiece is pulled in tension but the res~sting erates compression in the reglon undergomg deformation.

Rollmg operations reduce the thickness or change the cross section of a matenal compressive forces exerted by tolls. As shown in Figure 16-1, rolling is often t process that is used to convert mater~al into a finished wrought product. Th~ck stock can be rolled into blooms, billets, or slabs, or these shapes can be obtained from continuous castmg. A bloom has a square or rectangular cross section,WI ness greater than 15 cm (6 in.) and a w~dth no greater than twice the thickness usually smaller than a bloom and has a square or circular cross sectlon B~llets ar ly produced by some form of deformation process,such as rolhng or extrusi0n.A rectangular sohd where the w~dth is greater than tnnce the thickness. Slabs can h rolled to produceplate,sheet, andstrrp. Plates have th~ckness greater than 6 mm whtle sheet and stnp range from 6 mm to 0.1 mm inch to 0.004 inch). These hot-rolled products often form the startmg material for subsequent p such as cold forming or machining. Sheet and strip can be Cabr~cated Into pro

(i

SECTION 16.4 Rolling

383

Large dlameter prpe

Tin plate

Seamless pipe

Cold drawn bars

tl~rrlisr cold rollsd irlto ihinn~r,srrong?r rn'lteri~l or G \ r,n into foil (thickncsscs lccs t h ; ~ n 0.1 mm). Nloc~rnz ;in11 hiller, i i i n hc furthcr rolled into iinislieJ products. such as srn,cfu?ulsha;pesor railroad rail, or they can be processed Into semifinished shapes, such as bar, rod, tube, or ptpe. From a tonnage viewpoint, rolling is clearly predominant among all manufactfirlng processes, with a p p r m a t e l y 90% of all metal products expenencmg at least one rolling operation. Rolling equipment and rolling practices are sufficientlyadvanced that standardized, uniform-quality produots can be produced at relatively low cost. Because shapc~l roll>211-2 both massive : I ~co,tly, J \hap<d product\ arc unly a\.ailltblc in i[illidard ic)rnis and s u c s where t h c r ~ is sufficient dcni;lnd to pcr~nit economical production.

BASIC ROLLING PROCESS In the basicr~lling proces.a,shown inFigure 16-2,metal is passed between two rolls that rotate in opposite directions, the gap between the rolls being somewhat less than the thickness of the entering metal. Because the rolls rotate with a surface velocity that exceeds the speed of the incoming metal, friction along the contact mterface acts to propel the metal forward.The metal 1s then squeezed and elongates to compensate for the decrease in thickness or cross-sectional area. The amount of deformation that can be achieved in a single pass between a given pair of rolls depends on the friction conditions along the interface. If too much is demanded, the rolls cannot advance the material andsimply skid over its surface. If too little deformation 1s taken, the operation will be successful, but the additional passes teqnired to produce a gven part mll Increase the cost of production.
In hot rollmg, as with all hot-working processes, temperature control is requiredfor success. The starting material should be heated to a uniform elevated temoerafure.If the temaerature is not uniform, the subsequent defOrmatron will not he unkorm. Consider a Gece being reheated for rolling. If the soaking time is insmcienk, the hottex exteriorwill flow in preference to the cooler, stronger interior. Conversely,if a uniform-temperature material is allowed to cool pnor to working or has cooled during previous working operations,

384

CHArreR 16

~ulk Forming Processes

FIGURE 16-2 Schematic representation of the hot-rolling process, showing the deformation and recrystallization of the metal being rolled.

the cooler surfaces will tend to resist deformation. Cracking and tearingoftRi?$ may result as the hotter, weaker interior tries to deform. It is not uncommon for high-volume producers to begin with continuous-castf stock. The cooling from solidification is controlled so as to enable direct insertion a hot-rolling operation without additional handling or reheating. For smaller operah or secondary processing, the starting material is often a room-temperature solid, as an ingot, slab, or bloom. This material must first be brought to the desired ro temperature, usually in g a s or oil-fired soaking pits or furnaces. For plain-carbon low-alloy steels, the soaking temperature is usually about 1200C (2200' F).For s er cross sections, induction coils may be used to heat the material prior to rolling. Hot-rolling operations arc usually terminated when the temperature falls to about to 100C (100" to 200F) above the recrystallization temperature of the material b rolled. Such afinishing temperature ensures the production of a uniform fine grainsii prevents the possibility of unwanted strain hardening. lf additional deformation is req a period of reheating will be necessary to reestablish desirable hot-working conditio Cold rolling can be used to produce sheet, strip, bar, and rod products with ext ly smooth surfaces and accurate dimensions. Cold-rolled sheel and strip can be obtain various conditions, including skin-rolled, quarter-hard, huv-hard, and full-hurd. Skin-I mctal is subjected to only a 0.5 to 1%reduction to produce a smooth surface and uni thickness, and to remove or reduce the yield-point phenomcnon (i.e., prevent formatio Luders hands upon further forming).This material is well suited for subsequent c ing operations where good ductility is required. Quarter-hard. half-hard,and full-h and strip experience greater amounts of cold reduction, up to 50%.7heir yield higher. properties have becomc directional, and ductility has decreased. Quarter can be bent back on itself across the grain without breaking. Half-hard and fullhe bent back 90" and 45",respectively, about a radius equal to the material thickness For products with a unilorm cross section and cross-sectional dimensions less about 5 cm or 2 inches, cold rolling of rod or bar may be an attractive alternative t trusion or machining. Strain hardening can provide up to 20% additional streng the material, and the process offers the smooth surfaces and high dimensionalpreci of cold working. Like the rolling of structural shapes, however, the process gene requires a series of shaping operations. Separate roll passes (and roll grooves) ma requircd for sizing, breakdown, roughing, semiroughing, semifinishing, and finis While the various grooves may be in a single set of rolls, a minimum order of of product may be required to justify the cost of tooling.

,Upper roll

w~
2- High

Lower roll

W
2 -High

nonreversing Steel returned

reversing

16-3 Various roll rations used in rolling

ns.

4- High

Cluster

ROLLING MILL CONFIGURATIONS As illustrated in Figure 16-3,roUing mill stands are available in a variety of roll configurations. Early reductions, often called primary, roughing, or breakdown passes, usually employ a two- or three-high configuration with rolls 60 to 140 cm (24 to 55 in.) in diameter.The two-high nonreversing mill is the simplest design, but the material can only pass through the mill in one direction. '4 two-high reversing mill permits back-and-forth rolling, but the rolls must be slopped, reversed, and brought back to rolling speed between each pass. A three-high mi0 eliminates the need for roll reversal but requires some form of elevator on each side of the mill to raise or lower the material and mechanical manipulators to turn or shift the product between passes. As shown in Figure 16-4, smaller-diameter rolls produce less length of contact for a given reduction and therefore require lower force and less energy to produce a given change in shape.The smaller cross section, however, provides reduced stiffness, and the rolls are prone to flex elastically since they are supported on the ends and pressed apart by the metal passing through the middle (a condition known as three-point bending). Four-high and cluster arrangements use backup rolls to support the smaller work rolls. These configurations are used in the hot rolling of wide plate and sheets, and in cold

Length of contact (small d~arneterroll) Length of contact (larger-diameter roll)

386

CHAPTER 16

Bulk Forming Processes

rohng, where even small deflections m the roll would result in an unaccepta tion In product thickness. Foil is almost always rolled on cluster mtfls since th~ckness requlres small-diameter rolls. In a cluster m~ll. the roll in contact w~tt can be as small as 6 mm mch) In diameter.To counter the need for even sm some foils are produced by pack rolling, aprocess where two or more layers of rolled s~multaneously as a means of providing a thicker input mater~al. Hous~ minum foil 1s usually rolled as a double sheet, as ev~denced by the one ~hiny sic tact with the roll) and one dull side (in contact with the other plece of foil) In the rolling of nonflat or shaped products, such as structural shapes an call, the sets of rolls contain wntoured grooves that sequentially form the desii rcduce the cross-sectional area, and control the metal flow. Figure 16-5 s h ~ typical roll-pass sequences used in the production of structural shapes. Length as the cross section is reduced.

(i

CONTINUOUS (OR TANDEM) ROLLING MILLS When the volume of a product justifies the investment, rolling may be perfor contmuou~ or tandem rollrng m i l l .Billets, blooms, or slabs are heated and t e ~ an integrated serles of nonreversing rolling m~ll stands Continuous m~lls f c rolllng of steel strip, for example, often consist of a roughmg train of approxim four-high mill stands and a finrshtng tram of SIX or seven additional four-h~gh a contlnuous structural mill, the rolls m each stand contam only one set r grooves, m contrast to the multigrooved rolls used when the product 1s prc back-and-lorth passes through a srngle stand. If a s~ngle piece of material 1s in mult~ple rolling stations at the same tim pelatrve that the same volume pass through each stand in the same amount I the cross section isreduced, speed must be increased proportionately.Therefor~ terial 1s reduced in size, the rolls of each successive stand must turn faster tha~ the preceding one. If a subsequent stand is runnlng too slow, material will ac between stands. If the demand for incoming mater~al exceeds the output of ous stand, the material is placed m tension and may tear or rupture. The synchronization of SIX or seven mill stands is not an easy task, especi key variables such as temperature and lubr~cat~on may vary during a single ru product may be cxitmg the final stand at speeds in exccss of 110 k~lometers (70 miles per hour). Computer control is basic to succe?sll rolling, and moden equipped with numerous sensors to provide the needed information. When u casting urnts feed directly into contlnuous rollmg nulls, the time lapse from ~ I I fication to fin~shed rolled product 1s often a matter of a few mmutes.
FIGURE 16-5 Typical roll-pass sequences used in producing structural shapes.

I.D.

roll

FIGURE 16-6 Schematic of a

O.D,roll
(driven)

horizontal ring rolling operation. A s the thickness of the ring is reduced, its diameter will Increase.

RING ROLLING Ring rolling is a special rolling process where one roll is placed through the hole of a thick-walled ring and a second roll presses in from the outside (Figure 16-6).As the rolls squeeze and rotate, the wall thickness is reduced and the diameter of the ring increases. Shaped rolls can be used to produce a wide variety of cross-section profiles.The resulting seamless rings have a circumferential grain orientation and find application in products such as rockets, turbines, airplanes, pipelines, and pressure vessels. Diameters can be as large as 8 m (25 ft) with face heights as great as 2 m (80 in.). THREAD ROLLING Thread r~~llzng is a deformation alternative to the cutting of threads: it is illustrated in Figure 16-5 and discussed in Chapter 29. CHARACTERISTICS, QUALITY, AND PRECISION OF ROLLED PRODUCTS Because hot-rolled products .are formed and finished above their recrystallization temperature, they have little directionality in their properties and are relatively free of deformation-induced residual stresses. These characteristics may vary, however, depending o n the thickness of the product and the presence of complex sections. Nonmetallic inclusions do not recrystallize, so they may impart some degree of directionality. In addition, residual stresses can be induced by nonuniform cooling from the temperatures of hot working.Thin sheets often show directional characteristics, whereas thicker plate (above 20 mm or 0.8 in.) will usually have very little. Because of high residual stresses in the rapidly cooled edges, a complex shape, such as an I- or H-beam, may warp in a noticeable fashion if a portion of one flange is cut away. As a result of the hot deformation and the good control that is maintained during processing, hot-rotled products are normally of uniform and dependable quality. It is quite unusual to find any voids, seams, or laminations when produced by reliable manufacturers.The surfaces of hot-rolled products are usually a bit rough, however, and are originally covered with a tenacious high-temperature oxide, known as m i l l scale. This can be removed by an acid pickling operation,resulting in a surprisingly smooth surface finish. The dimensional tolerances of hot-rolled products vary with the kind of metal and the size of the product. For most products produced in reasonably large tonnages, the tolerances are within 2 to 5% of the specified dimension (either height or width). Cold-rolled products exhibit superior surface finish and dimensional precision, and they can offer the enhanced strength obtained through strain hardening. FLATNESS CONTROL AND ROLLING DEFECTS If we are rolling a flat product with uniform thickness, the gap between the rolls must be a uniform one. Attaining such an objective, however, may be difficult. Consider the upper roll in a set that is rolling sheet or plate. As shown in Figure 16-7, the material presses upward in the middle of the roll, while the roll is held in place by bearings that are mounted on either end and supported in the mill frame.The roll, therefore,is loaded in three-point bending and tends to flex in a manner that produces a thicker center and

(a) Load~ng on a roll~ngmill roll. The top roll 18 pressed upward in the center wh~le be~ng supported on the ends (b) The elast~c resoonse to the th;ee-point bending.
FIGURE 16-7

FIGURE 16-8 Use of a

"crowned" roll to compensate for roll flexure When the roll flexes in three-point bending, the crowned roll flexes Into flatness
I

--------- . ,

CZ3J
(a)

a
(b)

thinner edge. Sincc the thicker center will not lengthen as much as the thinner edge, result will be a product with eithcr a wavy edge or a fractured center. ll' the rolls are always used to reducc the same material at the same temperat by the same amount. the forces and deflections can bc predicted and the roll ca signed to have a specified profilc. If a "crowned," or barrel-shaped, roll is subject thc designed load. it will deflect into flatness, as illustrated in Figure 16-8. If the ap load is not of the designed magnitude. however, the resulting profile will not be defects may result. If the correction is insufficient.for example, wavy edges or center tures will still occur. If the correction is excessive, the center becomes thinner andlo resulting in a wavy center or cracking of the edges. Since roll deflections are proportional to the forccs applied to the rolls, product ness can also be improved by measures that reduce these forces. If possible,fri could be reduced. smaller-diameter rolls could be used, and smaller reductions be employed. Heating the workpiece generally makes it weaker,so increased wo temperature will also reducc the l'orce on the rolls. Horizontal tensions can be to the piece as it is being rolled (strip tension in sheet metal rolling). Since these ten combine with the vertical con~pression to deform the piece (stretching while squ ing), the roll forces and associated deflections are less. Othcr techniques to improve ness include increasing the elastic modulus of the rolls themselves through mat selection or providing sonic Iorm of backup support to oppose deflection, as lour-high and cluster mill configurations. Successful rolling requires the balancing of many factors relating to the bcing rolled, the variables of the rolling process. and lubrication between the wor and the rolls. Common delects include the nonuniform thickness previously dis dimensional variations caused by changes in workpiece temperature. surface flaws as rolled-in scale and roll marks), laps,seams, and various types of distortions.

THERMOMECHANICAL PROCESSING AND CONTROLLED ROLLING As with most deformation processes, rolling is generally considered to be a way ofc ing the shape of a material. While heat may be used to rcduce forces and promot ity, the themral processes that produce or control product properties (heat treatments usually performed as subsequent operations. Thernaornechanicalproce.r.ring, of which tr-r~llc?d rolling is an example, consists of integrating deformation and thermal pro

SECTION 16.5 Forging

389

into a single process that w~llproduce not only the desired shape but also the deslredproperbes, such as strength and toughness.The heat for the property modification 1s the same heat used in the 1oUlng operation, and subsequent heat treatment becomes unnecessary. A successful thermomechan~cal operatron begins with process design. The startmg materlal must be specified and the compositton closely maintained. Then a timetemperature-deformat~onsystem must be developed to achieve the deslred objective. Poss~ble goals include producmg aunlform fine gram size, controlling the nature, size, and dlstributron of the varrous transformation products (such as ferrite, pearlite, bamlte, and martensile in steels); controlling the reactions that produce solid-solut~on strengthening or precipitation hardenlog, and producing a deslred level of toughness. Start~ng structure (controlled by composition and prior thermal treatments), deformatlon deta~ls, temperature durmg the various stages of deformation, and the conditions of cool-down from the working temperature must all be specified and controlled. Moreover, the attainment of unrform properties requires uniform temperatures and deformat~ons throughout the product. Computer-controlled facllitles are an absolute necessity d thermomechanical processmg~s to be successfully performed. Possible benefits of thermomechanical processing include improved product properties; substantial energy savings (by eli~mnatlng subsequent heat treatrnenl); and the poss~ble substitution of a cheaper, less alloyed metal for a highly alloyed one that responds to heat treatment.

Forging is a term applied to a family of processes that induce plastic deformation through localized compressive forces applied through dies.The equipment can take the form of hammers, presses, or special forging machines. While the deformation can be performed in all temperature regimes (hot, cold, warm, or isothermal), most forging is done with workpieces above the recrystallization temperature. Forging is clearly the oldest known metalworking process. From the days when prehistoric peoples discovered that they could heat sponge iron and beat it into a useful implement by hammering with a stone, forging has been an effective method of producing many useful shapes. Modern forging is simply an extension of the ancient art practiced by the armor makers and immortalized by the village blacksmith. High-powered hammers and mechanical presses have replaced the strong arm and the hammer, and tool steel dies have replaced the anvil. Metallurgical knowledge has supplemented the art and skill of the craftsman, as we seek to control the heating and handling of the metal. Parts can range in size from ones whose largest dimension is less than 2 cm (1in.) to others weighing more than 170 metric tons (450,000 lb). The variety of forging processes currently offers a wide range of capabilities. A single piece can be economically fashioned by some methods, while others can massproduce thousands of identical parts. The metal may he (1)drawn out to increase its length and decrease its cross section, (2) upset to decrease the length and increase the cross section, or (3) s q u e e ~ r d in closed impression dies to produce multidirectional flow. As indicated inTable 15-2, the state of stress in the work is primarily uniaxial or multiaxial compression. Common forging processes include: 1. Open-die drop-hammer forging 2. Impression-die drop-hammer forging 3. Press forging 4. Upset forging 5. Automatic hot forging 6. Roll forging 7. Swaging 8. Net-shape and near-net-shape forging

390

CHAPTER 16

Bulk Forming Prucesses

OPEN-DIE HAMMER FORCING I n concept, open-dze hamm of old, hut massive mechanic The metal is first heated to the proper te ban heating. An mpact is then delivered plest industrial hammer 1s a workpiece, and the energy of the blow is forglng hammers now employ some for fluids surized air, steam, or hydraul~c strikmg velocities are achieved, wlth and the ability to shape pieces up to several tons. Computer-controlled hamme provide blows of diffenng i various stages of a gven operation. process and also mirumize the amo mon outlets for the excess 16-9 shows a large double-f F~gure Open-die forging d shape, the operator must or may contact the workpiece direct making concave or conv Manipulators may be used Whlle some finished parts can b employed to preshape metal in p parts like turb~ne totors and ge and up to 1 m (3 ft) in diame

Double-framedrop hammer. (Cou~tesy of Erie Press Systems, Er~e,PA.) ( FIGURE 16-9 (Left) Schematic diagram of a fotg~ng hammer

SECTION 16.5 Forging

391

1 Preform mounted

2 Metal displacement-

on saddleimandrel.

reduce preform wall thickness to increase d~ameter

3 Progressive reduct~on 4 Mach~nlng to near of wall thickness to net shape produce rlng d~rnens~ons.

minMizes the amount of subsequent machining. Figure 16-10 ~llustrates the unrestricted flow of material along with the open-die forging o f a multldiameter cyhndr~cal shaft and a seamless metal ring.

IMPRESSION-DIE HAMMER FORGING Open-die hammer forging (or smlth forglng, as it has been called) is a simple and flexible process, but it is not practical for large-scale production. It is a slow operation, and the shape and dimensional precision of the resulting workpiece is dependent on the skill of the operator. As shown in Flgure 16-11, impression-die or closed-dze forging overcomes these dtfficulties by using shaped dies to control the flow of metal. Figure 16-12 shows a typical set of multicavity dies.The upper plece attaches to the hammer and the lower piece to the anvil. Heated metal is positioned in the lower cavity and struck one

392

CHAPLER 16

Bulk Forrrling Processes

FIGURE 16-12 Impresston

drop-forging dies and the product resulting from each Impression The flash IS trlmmed from the finished connecttng rod In a separate tnmmmg dte. The sectional vlew shows the gram flow result~ng from the forging process (Courtesy of Forging Industry Associatfon, Cleveland, OH.) or more blows by the upper die.The hammering causes the metal to flow and com fill the die cavity. Excess metal is squeezcd out along the parting line to form around the periphery of the cavity.Tnis material cools rapidly,increases in strengt by resisting deformation, effectively blocks the formation ol additional flash. By t. ping material within the die, the flash then ensures the filling of all oI the cavity det The flash is ultimately trimmed from the part in a final forging operation. Inflashlrss forging, also known as true closed-die Corging, the metal is defor in a cavity that provides total confinement.Accurate workpiece sizing is required complete filling of the cavity must be ensured with no excess material. Accurate wo piece positioning is also neccssary, along with good die design and control of lubncati The major advantage of this approach is the elimination of thc scrap generated d flash formation, an amount that is often between 20 and 45% of the starting mat Most conventional forgings are impression-dic with tlash and are producedin with a series of cavities, where one or more blows of the hatnmer are used for each in the sequence.The first impression is oftcn an edging,fullering, or bending impress to distribute the metal roughly in accordance with thc requirements of the later cavi Edging gathers material into a region, while fullering moves material away. Interm ate impressions are for blocking the metal to approximately its final shape, withge ous corner and fillet radii. For sniall production lots, the cost of further cavitiesma be justified, and the blocker-type Iorgings are simply finished by machining. More the final shape and size are imparted by an additional forging operation in ajina finisher impression, after which the flash is trimmed from the part. Figurc 16-12 sh an example of these steps and the shape of the part at the conclusion of each.S every part is shaped in the same die cavities, each mass-produced part is a closed cate of all the others. Conventional closed-die forging begins with a simple hot-rolled shape anduti rehcating and working to progressively convert it into a more complex geometry. shape of the various cavities controls thc flow of matcrial. and the flow, in turn,imp

STRIKING

RETURNING

Schemat~c diagram of an impactor in the str~king and returning modes. (Courtesy of Chambersburg rlng Compony, Chornbenburg, PA)
16-13

Convent~onal

the or~ented structure discussed in Chapter 16. (Grain flow that follows the external contour of the component is in the crack-arrestor orlentation, improv~ng strength, ductihty,and reslstance to Impact and fatigue.) Through forgmg, we can also control the size and shape of vanous cross section&so the metal can he diqbibuted as needed to resist the applied loads. Couple these factors with a fine recrystallized grain structure (hot working) and the absence of voids (compressive forming stresses), and we see why forgings often have about 20% h~gher strengkh-to-weight ratlos compared with cast or machmed parts of the same material. Board hammers, steam hammer&and alr hammers have a11 been used m impression-die forging. An alternative to the hammer and anvil arrangement is the counterblow machme, or tmpactor, illustrated m Figure 16-13,These machines have two horizontal hammers that simultaneously Impact a workpiece that is positioned between them. Excess energy simply becomes recod, ~ncontrast to the hammer and anvil arrangement, where energy is lost to the machine foundation, and a heavy machine base is requ~red. Impactors also operate with less noise and less vibration and produce d~stinctly different flows of marer~al, 16.14. as illustrated 111 F~gure Heat-treatment costs can be reduced by the dxect quenching or controlled cooling of the hot parts as they emerge from the forging operation. Energy conservation can also be achieved through several processes that have been designed to produce a product that is somewhere between a convent~onal forg~ng and a conventional castmg. In one approach, a forging preform 1s cast from liqu~d metal, removed from the mold while still hot, and then finlsh forged in a single-cavity d~e.The flash 1s then tnmrned and the part IS quenched to room temperature. Forging preforms can also be produced by the spray depos~t~on of metal droplets into shaped contamers, as descr~bed in Section 19-10.These preforms are then removed from the mold, and the final shape and properties are imparted by a final forging operation. St111another approach is sem~soltd forging,discussed in Chapter 13.

Dlskformed by

PRESS FORGING In hammer or impact forging, the metal flows to dissipate the energy imparted in the hammer-workpiece collision. Speeds are high, so the forming time is short. Contact times under load are on the order of mil1iseconds.There 1s little time for heat transfer and coo11ng of the workpiece, and the a d ~ a b a t heating ~c that occurs dur~ng deformation helps to minlrmze chilling.It is possible, however, that all of the energy can be dissipatadditional absorpt~on ed by deformation of just the surface of the metal (coupled w ~ t h remalns essenttally by the anvil and foundation), and the Interior of the workp~ece undeformed. Consider the deformabon of a metal wood-sphtting wedge after it has been struck repeatedly by a sledge hammer.The top 1s usually "mushroomed," while the

the slower squeeztng aetton penetrates completely through the metal, producing arnore un~form deformation and flow. New problems can arise, however, because of the longer

~ H A P T E R16

Bulk FOI.mlng Processes time ot contact between the dies and the w0rkpiece.A~ the surface of it becomes stronger and less ductile, and may crack if deformabon is continued. are generally used to reduce heat loss, promote surface flow, and enable the of finer details and closer tolerances. Periodic reheating of the workpiece quired. I f the dies are heated to the same temperature as the workpiece and pres ceeds at a slow rate, isothermalforglng can be used to produce near-net-shape cam w t h uniform microstructure and mechanical properties. Forging presses are of two basic types, mechanical and hydraulic, and are usu quite massive. Mechanical presses use cams, cranks, or toggles to produce a preset reproducible stroke. Because of their mechanical dr~ves, different forces are avalab the various stroke positions. Production presses are quite fast, capa per minute, and are available In capacities ranging from 300 to 1 MN). Hydraulic presses move in response to fluid pressure in a plston a ly slower, more massive, and more costly to operate. On the positive side, h presses are much more flexible and can have greater capacity. Since motion is in to the flow of pressurized drive fluids, hydraulic presses can be pro different strokes for d~fferent operations and even different spe Presses can be used to perform all types of forging, including open die and imp die. Impression-die press forgings usually require less draft than drop forgings an higher dimensional accuracy. In addition, press forgings can often be completed single closing of the dies as opposed to the multiple blows of a hammer. Machines capacities up to 50,000 tons (445 MN) are currently in operation in A third type of press is the screw press, which in many ways acts llke a A large flywheel stores a predetermined amount of energy. This mitted to a vertical screw,which drives a descending ram. Downward all of the energy from the flywheel has been dissipated. Additional informatlon about the various types of presses and dr~ve mech can be found in the closing section of Chapter 17

DESIGN OF IMPRESSION-DIE FORGINGS AND ASSOCIATED TOOLING Thegeometrical possibilities for impression-die forging are quite numerous, with plex shapes like connecting rods, crankshafts,wrenches, and gears being c duced. Figure 16-15 shows a forged-and-machined steel automotive c significantly outperformed similar components made of austemper Parts typically range from under 3 lb. up to about 750 lb., with the major being between 7 and 20 in. (20 and 50 cm). Steels, stainless steels, and all minum, copper, and nickel can all be forged with fair to excellent results The forging dies are usually made of high-alloy or tool steel and can be to design and construct. Impact resistance, wear resistance, strength at elev perature, and the ability to withstand cycles of rapid heating and cooling mus outstanding.In addition, considerable care is requued to produce and maintain a and accurate cavity and parting plane. Better and more economical results obtamed if the following rules are observed:

1 . The dies should part along a single, flat plane if at all possible. If not, the partin should follow the contour of the part. 2. The parting surface should be aplane through the center of the forgmg,notn upper or lower edge. 3. Adequate draft should be provided-at least 3"for aluminum and 5 ' to 7" fo 4. Generous fillets and radii should be provided. 5. Ribs should be low and wide 6. The vanous sections should be balanced to avoid extreme differences in met 7. Full advantage should be taken of fiber flow lines. 8. Dimensional tolerances should not be closer than necessary.

SECTION 1 6 . 5

Forging

395

16-1 5 A 'iorged-andd autornob~le engine aft that has been formed

ance ~isu~erior to cranks t ductile iron.

The vatious design details,such as the number of intermediate steps, the shape of each, the amount of excess metal required to ensure dte filling, and the dimensions of the flash at each step, are often a matter of expertence.Each component is a new design entity and brings its own unique challenges. Computer-aided design has made notable advances, however, and the development and accessibility of hgh-speed, immensememory computers have enabled the accurate modeling of many complex shapes. Good dimensional accuracy is a characteristic of tmpression-die forginn. With reasonable care,the dimensions foriteel productscan be &intained withinthetolerances of 0.02 to 0.03 m. (0.50 to 0.75 mm). It should be noted, however, that the dimensions across the parting plane are affected by closure of the dies and are therefore dependent on die wear and the thickness of the final flash. Dimensions contained entirely within a sin& die segment can be maintained at a significantly greater level o f accuraq. Surfatefinish values range from 80 to 300 pin. Draft angles can sometimes be reduced, occasionally approachingzero, but This is not recommended for general practice. Selectionof alubricant is also critical to successful forging.The Iubricant not only affects the friction and wear and associated metal flow, but it may also be expected to i e s )and a partact as a thermal barrier (restrictingheat flow from the workplea to the d ing compound (preventing the part from sticking in the cavities).
UPSET FORGING Upset forging involves increasing rhe diameter of a matedal by compressing its length. Because of its use with a myriad of fasteners, it i s the most widely used of all forging
processes when evaluated m terms of the number of pieces produced. Parts can be upset forsed both hot and cold, with the operation generally beine. verformed on speoial high.. spccd rn;ichincs.Tl~z forgin? molion is usually horizontal, and thc workpircc is rapidl! moved frorn 5tntion to tation.\\'hilr most opc!rati,)ns srarr nith \r,irc or rod, some machines call upseT bars up to 25 cm (10 in.) in diameter. Upset forging generally employs spht dies that contain multiple positions or cavities as seen in the typfcaldie set of Figure 16-16.The dies separate enough for the bar to advance between them and move into position. They are then clamped together

FIGURE 16-16 Set o f upsetforging dies andpunches. The f pwduct resulting from each o the four positions is shown along, thebottom. (Courtesy of Ajax
Manufocturinp Company Euclid,

OH.)

396

CHAPTF.R 16

Processes Bulk I'orrn~ng

Appl~catlons o f rule 1
FIGURE 16-17 Schernat~cs ~llustrat~ng the rules governing upset forg~ng(Courtesy of Nat~onol Machinery Cornpony, Tiffin,OH.)

Appl~cat~ons o f rule 2

Appl~cat~ons of rule 3

V~olat~on of rule 1

Vtolat~on of rule 2

V~olatlon of rule 3

and a heading tool or ram moves longitudmally against the bar, upsetting it ln cavity. Separat~on of the dies then permits transfer to the next positlon or remo the product. If a new piece is started wlth each dle separation and an operation1 formed m each cavity simultaneously, a finished product can bemade with each of the machine By mcludmg a shearing operation as the lnitral prece moves in sition, the process can operate wlth continuous coil or long-length rod as its inc feedstock. Upset-forging machines are often used to form heads on bolts and other fas as well as to shape valves, couplings, and many other small components. The up gion can be on the end o r central poruon of the workpiece, and the final diamete be up to three times the original. The following three ruleq illustrated in Figure should be followed when designing parts that are to be upset forged: 1. The length of unsupported metal that can be gathered or upset in one blow injurious buckling should be lrmited to three times the diameter of the bar. 2. Lengths of stockgreater than three times the diameter may be upset successful vided that the diameter of the upset is not more than 1limes the d~ameter of 3. In an upset requiring stock length greater than three tlmes the diameter of th and where the diameter of the cavity is not more than 1times the diameter oft (the conditions of rule 2), the length of unsupported metal beyond the face dre must not exceed the diameter of the bar.

AUTOMATIC HOT FORGING Several equipment manufacturers now offer highly automated upset equipment in mrll-length steel bars (typically 7 m or 24 ft long) are fed into one end at room ature and hot-forged products emerge from the other end at rates of up to 1 per minute (86,400 parts per eight-hour shift).These parts can be solid or hollow, or symmetrical, up to 6 kg (12 lb) in we~ght, and up to 18 cm (7 in.) in diameter. The process begins with the lowest-cost steel bar stock: hot-rolled and air carbon or alloy steel. The bar is first heated to 1200' to 1300C (2200" to 235 under 60 seconds as it passes through h~gh-power induction coils It is then des rolls, sheared Into individual blanks, and transferred through several successwe stages,during which it is upset,preformed,hnal forged, and pierced (if necessary . parts can be produced at up to 180 parts per minute, and larger parts at rates order of 90 parts per minute.F~gure16-18 shows a typical deformation sequence variety of hot-forged ferrous products.

.5

Forging

Upset pancake Sheared billet

~lockerforginq

Finished ;ear

blank

FIGURE 16-18

(a) Typical fo~r-step hequelse to pronucc d spur-gear forg~nq by abtornmc t blocker forging, hot forqiny. The ,neared bi let progrcss'vely shaped :ntn ari d p ~pancake, f ferrn.1~ parrs produicd oy ai.tomalic nor lorginq at rates and finished gear uank. (b) Samples o between 90 ;and 180 parts per minute. (Courtesy uf lvo[iotiol Lfocn:nery Cotnpuny, Itlfin, 011.)

',

The automatic hot-forging process has a numbel. oE attractive Ceatures. Low-cost input matcrial and high production speeds have already been cited. Minimum labor is rcquired, and since no flash is produced, material usage can be as much as 20 to 30% greatcr than with conventional forging. With a consistent finishing temperature near 1050C (1900F). an air cool can often produce a structure suitable for machining, eliminating thc need for an additional anneal or nol-malizing treatment.Tolerances are generally within 0.3 mm (0.012 in.), surfaces are clean, and draft angles need only he 0.5 to 1" (as opposed to the conventional 3" to 5").1'001 life is ncarly double that of conventional forging because the contact times are only on the order o l h of a second. Automatic hot formers can also be coupled with high-rate, cold-Iorming operations. Preform shapes can he hot formed at rates that approach 180 parts per minute. These products call then be cold formed to final shape on machines that operatc at speeds near 90 parts per minule.The benefits of the combined operations include highvolume production at low cost, coupled with the precision, surface finish, and strain hardening that are characteristic of a cold-finished material. To justify an automatic hot-forging operation. however, large quantities of a given product must be required.A single production line may well require an initial investment in excess of $10 million.

FIGURE 16-14 (Left) Roilforging mach~ne in operation. (R~ght) Rolls from a roll-forging machine and the varlous stages In roll forging a pa&. (Courtesy oi' Ajax Manufacturing Compan& Euclid, OH.)

ROLL FORGING In roll forging, round or flat bar stock is reduced in thickness and increased in1 to produce such products as axles, tapered levers, and leaf springs. As illustra Figure 16.19, roll forging is performed on machines that have two cylindrical o cylindrical rolls, each containing one or more shaped grooves. A heated bar is in between the rolls. When the bar encounters a stop, the rolls rotate, and the bar gressively shaped as it is rolled out toward the operator. The piece is then tra to the next set of grooves (or rotated and reinserted in the same groove), and th repeats until the desired size and shape are produced. Figure 16-19 also show rolls and the product formed by each set of grooves.Figure 16-20shows the cross of one set of grooves and a piece being formed. In most cases there is no flash, oriented structure imparts favorable forging-type properties. SWAGING Swaging (also knawn as rotary swaging or radial forging) uses external hamme reduce the diameter or produce tapers or points on round bars or tubes. Figur shows a typical swaging machine, and Figure 16-22 shows a schematic of its components. The dies, located in the center of the apparatus, consist of two b

Schematic of the roil-forging process showin the two shaped mlls and the stock being formed. (Courtesy Forging industry Associat;on, Cleveland, OH.)
FIGURE 16-20

FIGURE 16-14 (Left) Rollforging machine in operation. (Right) Rolls frdm a roll-forg~ng mach~ne and the various stages In roll forging a part. (Courtesy I Alox Manufocturrng Company, Euclid, OH.)

ROLL FORGING In vollfi~rgiv~g, I-ound or flat bar stock is reduced in thickness and increased inler to oroducc such nroducts as axles. t a ~ e r e d levers. and leaf sorings. As illustratei Figurc 16-19,roll forging is performed on ma1 cylindrical rolls, each containing one or more, between the rolls. When the b& encounters a stop, the rolls rotate, and the bar is1 gressively shaped as it is rolled out toward the operator.The piece is then transfe~ to the next set of grooves (or rotated and reinserted in the same groove), and the p r ~ repeals until the desired size and shape are produced. Figure 16-19 also shows ase rolls and the product formed by each set of grooves. Figure 16-20 shows the crosssecl of one set of grooves and a piece being formed. In most cases thcre is no flash, and oriented structure imparts favorable forging-type properties. SWAGING Swaging (also known as rotary swaging or radial forging) uses external hammerin reduce the diameter or produce tapers or points on round bars or tubes. Figure shows a typical swaging machine, and Figure 16-22 shows a schematic of its inta components. The dies, located in the center of lhe apparatus, consist of two blocd

the roll-forging I the two shaped

1621 Tube be~ng

m a rotary swaging
?.

(Courtesy of the Timkrn

y, danton, OH.)

hardened tool steel. They combine to form a central hole that generally has a conical input transitioning to a cyhnder. An external motor drlves a large, massive flywheel, which 1s connected to the central spindle of the machine H~gh-speed rotatlon of the central u n ~ generates t centrifugal force, whlch causes the matchlng die segments and backlng blocks to separate. As the spindle rotates, the backlng blocks are driven into opposing rollers that have been mounted in a massive machine housing.To pass beneath the rollers, the backer blocks must squeeze the dies tightly together. Once the assembly clears the rollers, the d ~ eonce s agam separate and the cycle repeats,generatlng between 1000and 3000 blows per mlnute. With the machlne In motion,the operator simply inserts a rod or tube between the &es and advances it during the periods of die separation. Because the dies rotate, the repeated blows are delivered from vanous angles,reduclng the dlameter and ~ncreasing the length. Slnce the rotatlng spindle is usually hollow, the workp~ece can be passed completely through the machine or withdrawn after a preset length has been reduced. Swaging operations can also be used to form tubular products with internal cavities of constant cross secti0n.A shaped mandrel is ~nserted Into a thick-walled tube (or hollow-end workplece), and the metal 1s collapsed around it to simultaneously shape and size both the interior and exterior of the product Swag~ng over a mandrel can be

FIGURE 16-22 A variety of

swaged parts, some w~th Inter nal details. (Courtesy of Oncinnoti Inc, Clncinnotr, O H ) M~locron, used to form parts with internal gears, splines, recesses, or sockets. Figure 16-23 sh variety of swaged products, many of which contain shaped holes. The term swaging has also been applied to a process where material is force material. Figure 16-24 shows a hot-swaging sequence being used to form the en pressurized gas cylinder.

NET-SHAPE AND NEAR-NET-SHAPE FORGING

effort has been made to develop processes that can form parts close enou dimensions that little or no final machining is required. These are known as

overall decrease in the amount of energy required to produce the product. Precision or near-net-shape forgings can now be produced with draft angles than 1" (or even zero draft). Complex shapes can be forged with such close tole that little or no finish machining is required. Since the design and implementation shape processing can be rather expensive,application is usually reserved for parts a significant cost reduction can be achieved.

Second swaging
(b)

Finished cylinder

(d)

FIGURE 16-24 Steps in swaging a tu form the neck of a gas cylinder. (Coorte United States Steel Corp., Pittsburgh, PA.)

Direct extrusion
1 Extrusion 2 Die backer 3 Die 4 B~llet

5 Dummy block 6 Presstng ram

7 Conta~ner her
8 Cotltalner body

:In the extrusion process, metal is compressed and forced to flow through a suitably shaped die to form a product with reduced but constant cross section. Although extrusion may be performed either hot or cold, hot extrusion is commonly employed for many metals to reduce the forces required, eliminate cold-working effect&andreduce directional properties. Basically, the extrusion process is like squeezing toothpaste out of a tube. In the case of metals, a common arrangement is to have a heated billet placed inside a confining chamber. A ram advances from one end, causing the billet to first upset and conform to the confining chamber. As the ram continues to advance, the pressure builds until the material flows plastically through the die and extrudes, as depicted in Figure 16-25.The stress state within the material is one of triaxial compression. Aluminum, magnesium, copper, lead, and alloys of these metals are commonly extruded, taking advantage of the relatively low yield strengths and low hot-working temperatures. Steels, stainless steels, nickel-based alloys, and titanium are far more diificult to extrude.Their yield strengths are high, and the metals tend to weld to the walls of the die and confining chamber under the required conditions of temperature and pressure. With the development and use of phosphate-based and molten glass lubricants, however, hot extrusions can be routinely produced from these high-strength, hightemperature metals. These lubricants are able to withstand the required temperatures and adhere to the billet, flowing and thinning in a way that prevents metal-to-metal contact throughout the process. As shown in the left-hand segment of Figure 16-26, almost any cross-sectional shape can be extruded from the nonferrous metals. Size limitations are few because

Direct extiuston

extrus~onIn d~rect extrus~on, the ram and billet both move FIGURE 16-27 Direct and lnd~rect friction between the b~llet and the chamber opposesforward motton. For ~ndtrect extrusion, the b~llet a stationary There is no b~llet-chamber fr~ction, since there is no relatlve motton presses are now available that can extrude any shape that can be enclosed within ac of 75-cm (30-111.) diameter. In the case of steels and the other high-strength metals shapes and sues are a ba more limited, but, as the right-hand segment of Figure shows, considerable freedom st111exists. Extrusion has a number of attractive features. Many shapes can be produc extrusions that are not possible by rolling, such as otres contaimng reentrant ang longtudmal holes No draft is required, so extrusions can offer savings in both meta weight. Since the deformation is compressive, the amount of reduchon in a smgle s limited only by the capaciry of the equipment. Billet-to-product cross-sectional arear c can be ~n excess of 100-to-1 for the weaker metals. In addition, extrusion d~ev rela trvely inexpensive, and one d ~ may e be all that is required to produce a given pr Conversion from one product to another requires only a slngle die change, so small tities of a desired shape can be produced economically The major limitation of the p is the requirement that the cross sectton be uniform for the entlre length of the pro Extruded products have good surface finish and dimensional prec~sion. For shapes tolerances of 0.003 cmlcm or in.im. with a mmimum of 0.075 mm (0 003 in easily attainable. Grain structure is typlcal of other hot-worked n~etals. but stro rectional properhes (longitudinal versus transverse) are usually observed. St product lengths are about 6 to 7 m (20 to24 ft), but lengths in excess of 12m (40 R been produced. Since little scrap is generated, b~llet-to-product yields are rather

EXTRUSION METHODS
Extrusions can be produced by various techniques and equipment configuration extrusion is usually done by elfher the direct or indirect method, both of which lustrated in Figure 16-27. In direct extrusion,a solid ram drives the entire billet through a stationary die andmust provide additional power to overcome the fric resistance between the surface of the moring billet and the confining chambe md~rect extrrrsron, also called reverse, backward. or inverted extrusion, a hollo pushes the dle back through a stationary,confined billet. Since there is no relabve friction between the billet and the chamber 1s eliinated.The requ~red force is I and longer billets can be used with no penalty in power or efficiency. F~gure 16-28shows the ram force versus ram posltion curves for both d~re indirect extruslon.The areas below the lines have units of newton-meters or foot-p and are therefore proportional to the work requrred to produce the part. The ar tween the two curves is the work required to overcome the billet-chamber f~ during direct extrusion, an amount that can be saved by converting to indirect ex Unfortunately, the added complex~ty of the indirect process (applying force t a hollow ram, extracting the product through the hollow, and removing res~dual material at the end of the stroke) serves to increase the purchase price and maint cost of the required equipment. With either process, the speeds of hot extrusion are usually rather fast,s minimize the cooling of the billet w~thln the chamber. Extruded products can em rates up to 300 mlmin (1000 ft/min).The extrusion speed may he restricted, howe

-D ~ r e cextrustan t
lnd~rect extruslon

D~agrarn of the ram pos~tion for ndlrect ,ame product ie curve ie amount of ,tance) iifference CUNeS IS ,t-chamber

Deformation
and

die friction

Ram position

the large amounts of heat that are generated by the massive deformation and the associated rise in temperature. Sensors are often used to monitor the temperature of the emerging product and feed this information back to a control system. For materials whose properties are not sensitive to strain rate, ram speed may be mamtained at the highest level that will keep the product temperature below some predetermined value. Lubrication is another important area of concern I f the reduction ratio (cross section of billet to cross section of product) IS 100, the product will be 100 times longer than the starting billet. If the product has a complex cross section, its perimeter can be significantly greater than a circle of equivalent area. Since the surface area of the product is the length times the penmeter, this value can be more than an order of magnitude greater than the surface area of the original billet. A lubricant that is applied to the starting piece must thin considerably as the matenal passes through the d ~ and e is converted to product.An acceptable lubricant is expected to reduce friction and act as a barrier to heat transfer at all stages of the process.

METAL FLOW IN EXTRUSION

The flow of metal during extrusion is often complex, and some care must be exerc~sed to prevent surface cracks, interior cracks, and other flow-related defects. Metal near the center of the chamber can often pass through the die with little distortion, while metal near the surface undergoes considerable shearing. In dlrect extruslon, fnction between the forward-moving billet and both the stabonary chamber and die serves to further impede surface flow.The result is often a deformation pattern similar to the one shown in Figure 16-29. If the surface reglons of the billet undergo excessive coohug, surface deformation is further impeded, often leading to the formation of surface cracks. If quality 1s to be maintained, process control must be exercised in the areas of deslgn, lubrication, extruslon speed, and temperature.

EXTRUSION OF HOLLOW SHAPES Hollow shapes, and shapes with multiple longitud~nal cavities, can be extruded by several methods. For tubular products, the stationary os movmg mandrd processes of Figure 16-30 are quite common. The dle forms the outer profile, while the mandrel shapes and slzes the interior.

FIGURE 1629 Grid pattern

zhowrns the metal flow in a a rect cxtrLs 00 ihe billrt wd> sect~oned and me qr~d parterr! was engraved priorto extrusion. Mandrel

Die First step la)

Die

Second step

FIGURE 163Q Two methob of extruding hollow shapes uslng ~nternal mandrels. In part (a) the mandrel and ram have ~ndependent rnotlons; in part (h) they move as aslngle unlt.

Ram

mandrel

Ftrst step

Second step

For products w~th multiple or more complex cavrtres, a spider-mandreld ~ (a e it as a porthole, bndge, or torpedo die) may be required. As illustrated m F~gure flow5 around the arms of a "sp~der,"and a further reduct~on then forces the mat together. Since the metal is never exposed to contamination, perfect welds resL tunately, luhncanr~ cannot be used slnce they WIU contaminate the surfaces to t The process is therefore ltmited to materials that can be extruded wthout lubric can also be easily pressure welded. Since additional tooling 1s required, hollow extrusions will ohv~ously I than sohd ones, but a wide variety of continuous cross-secuon shapcs can bc that cannot he made econom~cally by any other process.

HYDROSTATIC EXTRUSION Another type of extrusion, known as hydrastutrc exrrusmn, is ~llustrated sche fluid surrounds the workp~ece and appl~e! in hgure 16-32.Here high-pressu~e necessary to extrude it through the die. The product emerges into either atr pressure or a lower-pressure fl~d-filled chamber.The process resembles d~rect but the pressurized fluid surrounding the b~llet prevents any upsetting. Slnce does not come into contact with the surround~ng chamber, b~llet-chamber el~mmated. In add~tion, the pressur~zed fluid can also emerge between the bill d ~ eactmg . in the form of a lubricant.

16.6 Extrusion SECTION

Billet chamber

405

16-51 Hot extrusion ow shape using a spider-

Seals

While the efficiency can be significantly greater than most other extrusion processes, there are problems related to the fluid and the associated high pressures (which typically range between 900 and 1700 MPa or 125 to 250 ksi). Temperatures are limited since the fluid acts as a heat sink, and many of the pressurizing fluids (typically light hydrocarbons and oils) bum or decompose at moderately low temperatures. Seals must be designed to contain the pressurized fluid without lcaking, and measures must he taken to prevent the complete ejection of the product, often referred to as blowout. Because of these features, hydrostatic extrusion is usually employed only where the process offers unique advantages that cannot be duplicated by the more conventional methods. Pre.~.~ure-ro-pressrire extrusion is one of the unique capabilities. In this variant, the product emerges from one pressurized chamber into a second high-pressure chamber. In effect, the metal deformation is performed in a highly compressed environment. Crack formation begins with void formation, void growth, and void coalescence. Since voids are suppressed in a compressed environment, the result is a phenomenon known as pressure-induced ductility. Relatively brittle materials such as molybdenum, beryllium, tungsten, and various intermetallic compounds can be plastically deformed without fracture, and materials with limited ductility become highly formable. Products can he madc that could not be otherwise produced, and materials can be considered that would otherwise have been rejected because of their limited ductility at room temperature and atmospheric pressure.

Bulk F ormlng Processes

Groove in wheel

Sectjon x- x

FIGURE 16-33 Cross-sectional schemabc of the Conform continuous extrusion process matenal upsets at the abutment and extrudes. Section x-x shows the material in the sho

CONTINUOUS EXTRUSION
finite-length products If the pushing force could be apphed to the periphery of

process is established. forms, including solid rod, metal powder, punchouts from other forming op chips from machining. Metalhc and nonmetalhc powders can be intimately co-extruded. Rapidly solidified material can be extruded without exposure to vated temperatures that would harm the properties. Polymer~c materials and e re~nforced plastics have been successfully extruded The most common feed, is coiled alumlnum or copper rod. Continuous extrusion complements and competes with wire drawing an rolhng as a means of producing nonferrous products with small, but unrform, c or more holes. Since extrusion operations can perform massive reductions single die, one Conform operation can produce an amount of deformat~on to 10 conventional drawing or wld-rolling passes. In addition, sufficient heat c erated by the deformation that the product wiU emerge in an annealed conditio for further processing without mtermed~ate heat treatment.

AND TUBE DRAWING

Wire-, rod-, and tube-drawing operations reduce the cross section of a material it through a die.In many ways, the processes are similar to extrusion,but theapp es are now tensile, pulhng on the product rather than pushing on the workpiece.Ro

SECTION 16.7 Wire, Rod, and Tube Drawing

407

d~uwmg, illustrated schematically in Fagme 16-34,is probably t h e simplest af these operations, One end of a rod is reduced or pounted,so that it can p m through a die of somewhat smaller cross section.?he protruding material is then placed in grip and pnlled in tension, drawing the remainder of the rod through the die. The rods reduce in seaon, elongate, and become stronger (strain harden).Since the product cannot be readily bent or coiled, straight-puUdraw benches are generally employed with finite-1engthfeedsYock. Hydraulic cylinders ean be used to providz the pull for short-length products,while cham drxves as depicted in Figure 16-35,can he used to draw products up to 30 rn (100 ft) in length. The reduction in area is usuaIly restricted to between 20 and 50%, since higher values require higher pulling forces that may exceed the tensile strength of the reduced product.To produce a desired size oc shape,multiple draws may be required through a series of progressively smatler dies Intermediate anneals m y also be requned to restore ductility and enable further deformation. Tube drawing can be used to produce high-quality tubing where the product requires the smooth surfaces,thin walls, accurate dimenstons,and added strength (horn the strain hardening) that are characteristic of cold forming. Internal mandrels are often used to control the inside diameter of tubes,which range from about 12Xo 250 rnm (0.5 to 10 in.) m d~amter. As shown in Figure 16-36,these mandrels are inserted through the incoming stock and are held in place dmng the drawing operation.

FLGURE 16-36 Cold-drawlng smaller tubing from larger

tubing The die seCs the Outer dimension while the stationary mandrel sizes the lnner diameter.

408

CHAPTER 16 Bulk Forming Processes

Statlonary

Feed

matepal
FIGURE 16-37 Tube drawing with afloating plug.

Thick-walled tubes and those less than 12m m (0.5 in.) in diameter are often without a mandrel in a process known as tube smking. Precise control of the ameter is sacrificed in exchange for process simplicity and the ability to draw long of product. If a controlled internal diameter must be produced in a long-length it is possible to utilize a poutingplug, like the one shown in figure 16-37.This p be designed for the specific conditions of material, reduction, and friction.If the on the plug surface is too great, the flowing tube will pull it too far forward or fracturing the tube wall. If the amount of friction is insufficient, the plu or vibrate within the tube and wrll not assume a stable position. If properly desi floating plug will assume a stable position within the die, and size the internal while the external die shapes and s~zes the outside of the tube. The drawng of bar stock can also be used to make products with sections. By using cold drawing instead of hot extrusion, the material eme cise dimensions and excellent surface finish. Inexpensive materials s strain hardening can often replace stronger alloys or ones that would require a a1 heat treatment. Smdl parts with complex but constant cross sections c nomically made by sectioning long lengths of cold-drawn shaped bars to pr individual products Steels,copper alloys, and aluminum alloys have all been col into shaped bars. W i r e drawing is essentially the same process as bar drawing except smaller-diameter material. Because the material can now be coiled, the conducted in a somewhat continuous manner on rotating draw blocks, like the lustrated schematically in Figure 16-38. Wire drawing usually begins with large hot-rolled rod stock approximately 9 mm ($in.) in diameter. After descaling o forms of surface preparation, one end of the coil is pointed, fed through a die, and the drawing process begins. Wire dies generally have a configuration similar to the one shown in Figur The contact regions are usually made af wear-resistant tungsten carbide o talline, manufactured diamond. Single-crystal diamonds can be used for the very fine wre, and wear-resistant and low-friction coatings can be applied to the

Starting stock

FIGURE 16-38 Schematic of

wire drawing with a rotating draw block. The rotating motor on the draw block provides a continuous pull on the incomina wire.

SECTION 16.8 Cold Formlng, Cold Forglng, and Impact Extrusion

Carbide die

409

Steel casing
Back relief

Bell radius
Appraach

Entrance angle angle

Land

FIGURE 16-39 Cross sect~on through a typical carblde wiredrawlng d ~ show~ng e the character~sbc reglons of the contour

Speed control

Finished

Draw block

Draw block

Finish draw block

F/GURE16-40 Schemat~c of a mult~statlon synchron~zed w~re-draw~nq machine. T o prevent accumulation or brcdkage, i t s nece,sary tu ensure that the sdmc volume of materal pci,secilirougli each 5tdt.on iri a q vcn tinic. The
aria feenback eiertronirs to provlde tnc nercsdry loopr around the sheaves behuccri the ,idl;on5 J,e w:re ~ens;oris sjbeed control.

die material substrates Lubrication boxes often precede the individual dies to help raduce friction drag and prevent wear of the dies. Because the tensile load is applied to the already reduced product, the amount of reduction is severely limited Multiple draws are usually required to affect any significant change m me.To convert hot-rolled rod stock to the fine wire that is used in household telephme lines requires passes through as many as 29 or 30 individual dies. To minimize handling and labor, these operations are usually performed on tandem machines, like the one shown schematically in Figure 16-40. Between 3 and 12 dies are mounted in a single machine, and the material moves continuously from one station to another in a synchronized manner that prevents any localized accumulahon or tension that might mduce fracture. After passing through all the d i in a tandem machine, the material usually requres an intermediate anneal before it can be subjected to further deEormation. By controlling the placement of the last anneal so the final product has a selected amount of cold work, wires can be made with a wide range of strengths (or tempers).Whenmaximum ductility and conductivity are desited, the wne should be annealed m courrolledatmosphere furnaces, after the final draw.

.8 COLD FORMING, COLDFORG~NG, AND IMPACT EXTRUSION

Large quantities of products are now being made by old formmg, a family of processes in which slugs af material aresqueezed into or extruded from shaped die cavities to produce finished parts of precise shape and size. Workpiece temperature varies from rbom temperature to several hundred degrees Fahrenheit. Cold herldmbtng is a form of the previously discussed upset farging.~s illustrated m F~gugure16-41,it is used for making enlarged seetions on the ends of rod or wise, such .as the heads of nails, balts, rivets, or other fasteners Two variations of the process are

410

CHAPTER 16 Bulk Forming Processes

Ejector stop

FIGURE 16-41 Typical steps In

a shear~ng and cold-headlng operation.

common. In the first, a piece of rod is first sheared to a preset length and then trans

tractive for producing nails, since the point can bc formed in the shearing or c operation. Enlarged sections can also be produced at locations other than the end rod or wire, in the manner illustrated in Figure 16-42.

diameter, speeds of 400 to 600 pieces per minute are typical. For larger diameter speeds may reduce to 40 to 100 pieces per minute. Alloys of aluminum and copper excellent Cormabilily. while mild steel and stainless steel are rated fair to good. A variety of extrusion operations, commonly called impad extrusion, can a incorporated into cold forming. In these processes, a mctal slug of predetermine is positioned in a dic cavity, where it is struck a single blow by a rapidly movingp The metal may flow forward through the die. backward around the punch, or ina bination mode. Figure 16-43 illustrates the forwurd and hrrckwaril variations, using

by the clearance between the punch and die, and the bottom thickness is set by the ward. and combination impacting.Typica1 production speeds range from 20 to 60 str per minule.

Ejector stop

+ -

FIGURE 16-42 Method of upsetting the center portlon of a rod. The stock is supported in both d~es durlng upsettlng.

SecrnoN 16.8 Cold Forming, Cold Forging, and Impact Extrusion

41 1

open die

closed die

AGURE 16-43

Backward and
Forward extrusion
open die

xed le ex:ce is : otate ; Is the , ,duce 3 secly atzltoff sofa gions :h the n.) in :s, the .have Alloy Iso be d size lunch.

I lorward extrusion with open and cloied dies.

Forward extrusion

closed die

FIGURE 1 6 - (a) Reverse, (b)forward,and (c) combined forms of cold extrusion. [Courtesy the Aluminum Anociation, Arlington, V A . )
I

1 & i d !

lbl

(cl

1 i,
' I
I,

5 both Length ' h conmined , le stop back- I trokes

AGURE 16-45 Steps in the lorming o f a bolt by cold extrusion, cold heading, and thread rolling. (Courtesy of Notional Machinery Co., Tiffin, OH.)

~
I

!
I

The impact extrusion processes were first used to shape low-strength metals such as lead, tin, zinc, and aluminum into products such as collapsible tubes for toothpaste, medications, and other creams; small "cans" for shielding electronic components; zinc cases for flashlight batteries; and larger cans for food and beverages. In recent years, impact extrusion has expanded to the forming of mild steel parts, where it is often used in combination with cold heading, as in the example of Figure 16-45. When heading alone is used, thereis a definite limit to the ratio of the head and stock diameters (as presented in Figure 16-17and related discussion).The combination of forward extrusion and cold heading overcomes this limitation by using an intermediate starting diameter.The shank portion is reduced by forward extrusion while upsetting is used to increase the diameter of the head. By using various types of dies and combining high-speed operations such as heading, upsetting, extrusion, piercing, bending, coining, thread rolling. and knurling, a wide

CHAPTER 1 6 Bulk l=armingProcesses

FlGUf a spar

~mac stock; Note (Courl Co., T

again

FIGURE 16-16 Cold-forrnlng sequence rnvolv~ng cutoff, squarmg, two extrusions, an upset, and a trimming operation. Also shown are the f~nished part and the trtrnrned scrap. (Courtesy of Nabonol Machinery Co., T,ff!n, OH.)

Trim hex

varlety of relatively complex parts can be cold formed to close tolerances Ftg illustrates an operation that incorporates two extrus~ons,a central upset, and eration to shape and trtm that upset. F i g r e 16-47 presents an array of truded products. The larger parts are generally hot formed and machi smaller ones are cold formed. Since cold forming is a chipless manufacturing process, producing parts by matJon that would otherwise be machined from bar stock or hot forgngs, the m

FIGURE 16-47 Tvw~cal Parts

maae by upsetting'jnd related or>er;ltions.fCourfe,v uf Noriono~ niachinery Co., Tlffm, OH.)

Cuffing (7k% waste)

Cold forrnlng (6% waste)

is used more efficiently and waste is reduced. F~gure 16-48compares the manufacture

of a spark plug body by machining from hexagonal bar stock w ~ t h manufacture by cold forming. Material is saved, dlachining time and cost are reduced, and the product is stronger, due to cold work, and tougher, as illustrated by the flow lines revealed ~n Figure 16-49. By converting from screw machining to cold formmng, a manufacturer of cruise-control housings was able to reduce material usage by 65% whlle simultaneously increasing production rate by a factor of 5. Wh~le cold forming is generally associated with the manufacture of small. symmetrical parts from the weaker nonferrous metals, the process is now used exten'sively on steel and stainless steel, with parts up to 45 kg (100 1b) in weight and 18cm (7 in.) in d~ameter. A T the small end of the scale, micro former^ are now cold forming extremely small electronic components with d~mensional accuracies w~thin 0.005 mm (0.0002 m.). Cold-formed shapes are usually axisymmetric or those w ~ t h relatively small departures from symmetry. Production rates are hlgh; dimensional tolerances and surface finish are excellent; and there are no draft angle&parting lines, or flash to trim off. There is almost no matenal waste, and a comiderable amount of machining can often be eIiminated when used in place of alternate processes. Strain hardening can provide additional strength (up to 70% stronger than machined parts),and favorable grain flow can enhance toughness and fatigue life. As a result, parts can often be made smaller or thinner, or from lower-cost materials. Unfortunately, the cost of the required toollng, coupled ~ i t the h high production speed, generally requires large-volume production, typically in excess of 50,000 parts per year.

Thick-walled seamless tubtng can be made by rotary pzrrcing, a process illustrated in Figure 16-50 A heated blllet 1 s fed longitudinally into the gap between two large, convex-tapered rolls These rolls are rotated in the same d~rectlon, hut the axes of the rolls are offset from the axis of the billet by about 6', one to the nght and the other to the left. The clearance between the rolls is preset at a value less than the d~ameter ot the incoming billet. As the billet is caught by the rolls, it is simultaneously rotated and driven forward. The reduced clearance between the rolls forces the blllet to deform into a rotating ellipse. As shown in the right-hand segment of Figure 16-50, rotation of the elliptical sectlon causes the metal to shear about the major axis A crack tends to form down the center axls of the blllet, and the cracked matenal IS then forced over a polnted mandrel that enlarges and shapes the opening to create a seamless tube.The result 1s a short length of th~ck-walled seamless tubing, which can then be passed through sizing rolls to reduce the dlameter andlor wall th~ckness Seamless tubes can also be expanded in dlameter by passlng them over an enlarging mandrel. As the diameter and arcumference increase, the walls correspondingly thin.

414

CHAMER 16 Bulk Forming Processes

FIGURE 16-50 (Left) Principle

of the Mannesmann process of producing seamless tubing.

(Courtesy of Amencan Brass Company, Cleveland, OH.) (Right)

Mechan~sm of crack formation in the Mannesmann process. The Mannermann mrlls common ing up to 300 mm (12 in )in diameter. L mtlls,which use the same prrnciple but replace the convex rolls of the Mannesm with larger-diameter conical disks.

CI 16.10 OTHER SQUEEZING PROCESSES

ROLL EXTRUSION Thin-walled cyhnders can be produced process. In the variant depicted in Figu ameter as they squeeze the rotating ma elongates as the wall th~ckness is reduced. In Figure 16-51b, the internal dl maintained as external rollers squeeze the material aganst a rotahng mandrel. the process has been used to produce cyhnders from 2 cm to 4 m (0.75 to 156 in ameter, most products have diameters between 7.5 and 50 cm (3 and 20 in.).

SIZING
Szzrag ~nvolves squeezing all or select metallurgy products to achieve a prescnbed thickness or enhanced dimensi sion. By Incorporating sizing, designers can make the initial tolerances of a liberal, enabling the use of less costly product~on methods Those dimensions be precise are then set by one or more sivng operations that are usually perfo simple, mechanically driven presses.

External

--

FIGURE 16-51 The roll-extrusion process: (a) w~th ~nternal rollers expanding the inner d~arneter; with external rollers reducing the outer diameter.

SELTON 16.10 Other Squeezing Processes

415

punch ~~~~~i~~

optional
iveted

RE 16-52 Joining ponents by riveting.

RIVETING In riveting, an expanded head is formed on the shank end of a fastener to permanently join sheets or plates of material. Although riveting is usually done hot in structural applications, it is almost always done cold in manufacturing. Where there is access to both sides of the work, the method illustrated in Figure 16-52 is commonly used.The shaped punch may be driven by a press or contained in a special, hand-held riveting hammer. When a press is used, the rivet is usually headed in a single squeezing action, although the heading punch may also rotate so as to shape the head in a progressive manner, an approach known as orbital forming. Special riveting machines, like those used in aircraft assembly,can punch the hole, place the rivet in position, and perform the heading operatio-all in about 1second. It is often desirable to use rivetingin situations where there is access to only one side of the assembly. Figure 16-53 shows two types of special rivets that can be used for one-side-access applications. The shank on the "blind" side of an explosive rivet expands upon detonation to form a retaining head when a heated tool is touched against the exposed segment. In the pull type, orpop-rivet, a pull-up pin is used to expand a tubular shank. After performing its function, the pull pin breaks or is cut off flush with the head.

Stak~ng punch

Work-

Shah
Before After

@ @ @ Oprlonal patterns
FIGURE 16-54 Permanently attaching a shaft to a plate by staking.

U R E 16-53

Rlvets for use in "bl~nd" riveting (left) explosive type; (right) k-type pull-up. (Courtesy ofAkoa Fartening Systems, P~ttsburgh,PA)

STAKING Staking is a method of permanently joining parts together when a segment of one part protrudes through a hole in the other.& shown in Figure 16-54, a shaped punch is driven into the exposed end of the protruding piece. The deformation causes radial expansion, mechanically locking the two pieces together. Because the tooling is simple and the operation can be completed with a single stroke of a press,staking is a convenient and economical method of fastening when permanence is desired and the appearance of the punch mark is not objectionable. Figure 16-54 includes some of the decorative punch designs that are commonly used. COINING The term coining refers to the cold squeezing of metal while all ofthe surfaces are confined within a set of dies. The process, illustrated schematically in Figure 16-55, is used to produce coins, medals, and other products where exact size and fine detail are required and where thickness varies about a well-defined average.Because of the total confinement (there is no possibility for excess metal to escape from the die), the input material must be accurately sized to avoid breakage of the dies or press. Coiningpressures may be as high as 1400 MPa or 200,000 psi.

pa,t

RE 16-55 ~h~ coining

416

CHAPTER 16 Bulk Forming Processes

FIGURE 16-56 Hubbing a die btock in a hydraulic press. Inset

shows close-up o f the hardened hub and the impression in the die block. The die block is contained in a reinforcing ring. The upper surface of the die block is then machined flat toremove the bulged metal.

HUBBlNG
flubbingLis a cold-working process that is used to plastically form recessed cavi a w0rkpiece.A~ shown in Fi y r e 16-56,amale hub (or master) is made with ther profile of the desired cavity.After hardening,the hub is pressed into an annealed (usually by a hydraulic press) until the desired impression is produced, (Note: P tion of the cavity can aften be aided by machining away some of the metal in re where large amounts of material would be displaced.) The hub is withdrawnran placed metal is removed.by a facing-type machining tut.The workpiece,.which tains the desired cavity,is then hardened by &eattreatment. Hubbingis often more economical than die sinking (machining the.cavity cially when multiple impressions are to be pmduced. One hub can be used to number of identical cavities,and it is generally easier to machine a male profile [wi posed surfaces) than a female cavity (where you are cutting in a hole).

11 16.11 SURFACE IMPROVEMENTBY DEFORMATION PROCESSING

Deformation processes can also be used to improve or alter the surfaces of met ucts. Pmrliag is the mechanical working of surfaces by repeated blows of impel or a round-nose fool.The highly localized impacfs flatten and broaden the metal but the underlying material restricts spread, resulting in a surface with residual pression. Since the net loadimg on a material surfaceis the applied loadminus the ual ,compression,peening tends to enhance the fracture resistance and fatigue tensile-loaded components For this reason, shot impellers are frequently used shafting,mankshafts, connecting rod's, gear teeth,and other cyclic-loaded corn Manual or pneumatic h a m e r s a r e frequently used to,peen the surfaces weldmmts, Solidification shrinkage and thermal contraction produce surfaces with ual tension. Peening can reduce a r cancel this effect, thereby reducing associat tortion and preventing cracking. Burnishing involves rubbing a smooth, hard object (under considerable pr over the minute surface;irregularities that are.produced during machining or s The edges of sheet metal stampings Can be burnished by pushing the stamp through a slightly tapered die having its entrance end a littlelarger than the w and its exit slightly smaller.As the part rubs along the sides of the die, the pre ficient to smooth the slightly rough edges that are characteristicof a blankin (see Figure 182).

' This pracws should not he eqnhsed with "hobbing:'a

machining process used fer cutting gears

'

I,

r. (Courtesy of Modison Inc., Sumfer, SC.)

Roller burnishing, illustrated in Figure 16-57,c a n b e used t o improve t h e size a n d finish of internal and external cylindrical a n d conical surfaces. T h e hardened rolls of a burnishing tool press against thesurface a n d deform the protrusions t o a more-nearlyflat geometry.The resulting surfaces possess improved wear a n d fatigue resistance,since they have been cold worked, a n d a r e n o w i n residual compression.

E1I Key Words

drop-hammer forging extrusion finisher impression finishing temperature flash flashless forging floating plug forging hammer hubbing hydraulic press hydrostatic extrusion impactor impression-die forging indirect extrusion isothermal forging mandrel Mannesmann mill mechanical press mill scale near-net-shape forging net-shape forging open-die forging pack rolling piercing pipe plate press forging pressure-induced ductility pressure-to-pressure extrusion ring rolling riveting rod roll extrusion roll forging rolling rotary piercing seamless tubing sheet sheet forming sizing slab spider-mandrel die staking Stiefel mill strip structural shape swaging thermomechanical processing thread rolling tube tube drawing tube sinking upset upset forging wire drawing

rmation processes

Review Questions
tion of forming equipment from anaresome of the possible means of classifyingmotaldeanalysis he different for a process forging and a process like rolling? are some of the common terms applied to the various s of rolled products? are hot-rolled p r v d u ~ t s generally limited to standard
f

7. Why is it desirable to have uniform temperature when hot

rolling a material'?

8. Why is it important to control the finishing temperature of a

hot-rolling operation? 9. What are some of the attractive attributes of cold rolling? 10, Discuss the relative advantages and typical uses of two-high rolling milkwith large-diameter rolls,three-highmills,and fourhigh mills. I I. Why is foil almost always rolled on a cluster mill? 12. Why is speed synchronization of the various rolls so vitally important in a conlinuous or multistand rolling mill? 13. What types of products are produced by ringrolling?

is it undesirable to minimize friction between the workand tooling in a rolling operation?

418

CHAFTLR I6

Bulk Forrn~ng Processes

14. Explain how hot-rolled products can have directional prop-

erties and residual stresses. 15. Discuss the problems in maintaining uniform thickness in a rolled product and some of the associated defects. 16. Why is a "crowned': roll always designed for a specific operation on a specific material? 17. What are some of the methods of improving the thickness uniformity of rolled products? 18. What is thennomechanical PI-oeessing,and what are some of its p~lssiblc advantages? 19. What arc some of the types of flow that can occur in forging operations? 20. Why are steam or air hammers more attractive than gravity drop hammers for hammer forging? 21. What are some of the attractive features of computercontrolled forging hammers? 22. What is the diffe~cnce between open-die and impression-die forging'? 23. Why is open-die forging not a practical technique for largescale production of identical products? 24. What additional controls must he exercised to perform flashless forging? 25. What is a blocker impression in a forging sequence? 26. What attractive featuresare offered by counterblow forging equipment, or impactors? 27. For what types offorgrng productsor conditions might a press be preferred over a hammer? 28. Why are heated dies generally employed in hot-press forging operations? 29. Describe some of the primary differences among hammers. mechanical presses, and hydraulic presses. 30. Why are different tolerances usually applied to dimensions conlained within a single die cavity and dimensions across the parting plane? 31. What are some of the roles played by lubricants in forging operations? 32. What is upset forging? 13. What are some of the typical products produced by upsetforging operations? 34. What are some o f the attractive [ealures of automatic hot forging? What is a 111ajorlimitation? 35. How does roll forging differ from a conventional rolling operalion? 36. What is swaging? What kind of products are produced?

37. How can theswaging process impart different sizes andsb to an interior cavity and the exterior of a product? 38. What are some possible objectives of near-net-shape for 39. What metals can he shaped by extrusion'? 40. What are some of the attractive features of the extr process? 41. Whatis the primary shape limitation of the extrusion 42. What is the primary benefit of indirect extrusion? 43, What property of a lubricant is critical in extrusion that not be required for processes such as forging? 44. What types of products are made using a spider-mandr Why can lubricants not b e used in spider-mandrel extr 45. What are some of the unique capabilities and special tious of hydrostatic extrusion? 46. What is the unique capability provided by pressu pressure hydrostatic extrusion? 47. How is the feedstock pushed through the die in cont extrusion processes? 48. Why are rods generally drawn on draw benches, while drawn on draw block machines? 49. What is the difference between tube drawing and tube ing'? 50. For what types of products might a floating plug be empl 51. Why are multiple passes usually required in wire-dr operations? 52. What types of products are produced by cold heading?, 53. What is impact extrusion and what vdriations exist? 54. If a product contains a large-diameter head and a diameter shank, how can the processes of cold extiusi cold heading he combined to save metal? 55. What are some of the attractive properties o r charx of cold-forming operations? 56. How might cold forging be used to substantially redu terial waste? 57. What processes can be used to produce seamlesspipe or 58. What type of products can be made by the roll-ex process? 59. What types of rivets can he used when there is access one side of a joint? 60. Why might hubbing be an attractive way to produce a of identical die cavities? 61. How might a peening operation increase the fractur tance of a product? 62. What is burnishingl

3blems
I Some snack foods such as rectangular corn ch~ps, are often formed hy a rolhng-type operation and are sub~ecl to the same types of detectr common to rolled sheet and strip. Obtam a bagot such a qnack and exanune the c h ~ p s L o ~dentify examples of roll~ng-relateddefects such as those discussed ~nSection 164 and shown in Figure 16-A

2. Consider the extrusion of a cylindrical billet, and comp

following. a. Assume the starting billet to have a length of 0.3 diameter of 15un.This is extruded into a cylindric uct that is 3 cm in diameter and 7.5 m long (a re ratio of 25).Neglecting the areas on the two ends, the ratio between the product surface area (map cylinder) and the surface area of the starting billet 6. How would this ratio change if the product were a with the same cross-sectional area as that of the3 ameter circle? c. Consider a cyljnder-to-cylinder extrusion with a ratio of R. Derive a general expression of the re1

FIGURE 16-A Some typical defects that occur during rolling wavy edges, edge crack~ng, and center crack~ng.

$ace areas of product to billet as a function of R. (Hint: Start 6 t h a cylinder with length and diameter both equal to !: unit. Since the final area will be lIR times the original, fhe final length will be R units and the final diameter will beproportional to l / f i ) . If the final product had a more complex cross-sectional &ape than a cylinder, would the linal area be greater thsn s less than that computed in part c? Relate your answers above to a consideration of lubrication during large-reduction extrusion operations. ie force required to compress a cylindrical solid between flat lrallel dies (see Figure 16-10) has been estimated (by a theory :plasticity analysis) to be: force here:
T
= thickness =

4 Mathematrcal analysls of the rolling of flat strip reveals that the roll-separation force (the squeezing force reqrured to deform the stnp) isdlrectly proportional to the term:

KlmL 1 +where.
f,,,,

K , = geometnc Constant m = frlctlon factor L = length of contact 1," = average thickness of the stnp in the roll b ~ t e

Since L is proportional to the roll radms, R, KIL can be replaced by K,R,so the force becomes proporhonal to the tern:

nR2u,

1 +2mR
7 .

4~ . .
If ti, and m are both positive numbers, how will the roll-separation force change as the strip becomes thinner? How can this effect he minimized? Relate your observations to the types of rolling mills used for various thicknesses of product. 5. In Figure 16-28, the vertical axis is force and the horizontal axis is position. If force is measured in pounds and position in feet, the area under the curves has un~ts of foot-pounds and is a measure of the work performed If the area under the mdirect extruslon curve 1s proport~onal to the work requ~red to extrude a product w~thout blllet-chamber fnctional reststance, regions of the d~rect extrusion curve how could the relat~ve be used to determ~ne a crude measure of the mechanical "efficiency" of direct extruslon7 6. Compare the fornung processes o l wlre drawmg, conventlonal extruslon, and contmuous extrusion w ~ t h respect lo conhnulty,reduct~on m area possible m a smgle operat~on,posslble materials, speeds, typ~eal temperatures, and other important processing varmbles 7. Ftgure 16-B shows the rolling of a wlde. thin strip where the width remalns constant as thickness 1sreduced Materlal enters the mlll at a rate equal to t,w,v, and exlts at a rate of f,w,v,. Since maEenal cannot be created 01 destroyed,these rates must be equal, and the w, terms wlll cancel. As a result, vf is equal to (id?) v,, The matenal enters at veloclly v, and accelerates to veloclty vras the matenal passes through the mlll.For stable v,, whch 1s a conatant, rolhng, the veloc~ty of the roll ~urface, must be a value between v, and vf. For these cond~txons, deand the rolls as scr~be the relatrve sliding between the s t r ~ p the stnp moves through the reglon of contact.

R = r a d ~ u of s the cylrnder

of the cyhnder

v, = yleld strength of the material

m = frlct~on factor

(between 0 and 1 where 0 is frlct~onless and 1 is complete shcking)

n engineering student 1s attempting to Impress h ~ date s demonstraung some of the neat aspects of metalformmg. eplaces a shmy penny between the platens of a 60,000-lb-cawty press and proceeds to apply pressure.Assume that the >inhas a,?-in dlameter and is 4 in thick.The yield strength estimated as 50.000 psi, and since no lubricant is applied, iction is that of co~nplete sticking, or rn = 1.0. Compute the force required to induce plastic deformation. If this force is greater than the capacity of the press (60,000 lb), cornpute the pressure when the full-capacity force of 60,000 lb is applied. If the press surfaces are made from thick plates of steel with a yield strength of 120,000 psi, describe the results of the demonstration. , A simple model of forging force uses the equation: force 'here:
no= yield
=

KO, A

ti = a dimensionless multiplying factor strength of the material A = projected area of the forging

:is assigned avalue of 3-5 for simple shapes without tlash,5-8 ,r simpleshapes with flash, and 8-12 for complexshapes with ash. Consider the two equations for forging force and disuss their similarities and differences.

FIGURE 16-8 Strip roll~ng where

the w ~ d t h of the strlp remains unchanged. The llnes across the workp~ece block the area of contact w~th the roils. The top roll has been removed for ease of virual~zatlon.

segment of a relatively large ratchet wrench, such as those used with conventional socket sets. The design specifications require a material with a minimum yield strength of 50,000 psi and an elongation of at least 2% in all directions. Additional consideration should be given to weight minimization (because of the relatively large size of the wrench), corrosion resistance (due to storage and use environments), machinability (if finish machining is required), and appearance. 1. Based on the size and shape of the product, describe several methods that could be used to produce the component For each method, briefly discuss the relative pros and cons.

question 1 , select an appropriate material from the alternatives discussed in question 2, making sure that the process and material are compatible. 4. Which of the combinations do you feel would be the "best" solution to the problem? Why? 5. For this system, outline the specific steps that wouldb required to produce the part from reasonable starting material. 6. For your proposed solution, would any additional heat treatment or suttace treatment be required? If so, w h a would you recommend? 7. If a variation of this tool were to be marketed as a "safety tool" that could be used in areas of gas leaks where a spark might be fatal, how would you modify your previous recommendations? Discuss briefly.

&REARING OPERATIONS

Seaming andFlanging Stra~ghtening 17.4 D R A W ~ G S A~~E T ~ I W O

&OCEWES

Design Aids for Sheet Metal Forming 17.5 ~ T E R N A ~ V METHODS E O F PRODUCIN~ SHEBT-WE Electrofotming 'spray Forming 17.6 PIPEWELDING Butt-Welded Pipe Lap-Welded Pipe 17.7 PRESSES Class~fkation of Presses Types of Press Frame SpcialTypes of Presses Press-Feeding Deviws OF A Case SLudy: FABRICATION ONE-PIECE BRA&
FLASHLIGHT ( 3 A 5 ~

Tmls and Dies for Piercing and

Angle Bending (Bar Folder and

Spinning Shear Fofmmg or E l m Turmng Stretch Forming Deep Drawing and Shal/ow Drawing Forming with WubberTogli~@ or Fluid Prassure Sheet Hydrofarming Tube Hydrbformg Hat-Drawing Operatious

PRODUCTS

Superplastk Sheet Forming Propeffies of Sheer hlate&+l

1 17.1 INTRODUCTION
The various classification schemes for metal deformabon processes have been presented at the beginning of Chapter 16, wlth the indlcat~on that our text will be grouplng by bulk (Chapter 16) and sheet (Chapter 17). Bulk forminguses heavy machinery to apply three-dimensional stresses, and most of the processes are considered to be primary operations. Sheet metal processes, on the other hand,generally involve plane stress loadings and lower forces than bulk forming.Almost all sheet metal forming is considered to be secondary processing. The classification into bulk and sheet n far from dlstlnct, however. Some processes can be cons~dered as e~ther, depending on the slze, shape, or thickness of the workpiece.The bending of rod or bar IS often considered to be bulk forming, while the bending of sheet metal is sheet forming.Tube bending can be e~ther, depending on the wall thickness and diameter of the tube. Similar areas of confusion can be found 1 x 1 deep drawmg, roll formlng, and other processes.T'he squeezing processes were rlescnbed in Chapter 16 Presented here will be the processes that involve shearing, bendtng, and drawmg. Table 17-1 lists somc of the processes that fit these categories

7.2 SHEARING OPERATIONS

Shearing IS the mechanical cutting of materials without the formation of ch~ps or the u%eo f burning or meltlng. It IS often used to prepare material far subsequent operatrons, and its success helps to ensure the accuracy and precision of the Iinished product. When the two cutting blades are straight, the process is called shear~ng.When the blades are curved, the processes have special names, such as blanking, plerczng, notching, and trrmmlng. In terms of tool design and materlal behavior, however, all are shearing-type operatlons.

421

tSimple shearing
2. Sliuing 3. Pieru'ng 4. Blanking 5. Fwblanking 6. Lancing 7. Norching 8, Nibbling 9.Shavtng 10. Trimming 11.Cutoff 12.Dinkhg
Z Roll bending

3.Draw bending 4. Compression bending 5. Press bending 6.Tuhe hending 7. Roll forming 8. Seaming 9 . Ranging lo. Strai@tening

3. Stretch fotming 4. Deep drawing and shaliowdra 5. Rubber-tool forming 6. Sheet hydroforming 7.X1b hydroforming '8,H Qdriiwing ~ 9. High-energyrateforming

10. Ironing 11.Embossing I t . Supeplaaicsheet farming

A simple type of shearing operation is illustrated in Figure 17-1.As the punc upper blade) pushes on the workpiece, the metal responds by flowing plastically into

Because of the normal inhomogencitics in a mctal and the possibility of non

edge. which combined with possible ~nicrocracksand work hardening of the shea edge can adversely affect subsequent forming processes. Changing the clearance between the punch and the die can greally change condition of the cut edge. If the punch and die (or upper and lower shearing bla have proper alignment and clearance, and are maintained in good condition, shea edges can be produced that have sufficient smoothness to permit use without lurther

perimeter of the cut.

smooth, deformed metal. Figure 17-1 shows one method of producing a compres environment. In the fineblanking process. a V-shaped protrusion is incorporated

squeeze the material from above and below, and dcsccnd in unison. extracting the sired segment. With punch-die clearances of about those of conventional bla '' thc shcared edges are now both smooth and square, as show11in Figure 17-4.

Die

-Burr

-Break Burnished land \ Roll over (die roll1

'

,Pressure plate with protrusion

Workpie

Die

424

CHAPTER 17 Sheet-Form~ng Processes

mc

Punch

Rod

Force

FIGURE 17-5 Method of smooth shearing a rod by putting it into compression during shearing.

Fineblanked parts are usually less than 6 mm inch) in thickness and typicall, complex-shaped perimeters Dimensional accuracy is often within 0.05 mm ( 0 00 and holes, slots, bends, and semipierced projections can be incorporated as part fineblanking operation. Secondary edge finishing can often be ehminated, and the hardening that occurs during the shearlng process enhances wear reastance. In finel ing, however, a triple-actionpress is generally required.The heblanking force is abou greater than conventional blanking of the same contour,and the extra malenal ref for the impingmg protrusion often Iorces a greater separatton between nested par Figure 17-5 illustrates another means of shearing under compression. Bar st pressed agalnst the closed end of a feed hole, placing the slock in a State of compre A transverse punch then shears the material into smooth-surface, burr-free slugs, for further processing.

(a

SIMPLE SHEARING
When sheets of metal are to be sheared along a straight he,squarmgshears, hke tt shown in Figure 17-6,are frequently used. As the upper ram descends, a clampb or set af clamping fingers presses the sheet of metal agalnst the machine table t r it firmly In position. A movtng blade then comes down across a fixed blade and' the metal. On larger shears, the moving blade is often set at an angle or "rocks" at scends, so the cut is made m a progressive fashion from one s ~ d e of the material other, much like a pair of household scissors This action significantly reduces the ai of cuttmg force required, replacmg a high force-short stroke operation w t h one o f force and longer stroke. The upper blade may also be incl~ned about 0.5" to 2.5" with respect to the blade and descend along this line of inclinat~on. While squareness and edge qualit be compromised, this action helps to ensure that the sheared material does not b r wedged between the blades.

SLITTING
Slztnng is the lengthw~se shearing process used to cut coils of sheet metal into sever, of narrower width. Here the shearing blades take the form of cylindrtcal rolls w t cumferential mating groovesThe raised ribs of one roll match the recessed groo

FIGURE 17-6 A 3-m (1 0-ft)

power shear for 6.5 mm (i-~n.) steel. (Courtesy of Ononnoti Incorporated, Cincinnati, OH )

SECTION 17.2

Shear~ng Operationc

425

Piercing

FIGURE 17-7 Schematic showing the difference between piercing and blanking.

the other.Theprocessisnow eantinuousand can be performed rapidly and economically. Moreover, since the distance between adjacent shearing edges IS B e d , the resultant strips have accurate and constant width, more consistent than that obtained from alternabye cuttmg processes.

PIERCING AND BLANKING

Pierctng and blankzng are shearing operations where a part is removed from sheet material by forcing a shaped punch through the sheet and into a shaped die. Any two-dimensional shape can be produced, with one surface having a slightly rounded edge and the other surface containrug a slight burr. Since both processes involve the .same bas~c cutting action, the p m a r y difference is one of definition.Fxure 17-7 shows . Lha~ in hlankiug, the piecc bding punched clut becorn's rhc wc~rkpiecc. In piercing. the pur~chuur is the hcrap and thc rcmainmg strip ir thv ~vorkpicce. Picrc~ns .~nd blanking nrd usually done on some form of mechanical press. Several variations of piercing and blanking are known by specific names. Laming is a piercing operation that forms either a 1me cut (slii] or hole, Like thoseshownin the left-hand portion of Figure 17-8.The primary purpose of lancing is to permit the adjacent metal to flow more readily in subsequent forming operations. In the case illustrated m Figure 17-8,the lancing makes it easier to shape the recessed grooves,which were formed before the ashtray was blanked from the strip st& and shallow drawn. Perfar a n g consists of piercing a large number of closely spaced holes. Notching is used to remove segments from along the edge of an existing product. In nibbling, a contour is progressively cut by producing a series of overlapping s11ts or notches, as shown in Figure 17-9. In this manner, simple tools can be used to cut a complex shape from sheets of metal up to 6 mm (i in ) thick,The process 1s widely used when the quantities are insufficient to justify the expensed a dedicated blanking die. Edge smoothness is determined by the shape of the tooling and the degree of overlap m successive cuts Shnv~ng is a fin~shimg operation in which a small amount of metal is sheared away from the edge of an already blanked part. Its primary use is to obtain greater dlmenslonal accuracy,but it may also be employed to produce a 7quared or smoother edge,Because only a small amount of metal is removed,the punches and dies must be made with very little clearance. Blanked parts, such as small gears, ean be shaw.d to produce d ~ m n sional accnracies within 0.025 mm (0.001 in.), In a cutoff operation, a punch and die are used to separate a stamping or other product from a strip of stock. The contour of the cutoff frequently completes the periphery of the workpiece. Cutoff operations are quite common in progressive die sequences, like several to be presented shoxtly.

Dinklng is a modified shearing operation that is used to blank shapes from1 strength materials, such as rubber, fiber, or c1oth.A~ ilIustrated inFigure 17-10,the sh of a die is either struck with a hammer or mallet or the entire die is driven downw by some form of mechanical press.

Hardwood block

FIGURE 17-16 The dlnking

process.

TOOLS AND DIES FOR PIERCING AND BLANKING As shown in Figure 17-11,the basiccomponents of a pierc~ng and blanklng die set which is attached above the die to keep the strlp apunch, a dle, and a str~pperplate, terial from ascending with the retracting punch. The positlon of the stripper plate the sue of its hole should be such that it does not interfere with either the horizontal tion of the strlp as it feeds into position or the vertical motion of the punch. Theoretically, the punch should fit within the die with a uniform clearance approaches zero. On its downward stroke, it should not enter the d ~ but e should s as its base aligns wrth the top surface of the d ~ eIn . general practice, the clear from 5 to 7% of the stock thickness and the punch enters slightly into the die cav If theface of the punch is normal to the axis of motion, the entire perimeter1 s~multaneousl~ By tilting the punch face on an angle, a feature known as shear or angle, the cutting force can be reduced substantial1y.A~ shown in Flgure 17-12,th nphery is now cut in a progressive fashion, similar to the action of a p a r of s c ~ s the opening of a "pap-top" beverage can. Variation in the shear angle contro amount of cut that IS made at any given t ~ m and e the total stroke that 1s necessary to plete the operation. Adding shear reduces the force but increases the stroke It is tractive way to cut thicker or stronger material on an existingpiece of equipmen Punches and dies should also be in proper alignment so that a unlform cle is ma~ntainedaround the entire periphery The die is usually attached to the plate of the press, which, m turn, is attached to the main press frame.The pun tached to the movable ram, enabling motion in and out of the die with each str the press. Punches and dies can also be mounted on a separate punch holder a shoe,like the one shown m Figure 17-13, to create an mdependent die set.The hold shoe are permanently aligned and guided by two or more gu~de pins. By aligning a and diq and fastening them to the die set, an entire unit can be inserted into without having to set or check the tool alignment This can s~gnificantlyred amount of production trine lost dunng tool change. Moreover, when a given pun d ~ are e no longer needed, they can be removed and new tools attached to the sh holder assembly. In most cases the punch holder attaches directly to the ram of the press, an motion acts to both raise and lower the punch. On smaller die sets, springs can corporated to provide the upward motion The ram simply pushes on the top punch holder, forclng it downward. When the ram retracts, the springs cause th to return to its starting position This form of construction makes the die se contained. It 1s simply posit~oned in the press and can be eas~ly remove reducing setup bme.

FIGURE 17-11 The basic components of piercing and blanking dies.

Blankfall~ng through die and bolster plate

Blanking with lunch (left) and ~ngular shear difference in and contact work (the area ) i s the same for

Maximum force

Maximum
Contact stroke force

Pos~t~on

Position

A wide variety of standardized, self-contained die sets have been developed. Known as subpress dies or modular tooling, these can often be assembled and combined on the bed of a press to pierce or blank large parts that would otherwise require large and costly complex die sets Figure 17-14 shows an assembly of subpress dies where a piece of sheet is inserted between the tooling and the downward motion of the press produces a variety of holes and slots, all in proper relation to one another.

RE 17-13 Typical die set having

~lignrnent guideposts. (Courtesy of 1/ EM, Cleveland, OH )

FIGURE 17-14 A piercing and blanking setup using self-contained subpress tool units. (Courtesy of Strippit Division, Houdaille Industries, lnc., Akron, NY)

428

CHAPTER 17

Sheet-Forming Processes Punches and dies are usually made from low-distortion or air-hardenable too so they can be hardened after machining with minimal warpage.The die profile is tained for a depth of about 3 mm inch) from the upper face, beyond which an lar clearance or back relief is generally provided (see Figure 17-11) to reduce fn between the part and the die and to permit the part to fall freely from the die afterb sheared. The 3-mm depth provides adequate strength and sufficient metal so tha shearing edge can be resharpelled by grinding a few thousandths of an inch from the of the die. Dies can be made as a single piece, or they can be made in component sections are assembled on the punch holder and die shoe.The component approach simpIifi duction and enables the replacement of single sections in the event o E wear or fra Complex dies like the one shown in Figure 17-15 can often be assembled from the standardized punch and die components that are available. Substantial sa be achieved by modifying the design of parts to enable the use of standard die nenls. A further advantage of this approach is that whe ed, the components can be removed and used to construct tooling for a When the cut periphery is composed of simple lines, and the mat either soft metal or other soft material (such as plastics, board),"steel-rule" or "cookie-cutter" dies can often be used.The cuttin from hardened steel strips, known as steel rule, that are mounted on edge in groo have been machined in the uppes die block. The mating piece of tooling can be flat piece of hardwood or steel, a male shape that conforms to the part the protruding strips descend around it),or aset of matching grooves into which die can descend. Rubber pads are usually inserted between the strips to replace per plate. During the conrpression stroke, the rubber compresses and all tion to proceed.As the ram ascends, the rubber then expands L o push steel-rule cavity. Steel-rule dies are usually less expensive to construct than solid are quite attractive for producing small quantities of parts. Many parts require multiple cutting-type operations, and it is duce a completed pal-twith each cycle of apress. Several types of dies have been to accomplish this task. For simplicity, their operations are discussed in terms o Eacturing simple,flat washers from a continuous strip of metal. The progressive die set, depicted in Figure 17-16, sically, it consists of two or more sets of punches an stock is fed into the first die, where a hole is pierced as the ram descends.When raises, the stock advances and the pierced hole is positioned under the blankin

(a

-punch Ram

holder

17-16 Progressive and biankin die far a square waier. Note punches are of different

31-1
I

Flat

o n , a t ; & ::
I

fed
stock

Direction of travel

Upon the second descent, a pilot on the bottom of the blanking punch enters the hole that was pierced on the previous stroke to ensure accurate alignment. Further descent of the punch blanks the completed washer from the strip and, at the same time, the first punch pierces the hole for the next washer. As the process continues, a finished part is completed with each stroke of the press. Progressive dies can be used for many combinations of piercing, blanking, forming, lancing, and drawing, as shown by the examples in Figures 17-8,17-15,and 17-17.They are relatively simple to construct and are economical to maintain and repair, since a defective punch or die does not require replacement of the entire die set. The material moves through the operations in the form of a continuous strip. As the products are shaped, they remain attached to the strip or carrier until a final cutoff operation. While the attachment may restrict some of the forming operations and prevents part reorientation between steps, it also enables the quick and accurate positioning of material in each of the die segments. If individual parts are mechanically moved from operation to operation within a single press,the dies are known as transfer dies. Part handling must operate in harmony with the press motions to move, orient, and position the pieces as they travel through the die. In c o m p o u n d dies, like the one shown schematically in Figure 17-18,piercing and blanking, or other combinations of operations, occur sequentially during a single stroke of the ram. Dies of this type are usually more expensive to construct and are more susceptible to breakage, but they generally offer more precise alignment of the sequential operations. If many holes of varying sizes and shapes are to be placed in sheet components, numerically controlled turret-type punch presses may be specified. In these machines, as many as 60 separate punches and dies are contained within a turret that can quickly be rotated to provide the specific tooling required for an operation. Between operations, the workpiece is repositioned through numerically controlled movements of the worktable. This type of machine is particularly attractive when a variety of materials and thicknesses (0.4 to 8.0 mm) are being processed. Still greater flexibility can be achieved by single machines that combine punch pressing with laser cutting or water-jet cutting.

progressive die. (Courtesy of the Minster Machine Company,

430

CHAPTER 17

Sheet-Forming Processes

FIGURE 17-18 Method for making a simple washer in a compound piercing and blanking die. Part is blanked (a) and subsequently pierced (b) in the same stroke. The blanking punch contains the die for piercing.

Scrap

(a1

I bi

DESIGN FOR PIERCING A N D BLANKING The construction,operation, and maintenance of piercing and blanking dies can be grea
facilitated if designers of the parts to be fabricated keep a few simple rules in mind:

1 . Diameters of pierced holes should not be less than the thickness of the metal. wi minimum of 0.3 mm (0.025 in.). Smaller holes can be made, but with difficulty.
2. The minimum distance between holes, or between a hole and the edge of the s: should be at least equal to the metal thickness. 3. The width of any projection or slot should be at least 1 times the metal thick never less than 2.5 mm (& in.). 4. Keep tolerances as Large as possible. Tolerances below about 0.075 mm (0.003 will require shaving. 5, Arrange the pattern of parts on the strip to minimize scrap.

side

Compressio

FIGURE 17-19 (Top) Nature of a bend in sheet metal showing tension on the outside and compression on the inside. (Bottom) The upper portion of the bend region, viewed from the side, shows how the center portion will thin more than the edges.

Neutral

axis

Tension side

Bending is the plastic deformation of metals about a linear axis with little or no cha in the surface area. Multiple bends can be made simultaneously, but to be cla true bending, and treatable by simplc bending theory, each axis must be line dependent of the others. If multiple bends are made with a single die, the pr often calledfurming. When the axes of deformation are not linear or are not inde dent, the processes are known as drawing andlor stretching. and these operations w treated later in the chapter. As shown in Figure 17-19, simple bending causes the metal on the outside t stretched while that on the inside is compressed.The location that is neither stret nor compressed is known as the neutraluxis of the bend. Since the yield strengthof als in compression is somewhat higher than the yield strength in tension, the met the outer side yields first, and the neutral axis is displaced from the midpoint of th terial. The neutral axis is generally located bctwecn one-third and one-half of the from the inner surface. depending on the bend radius and the material heing bent. cause of this lack of symmetry and the dominance of tensile deformation, thc metalis erally thinned at the bend. In a linear bend. thinning is greatest in the center of thes and less near the frcc edges, where inward movement can provide some compe On the inner side of a bend, the compressive stresses can induce upsetti companion thickening of material. While this thickening somewhat offsets the thin of the outer section, thc upsetting can also produce an outward movement oft edges.This contraction of the tensile segment and expansion of the compressions can produce significant distortion of the edge surlaces that terminate a linear h distortion is particularly pronounced when bends are produced across the width of but narrow plates. Still another consequence of the combined tension and compression is thee recovery that occurs when the bending load is removed.The stretched rcgion conV

Wedge adjustment for raising and lowering wingAdjusting screw for

thickness of material

and the compressed region expands,resultrng rn a small amount of "unbending," known as sprzngback To produce a product with a specified angle, the metal must be overbent by an amount equal to the subsequent springback.The actual amount of spnngback W U vary with a number of factors,includrng the type of material and material thickness.

ANGLE BENDING (BAR FOLDER A N D PRESS BRAKE) Machines like the bar folder, shown in Flgure 17-20, can be used to make angle bends up to 150" in sheet metal under 1.5 mm in ) thick.The workpiece is inserted under the folding leaf and aligned in the proper posit~on. Raising the handle then actuates a cam, causmg the leaf to clamp the sheet. Further movement of the handle bends the metal to the desired angle.These manually operated machines can be used to produce linear bends up to about 3.5 m (12 ft) In length Bends m heavrer sheet or more complex bends in thin material are generally made onpresr brakes, lie the one shown in Figure 17-21.These are mechanical or hydraulrc presses with a long, narrow bed and short strokes. The metal is bent between interchangeable dies that are attached to both the bed and the ram As ~llustratedin Figures 17-21 and 17-22, different dres can be used to produce many types of bends The metal can be repositioned between strokes to produce complex contours or repeated

(6

432

CUAWFR 1 7 Sheet

rming Processes

FIGURE 17-22 Press brake d~es can form a varlety of angles and contours (Courtesy of Crncmnatr Incorporated, C~ncrnnatr, OH.)

FIGURE 17-23 D~es and operattons used tn the press brake form~ng of a roll bead (Courtesy of C~ncmnatr Incorporated, C~ncrnnat!,OH )

bends, such as corrugations. Figure 17-23 shows how a roll bead can be Eorme repeated strokes,repositioning,and multiple sets of tooling. Seaming, embossing,p and other operations can also be performed with press brakes, but these operatio usually be done more efficiently O n other types of equipment. The tools and support structures on a press brake are often loaded in th three-point bending discussed in Chapter 16 for rolling mill rolls. Elastic deflecti cause a variety of bend deviations and defects, and a number of means have b veloped to overcome the problems.

DESIGN FOR BENDING

Several factors must be considered when designing parts that are to he shaped b ing. One of the primary concerns is determining the smallest bend radius th formed without metal cracking (i.e., the minimum bend rudius).This value is d on both the ductility of the metal (as measured by the percent redudion in area o in a standard tensile test) and the thickness of the material being bent. Figure 17how theratis of the minimum bendradius R to the thickness of the material t v material ducti1ity.A~ this plot reveals, an extremely ductile material is required 1 to produce a bend with radius less than the thickness of the metal. If possibl should be designed with large bend radii. This permits easier forming and all designer to select from a wider variety of engineeringmaterials.

-1-

: ! ! ! L
00

lo

20

30

40

50

60

Relationship between minimum bend radius (relative to thick and measured the ductility by the of reduction the metal in being area in ben a
FIGURE 17-24

Reduction in area, %

uniaxial tensile test).

Transverse bends (poor)

Roll~ng

d~rect~on

Longltud~nalbends

(good)

If the punch radius is large and the bend angle is shallow,large amounts of springback are often encountered. The sharper the bend, the more likely the surfaces will be stressed beyond the yield point. Less severe bends have large amounts of elastically stressed material and large amounts of spnngback. In general, when the bend radius 1s greater than 4 hmes the material thickness, the tooling or process must provide springback compensation. If the metal has experienced previous cold work or has marked directional properties, these features should be considered when designing the b e n b g operation Whenever poss~ble, it is best to make the bend axls perpendicular to the direct~on of previous working, as shown in the upper portion of Figure 17-25. The explanation for t h ~ s recommendation has llttle to do with the grain structure of the metal, but is more closely related to the mechanical loading applied to the weak, oriented mclusions. Cracks can easily start along tensile-loaded ~uclusions and propagate to full cracking of the bend. If intersecting or perpendicular bends are required, lt 1s often best to place each at an angle to the rolling direction, as shown in the lower portion of Figure 17-25,rather than have one longitudmal and one transverse Another design concern is determinmg the dimensions of a flat blank that will produce a bent part of the desired precision.As d~scussed earlier in the chapter, metal tends to thin and lengthen when it is bent.The amount of lengthening is a function of both the stock thickness and the bend radius. Figure 17-26 illustrates one method that has been found to give satisfactory results for determming the blank length for bent products In addition, the minimum length of any protruding leg should be at least equal to the bend radius plus 1.5 times the thickness of the metal.

FIGURE 17-26 One nleUloa of delcrlrlining the $tart ng o ank , ' ~ e ( I . ) to, s ~ ~ c r a l oeno'nq opcrdrlons. Due to thlnn na. the uroduct will ,enathen u J r no forminu. ~ I.. , ,', . ano are 1r1rdtsred proddct amens or,s.'See tablr tn dGermrlc I ) &red on r 7e of r3diur R wnere I 's tnr ,rock rhckne,~.

material and bending method w~ll permlt them. Parts with multiple bends desiened wth most (or ~referablv all) of them of the same bend radiusTh~s will

included and a trimming operation performed after bendmg.

AIR-BEND, BOTTOMING, AND COINING DIES

I
d

angle of the resulting bend is set by the geometry of the toohng, adjusted for sub springback, and the mside bend radius 1s that machined on tbe nose of the Bottomng d ~ e are s designed for a specific material and material th~ckness, form bends of a single confieurabon. If the results are outs~de s~ec~ficat~o~ll ever, reproducibility of the bend geometry is excellent, provided there 1s con within the size and nroperties of the material being bent.

reprod&ibllity dependson the ab~llty to control the stroke of the press:~dapt trol and on-the-fly corrections are frequently used with air-bend t o o h g If bottoming dies continue to move beyond the full-contact pontion, the tl

Unfortunately, the loading is greatly increased onboth the press and the tools. Reproducible-stroke mecha~~ical presses are generally used for bottom and commg, while adjustable-stroke hydraulic presses are preferred for air he]

ROLL BENDING
Roll bendcng is a conknuous form of three-pomt bending where plates, sheets vioe, and even rolled shapes and extrusions are bent to a desired curvature usiit

formed, the linger rolls often reinire support on both ends.The support framc

end may be swung clear?however, to permit the removal of closed circular shapes or partially rolled product. Because of the variety of applications, roll-bending machines are available in a wide range of sizes, some being capable of bending plate up to 25 cm (10 in.) thick.

DRAW BENDING, COMPRESSION BENDING, AND PRESS BENDING Bending machines can also utilize clamps and pressure tools to bend material against a form block. In draw bending, illustrated in Figure 17-29, the workpieceis clamped against a bending form and theentire assembly is rotated to draw the workpiece along a stationary pressure tool. In compression bending, also illustrated in Figure 17-29, the bending form remains stationary and the pressure tool moves along the surface of the workpiece. Press bending, also shown in Figure 17-29, utilizes a downward-descending bend die, which pushes into the center of material that is supported on either side by wing dies. As the ram descends, the wing dies pivot up, bending the material around the form on the ram.The flexibility of each of the above processes is somewhat limited because a certain length of the product must be used for clamping. TUBE BENDING Quite often the material being bent is a tube or pipe, and this geometry presents additional problems. Key parameters are the outer diameter of the tube, the wall thickness, and the radius of the bend. Small-diameter, thick-mlled tubes usually present little difficu1ty.A~ the outer diameter increases, the wall thickness decreases or the bend radius becomes smaller, the outside of the tube tends to pull to the center, flattening the tube, and the inside surface may wrinkle. For many years, a common method of overcoming these problems was to pack the tube with wet sand,produce the hend,and then remove the sand from the interior. Flexible mandrels have now replaced the sand and are currently available in a wide variety of stylesand sizes.

Ram movement

(a) Draw bending

(bl Compression bending

(c) Press bending

17-29 (a) Draw bending, in which the form block rotates; (b) compression bending, in which a moving tool ses the workpiece against a srationaryform; (c) press bending, where the press ram moves the bending form.

CHA~E 17 R Sheet-Forming Processes

FIGURE 17-30 (a) Schematic

representation o f the cold roll-forming process being used to convert sheet or plate into tube. (b) Some typical shapes produced by roll forming.

ROLL FORMING
oped forming technique that competes directly with press brake forming, ext 5tamping.A~ shown in Rgures 17-30 and 17-31,the process involves the progress1 ing of metal strip as it passes through a series of forming rolls at speeds up to (270 ftlmin). Only bending takes place,aud all bends are parallel to one another. ness of the starting material is preserved,except for thinning at the bend radii.

ness and uniform cross section are now being formed. By changing the rolls, a single roll-forming machine can produce a wide different shapes. However, changeover, setup, and adjustment may take sew so a production run of at least 3000 m (10,000 ft) is usually required for any gi uct.To produce pipe or tubular products, a resistance weldingunit or seaming o is often integrated with the roll forming.

Section of stock

FIGURE 17-31 Eight-roll sequence for the roll forming of a box channel. (Courtesy of the Aluminum Associot Washington, DC.)

17-32 Var~ous types s used on sheet metal.

17-33 Mernod of etilny rod or ,heel bv g t thro.lgl~d ,et o f ten'ny rolls. For rod\ set of ro s ic i15ed to in tric ,rra:qt~~cn~nq re a : r w on.

SEAMING AND FLANGING Searning is a bending operation that can be used to join the ends of sheet metal in some form of mechan~cal interlock. Figure 17-32 shows several of the more common seam deslgns that can be formed by a serles of small rollers. Seamlng rnachrnes range from small hand-operated types to large automatic unlts capable of producing hundreds of seams per mlnute. Common products Include cans, pails, drums, and other similar contamers. Flanges can be rolled on sheet metal In essentially the same manner as seams. In many cases, however, the forming of both flanges and seams 1s a drawlng operation, since the bending can occur along a curved axls. STRAIGHTENING The objective of strarghtening or flatrenmg ir the opposlte of bending, and these operations are often performed before subsequent forming to ensure the use of flat or straight material that is reasonably free of res~dual stresses. Roll strarghtening or roller levelrng, illustrated in Figure 17-33, subjects the matenal to a serles of reverse bends. The rod, sheet, or m r e is passed through a series of rolls with progress~vely decreased offsets from a straight h e . As the malerial 1s bent, first up and then down, the surfaces ale stressed beyond their elastic hmit, replaclog any permanent set w ~ t h a flat or straight profile. Tens~on apphed along the length of the product can help induce the requlred

deformation
Sheet material can also be straghtened by a process calledstretcher leveling. Flmtelength sheets are grrpped mechan~callyand stretched beyond the elastlc hmit to produce the desned flatness.

The term drawcng can actually refer to two quite different operations. The drawing of wre, rod, and tube, presented In Chapter 16, refers to processes that reduce the cross sectlon of mater~al by pulling it through a die. When the starting mater~alis sheet, drawing refers to a fam~ly of operations where plastic flow occurs over a curved axis and the flat sheet IS formed into a recessed, three-dimensional part wlth a depth more than several hmes the thickness of the metal These operations can be used to produce a w ~ d range e of shapes, from small cups to large auton~obile and aerospace panels.

SPINNING Sp~nnlng 1s a cold-form~ng operation where a rotat~ng disk of sheet metal is progressively shaped over a male form, or mandrel, to produce rotationally ~ymmetrical shapes, such as cones, hemispheres,cyhnders,bells, and parabo1as.Aform block possessing the shape of the desired part is attached to a rotatlng spmdle, such as the drive sectlon of a s~mple larhe. A disk of metal is centered on the small end of the form and held tn place by a pressure pad. As the disk and form rotate, locallzed pressure is apphed through a round-ended wooden or metal tool or small roller that traverses the entlre surface of the part, causing it to

438

CHArnsn 1 7

sheet-Forming Processes
Steps in spinning

Follower held

Progress~ve of a sheet stages in the sp~nning metal product.

FIGURE 17-34

Form attached
to headstock spindle

Two stages in the sp~nning of a metal reflector. (Courtesy of Spincroft, Inc., New Berl~n,WI )

FIGURE 17-35

flow progressively against the form. Figure 17-34 depicts the progressive oper Figure 17-35 shows a part at two stages of forming. Because the final diameter of the part is less than that of the starting disk, the circumferential length decreases.Thisd must be compensated by either an increase in thickness, a radial elongation,or cir ential buckling. Control of the process is often dependent on the skill of thc op During spinning, the form block sees only localized compression, and the does not move across it under pressure. As a result, form blocks can often be m hardwood or even plastic.The primary requirement is simply to replicate the a smooth surface.As a result, tooling cost can be extremely low, making spinning tractive process for producing small quantities of a single part. With autom ning can also be used to mass-produce such high-volume items as lamp reflectors,co utensils, bowls. and the bells of some musical instruments. When large quantities produced, a metal form block is generally prelerred. Spinning is usually considered for simple shapes that can be directly with& from a one-piece form. More complex shapes, such as those with reentrant angle be spun over multipiece or offset forms. Complex form blocks can also be made frozen water, which is melted out of the product after spinning.

SHEAR FORMING OR FLOW TURNING Cones, hemispheres, and similar shapes are often formed by shear forming ing, a modification of the spinning process in which each element of the tains its distance from thc axis of rotation. Since there is no circumferent the metal flow is entirely by shear and no compensating stretch has to occ in Figure 17-36, the wall thickness of the product, I,, will vary with the angle of the ticular region according lo the relationship: t, = t , sin 0, where I,, is the thickness0

SECTION 17.4
Feed

Drawing and Stretching Processes

439

Finished part
-

Mandrel

FIGURE 17-36 Schematic representation of the basic shear-forming process.

FIGURE 17-37 Forming a conical part by reverse shear forrnlng.

Roller feed

t -

rnaterlal flow

FIGURE 17-38 Shear formlng a cylinder by the direct process.

FIGURE 1 7 - 4 0 Schematicof the deep-drawing process.

Because most of the deformation is induced by the tensile stretchin on the Corm block are far less than those normally encountered in bending orfo Consequently, there is very little springback, and the workpiece conforms to the shape of the tool. Since stretching accompanies bending or wrapping, pulled out before they occur. Because the forces are so low, the form blocks cano be made of wood, low-melling-point metal, or even plastic. Stretch forming, or stretch-wrup ,forming as it is often called. is quite popul the aircraft industry and is frequently used to form aluminum and stainless steel cowlings, wing tips, scoops, and other large panels. Low-carbon steel ca formed lo produce large panels for the automotive and truck industry. If mating and female dies are used to shape the metal while it is being stretched, the proce known as stretch-draw forming.

FlGUli dauhk blankf

press i

DEEP DRAWING AND SHALLOW DRAWING

The forming of solid-bottom cylindrical or rectangular containers from meta of the most widely used manufacturing processes. When the depth of the product isles its diameter (or the smallest dimension of its opening), the process is considered to be low drawing. II the depth is greater than the diameter,it is known as deep d Consider the simple operation of converting a circular disk of sheet metal into bottom cylindrical cup. Figure 17-40 shows the blank positioned over a die opening circular punch descending to pull or draw the material into the die cavity.The ma beneath the punch remains largely unaflected and simply becomes the bottom ofth The cup wall is formed by pulling the remainder of the disk inward and over the of the die, as shown in Figure 17-41.As the material is pulled inward, its decreases. Since the volume of material must remain constant, the decreas ential dimension must be compensated by an increase in another dimension, s ness or radial length. Since the material is thin, an alternative is to relieve the circ compression by buckling or wrinkling.Wrinkle formation can be suppressed,ho compressing the sheet between the die and a blankholder surface during the for

Sector of Compressive stress Metal flow p : m p r e s stress sive

FIGURE 17-41 Flow of material during deep drawing. Note the circumferential compression as the radius is pulled inward.'

Blankholder
ram

Blankholder
sl~de

Pressure ring
Blank

) /

'Punch

Formed piece

17-42 Draw~ng on a

ction press, where the der uses the second ion. In single-actionpresses, where there is only one movement that is available,springs or air pressure are often used to clamp the metal between the die and pressure ring. When multiple actions are available (two or more independent motions), as shown in Figure 17-42,the hold-down force can now be applied in a manner that is independent of the punch position. This restraining force can also he varied during the drawing operation. For this reason, multiple-action presses are usually specified for the drawing of more complex parts, while single-action presses can be used for the simpler operations. Key variables in the deepdrawing process include the blank diameter and the punch diameter (which combine to determine the draw ratio and the height of the side walls), the die radius, the punch radius, the clearance between the punch and the die, the thickness of the blank, lubrication, and the hold-down pressure. Once a process has been designed and the tooling manufactured, the primary variable for process adjustment is the holddown pressure or blunkholder force. If the force or pressure is too low, wrinkling may occur at the start of the stroke. If it is too high, there is too much restraint, and the descending punch will simply tear the disk or some portion of the already-formed cup wall. When drawing a shallow cup. there is little change in circumference, and a small area is being confined by the b1ankholder.A~a result, the tendency to wrinkle or tear is low. As cup depth increases, there is an increased tendency for forming both of the defects. In a similar manner, thin material is more likely to wrinkle or tear than thick material. Figure 17-43 summarizes the effects of cup depth, blank thickness, and

Blankholder force

FIGURE 17-43 Defect formation in deep draw~ng as a funct~on of blankholder force, blank thickness, and cup depth.

CHA~E 17 R Sheet-Forming Processes

FIGURE 17-44 Cup redrawing to further reduce diameter and increase wall height. (Left)

forward redraw; (right) reverse redraw. blankholder pressure. Note that for thin materials and deep draws, a defect-free uct may not be possible in a single operation, as the defects simply transition fron kling, to wrinkle plus tear, to just tearing as blankholder force is increased.Topr a defect-free product, the draw ratio is often limited to values less than about 2. The limitations of wrinkling and tearing can often be overcome by using mi operations. Figure 17-44 shows two alternatives for converting drawn parts into c cups. In the forward redraw option, the material undergoes reverse bending asit into the die. In reverse redrawing, the starting cup is placed over a tubular die,a: punch acts to turn it inside out. Since a l l bending is in one direction, reverse redra' produce greater changes in diameter. When the part geometry becomes more complex, as with rectangular or metric parts, it is best if the surface area and thickness of the material can remah tively constant. Different regions may need to be differentially constrained technique that can produce variable constraint is the use of draw heuds, vertic; jections and matching grooves in the die and blankholder.The added force of bc and unbending as the material flows over the draw bead restricts the flow of ma The degree of constraint can be varied by adjusting the height, shape, and size of t h ~ and bead cavity. Because of prior rolling and other metallurgical and process features, ths of sheet metal is generally not uniform, even in the simplest drawing operation.] material may be required to assure final dimensions, and trimming may be requi establish both the size and uniformity of the final part. Figure 17-45 shows ash drawn part before and after trimming.Trimming obviously adds to the productic because it not only converts some of the starting material to scrap but also adds; er operation to the manufacturing process, one that must be performed on an ak produced shape.

FORMING WITH RUBBER TOOLING OR FLUID PRESSURE Blanking and drawing operations usually require mating male and female, or upp lower, die sets, and process setup requires that the various components be propel sitioned and aligned. When the amount of deformation is great,multiple operatior

P~erceu, o~anken, and orawn part otlore d t i ~ ljfwr trimming.

FIGURE 17-45

....

FIGURE 17-46 The Cuer~n process for forming sheet metal products

be requ~red, each w ~ t h its own set of dedicated tooling, and intermediate anneals may also be necessary. Numerous processes have been developed that seek t o (1) reduce tooling cost, (2) decrease setup time and expense, or (3) extend the amount of deformation that can be performed wrth a smgle set of tools. Although most of these methods have distinct limitations, such as complexity of shape or types of metal that can be formed, they also have defm~te areas of apphcat~on Several forming methods replace either the male or female member of the die set with rubber or fluld pressure. The Guenn prutesa (also known as rubber-die forming) is deplcted In Figure 17-46. It is based on the phenomenon that rubber of the proper consistency, when tutully confined,acts as a fluid and transmits pressure uniformly in all directions. Blanks of sheet metal are placed on top of form blocks, which can be made of wood, Bakelite,polyurethane,epoxy,or low-meltmg-pornt metal.The upper ram coutarns a steel contamer. As the a pad of rubber 20 to 25 cm (8 to 10 m.) k k mounted w~tbin ram descends, the mbber pad becomes confined and transmits force to the metal, causing it to bend to the desired shape. Since no female d ~ is e used and Inexpensive form blocks replace the male die, the total tooling cost n quite low.There are no mating tools to align, process flexibility 1s quite high (different shapes can even he formed at the same time), wear on the mater~al and tooling 1s low, and the surface quality of the workpiece is easily mamtained, a feature that makes the process attractwe for forming prepainted or spec~ally coated sheet When reentrant sections are produced (as with product b in Flgure 17-46),rt must be possible to slide the parts lengthw~se from the form blocks or to d~sassemble a multipiece form from withln the product. The Guenn process was developed by the aircraft industry, where the production of small numbers of duplicate parts clearly favors the low cost of toolmg. It can be used on alummum sheet up to 3 mm in.) thick and on stainless steel up to 1.5mm (& in.). Magnesium sheet can also be formed if it is heated and shaped over heated form blocks. Most of the forming done w ~ t h the Guenn process is multiple-axis bending, but some shallow draw~ng can also be performed.The process can also be used to pierce or blank thin gages of alummum, as illustrated in Figure 17-47. For thls application the blanking blocks are shaped the same as the deslred workpiece, wlth a sharp face, or edge, of hardened steel Round-edge supporting blocks are positioned a short distance from the blanking blocks to support the scrap skeleton and permit the metal to bend away from the sheared edges. In bulgmg, flnrd or rubber transmits the pressure required to expand a metal blank or tube outward against a spht female mold or d~e.For simple shapes, rubber tooling can

(i

444

cHAae~ 17 sheet-~orming Processes

Springs Solid ring to hold split die together Split die

FIGURE 17-48 Method of

bulging tubes with rubber tooling. be inserted, compressed, and then easily removed, as shown in Figure 17-48. Fo alicated shaoes.fluid pressure may be required t o Lorm the bulge. - More compl( ment is required since pressurized seals must be formed and ma~ntained, M enabling the easy ~nsertion and removal of matenal that is required formass prod

SHEET HYDROFORMING

Sheet hydroformcngis really a fam~ly of p~ocesses in which a rubber bladder by fluid pressure, or pockets of pressurized liquid, replaces either the solid 1 female die of the traditional tool set. In a varlant known as hzgh-pressrtrefle forming. or flexforming, the rubber pad of the Guerin process 1s replac flexible rubber diaphragm backed by controlled hydraulic pressure at values 140 and 200 MPa (20,000 and 30,000 psi). As illustrated m Flgure 17-49,the sol d ~ ~ vthe e s sheet into the resisting bladder, whose preqsure is adjusted tbroug-, stroke. In a vanallon of this process that shares s~milarity to stretch forming, th? first clamped against the opening and the punch is retracted downward.Theflu pressurized, causing the workpiece to balloon downward toward the punch : pressure is uniformly distributed over the workpiece, the sheet is uniformly stret un~fomly thinned. The punch then moves upward, causing the conform to its profile.

I
-r

Pressure fluid inlet

D~aphragm

Alum~num
blank

Blankholder
Punch

F~rst operation

Second operatlon

Hiqh-pressure ex;o c-d~c formng,';nnwng ( I ) the blank in F, ace w;lh no pressure in the cavlty, (2) presx closed and cavity pressunred; (3) ram advanced with cav~ty maintarning flu~d pressure; and (4) pressure released and ram retracted. (Courtesy of Alum~num Association, Woshmgton, DC)
FIGURE 1749

1 1 1 1

Third operation

Find operation

,Rubber

bladder

-Sheet
1

metal blank

17-20 One form of ydroforming.

Pressurizee
fluid

The flexible membrane can also be used to replace the hardened male punch, as shown in Figure 17-S0.The ballooning action now causes the material to descend and conform fully to the female die. which may be made of epoxy or other low-cost material. Parallel-plate hydroforming, or pillow forming, extends the process to the simultaneous production of upper and lower contours. As shown in Figure 17-51, two sheet metal blanks are laser welded around their periphery or are firmly clamped between upper and lower dies. Pressurized fluid is then injected between the sheets, simultaneously forming both upper and lower profiles.This may be a more attractive means ofproducing complex sheet metal containers,since the manufacturer no longer has to cope with the problems of aligning and welding separately formed upper and lower pieces. While the most attractive feature of sheet hydroforming is probably the reduced cost of tooling, there are other positive attributes. Because a more uniform distribution of strain is produced, materials exhibit greater formability. Drawing limits are generally about 1.5 times those of conventional deep drawing. Deeper parts can bc formed without fracture, and complex shapes that require multiple operations can often be formed in a single pressing. Surface finish is excellent, and part dimensions are more accurate and more consistent. Because ihe cycle times for sheet hydroforming are slow compared to mechanical presses, conventional deep drawing is preferred when the draw depth is shallow and the part is not complex.The reduced tool costs of hydroloming make the process attractive for prototype manufacturing and low-volume production (up to about 10,000 identical parts),such as that encountered in the aerospace industry. Sheet hydroforming has also attracted attention within the automotive community because of its ability not only to produce low-volume parts in an economical manner but also to successfully shape lowerformability materials, such as alloyed aluminum shcct and high-strength steels.

TUBE HYDROFORMING
Tube hydroforming, illustrated in Figure 17-52, has emerged as a significant process for manufacturing strong, lightweight, tubular components. which frequently replace an assembly of welded stampings. As shown in Figure 17-53, current automotive parts include engine cradles, frame rails, roof headers, radiator supports, and exhaust components. In elementary terms, a tubular blank, either straight or preshaped, is placed in an encapsulating die, and the ends are sealed. A fluid is then introduced through one of the end plugs, achieving sufficient pressure t o expand the material to the shape of the die. A t the same time, the end closures may move inward to help compensate or

Instrument panel supports.

FIGURE 17-53 Use of

hydroformed tubes in
automotive appllcatlons. (Courtesy of MetalForming, a pubhcotron of PMA Servrces, Inc, for the Preosron Metalform~ng Assocrotron, Independence, OH )

SLCIION 17.4 Drawlng and Stretch~ng Processes

447

In low-pressure tube hydroformuzg (pressures up to 35 MPa or 5000 PSI), the tube is first filled with fluid and then the dies are closed around the tube.The pnmary purpose of the fluid is to act as a liquid mandrel that prevents collapsing as the tube is bent to the contour of the die. While the cross-sectlon shape can be changed, the shapes must be simple, corner radii must be large, and there must be mrnimal expansion of the tube diameter. In high-pressure tube hydroforming, an Internal pressure between 100 and 700 MPa (15,000 and 100,000 psi) is used to expand the dtameter of the tube,form~ng t~ght corner radii and significantly altered cross sections. Pressure-sequence hydroformrng begins by applymg low internal pressure as the die 1s cioslng.Th~s supports the inside wall of the tube and allows it to conform to the cavity.When the die 1s fully closed,h ~ g presh sure 1s then apphed to complete the formlng of the tube walls. Attractive features of tube hydroforming include the abllity to use lightwe~ght, high-strength materials; the increase In strength that results from strain hardenmg, and the ability to utilize designs w ~ t h varying thickness or varylng cross section.Welded assemblies can often be replaced by one-piece components, and secondary operations can often be reduced. Disadvantages include the long cycle time (low production rate) and relatlvely high cost of tooling and process setup, An emerging alternatlve to tube hydroforming is hot-metal-gas formmg. In this process, a straight or preformed tube is heated to forming temperature by Induction heatmg, and shaped using the applicahon of pressurized inert gas rather than fluid. Parts can then be quenched directly from the high-temperature forming operation. Expectations for the process include faster speed, lower cost, and greater flex~bility compared to tube hydroforming.

HOT-DRAWING OPERATIONS
Because sheet material has a large surface area and small thickness, it cools rapidly in a lower-temperature environment. For this reason, most sheet forming is performed at room or m~ldly elevated temperature. Cold drawlng uses relatlvely thin metal, changes the thickness very little or not at all,and produces parth m a wide varlety of shapes. In contrast, hot drawing is used for formmg relahvely th~ck-walled parts of emple geometries, and the material thickness may change signihcantly durlng the operation. As shown in Figure 17-54, hot-drawing operations are extremely similar to previomly discussed processes. The upper-left schematic shows a simple dnk-to-cup drawing operation w~thout a hold-down. Whlle the increased thickness of the material acts to reslst wrlnklmg, the height of the cup wall is still restr~cted by defect formation. When smaller diameter and higher waU height are desired, redraws, like that depicted in the upper-right schematic, can be used. The lower schemat~c of Figure 17-54 illustrates an alternat~ve where the cup is pushed through a series of dies \nth a single punch.

17-54 Method: of ving a cup-shaped part. eft) F~rst draw. (UFrper ,draw operation. (I .ewer) -d~e draw~ng. (COL ~rtery i States Steel Corp ,
8

fh, PA )

Sheet-Forming Processes

If the drawn products are designed to utilize part of the origmal disk as a around the top of the cup. the punch does not push the material completely throu d ~ but e descends to a predetermined depth and then retracts The partially drawn then ejected upward, and the perimeter of the remaiiung flange is trimmed to the d size and shape.
HIGH-ENERGY-RATE FORMING
A number of methods have been developed to form metals through the applicati

large amounts of energy in a very short time (hmgh stram rate).%ese are known as energy-rate fonnrng processes and often go by the abbreviation HERF. Many tend to deform more readily under the ultra-rapid load application rates used in processes. As a consequence, HERF makes it posslble to form large workpieces a ficult-to-form metals with less expensive equipment and cheaper tooling than otherwise be requued. HEW processes also produce less spnngback. 7111s is pro assowated with two factors: (1) the high compressive stresses that are created the elastic deformahon of the die produced by the ultra-high pressure &gh-energy-release rates can be obtained by five distmct methods: (1) und ter explosion^ (2) underwater spark discharge (electro-hydraulic techniques),(3) matic-mechanical means, (4) Internal combustion of gaseous mixtures, and (5) of rapidly formed magnetlc fields (electromagnetic techmques). Specmfic processe developed around each of these approaches and attracted considerable attentio ing the 1960s and 1970s when they were used to produce one-of-a-kind and small tihes of parts for the space program.They are currently playing a relatmvely musign' role in manufacturing technology.

IRONING lroncng ts the name given to the process that thins the walls of a drawn cylinder b ing it between a punch and h e where the gap IS less than the mcoming wall thic As shown in Figure 17-55,the walls reduce to a uniform thickness and lengthen the thickness of the base remains unchanged.The most common example of an product is the thin-walled beverage can. EMBOSSING Embosszng, shown m Figure 17-56, is a pressworking process in which raised lette other designs are Impressed in sheet material Baacally, rt 1s a very shallow dra eratlon where the depth of the draw is limited to 1to 3 tlmes the thickness of th and the material thickness remains largely unchanged. A common example of bossed product is the patterned or textured industr~al stair tread. SUPERPLASTIC SHEET FORMING Conventional metals and alloys typically exhibit tensile elongations in the range 30%. By producing sheet materials with ultra-fine grain size and performing th

FIGURE 17-56 Embossing.

I

FIGURE 17-55 The ~roning

process.

SECT~ON 17.4 Draw~ng and Stretch~ng Processes

449

mation at low strain rates and elevated temperatures, elongat~ons can exceed 100% behavlor can be used to form and may be as hlgh as 2000 to 3000% This s~pe~plasdlc matenal Into large, complex-shaped products w ~ t h compound curves. Deep or complex shapes can be made as single-piece, single-operahon presslngs rather than mult~step conventional pressings or multipiece assembl~es. At the elevated temperatures required to promote superplast~city (about 900C for taanium, between 450" and 520C for aluminum, or generally above half of the melting point of the material on an absolute scale), the strength of matenalis sufficiently low that many of the superplastlc f o n n g techniques are adaptahons of processes used to form thermoplastics (d~scussed in Chapter 14).The toohng doesn't have to he except~onally ~trong, so form blocks can often be used in place of die sets. In thermoforming, a vacuum or pneumatic presure causes the sheet to conform to a heated male or female die. Blow forming, vacuum form~ng, deep drawing, and combmed superplastic forming and diffus~on bonding are other poss~bil~ties. Precision is excellent. and fine deta~ls or surface textures can be reproduced accurately. Springhack and residual stresses are almost nonexistent, and the producls have a fine, uniform gram slze. The major limtation to superplast~c forming 1s the low forming rate that is required to maintain superplastic behavior. Cycle Limes may range from 2 mlnutes to as much as 2 hours per part, compared to the several seconds that is typlcal of conventional presswork. As a result, applications tend to be hmited to low-volume products such as those common to the aerospace industry. By makmg the products larger and eliminating assembly operations, the weight of products can often be reduced, there are fewer fastener holes to initlate fatigue cracks, tooling and fabricat~on costs are reduced, and there is a shorter production lead tlme.

PROPERTIES OF SHEET MATERIAL A wide variety of materials have been used in sheet-formmg operations, including hotand wld-rolled steel, stainless steel, copper alloys, magnesium alloys, alum~num alloys, and even some types of plastics Sheet mater~al can also be coated or pamted, or even a clad or laminated composite. The success or fa~lure of many sheet-fomng operations is strongly dependent on the properties of the startmg material. Sheet metal has already undergone a number of processes, such as casting. hot rolling, and cold rollmg, and has acqulred distinct propert~es and character1stics.A simple uniax~al tenslle test ofthe sheet mater~al can be quite useful by providing values for the yleld strength, tensile strength, elongation, and strain-hardening exponent The amount of elongation prior to necking, or the uniform elongation, is probably a more useful elongation number, since localized thinnmg is actually a form of sheet mela1 failure.A low-yield, hlgh-tensile, and high-uniform elongation all combine to indicate a large amount of useful plastlc~ty. A h~gh value of the stram-hardening exponent. n (obtamed by fitting the true stress and true stram to the that the materia1 will have greater allowable equat~on: stress = K (strain)"), ind~cates stretch and a more uniform stretch. In the uniaxial tenslle test, the sheet is stretched in one direction and is perm~tted to contract and compensate in both width and thlckness. Many sheet-formmg operatlons subject the material to stretching in more than one doection, however. When the material is in biax~al tension, as in deep drawing or the vanous stretch-forming operatlons, all of the elongation must be compensated by a decrease m thickness. The strain to fracture is typically about one-third of the value observed in a uniax~al test.As a result, some form of biax~al stretching test may he preferred to asse5s formahil~ty, such as a dome-height or hydraulic bulge test. Sheet metal 1s often qulte anisotropic-properties varying w ~ t h d~rection or orlentation. A useful assessment of this varlahon can be obta~ned through the plastic straln ratio, R. Dunng a umaxial tens~le test,as the length increaces, the w~dth and thlckness both reduce.The R-value is simply the ratlo of the w~dth strain to the th~ckness strain. Materials withvalues greater than 1tend to compensate for stretch~ng by flow witlun the sheet and resist the thinning that leads to fracture. Hence materials with high R-values have good formabihty. Sheet materials can alw have direct~onal vanatlons w ~ t h ~ the n plane of

the sheet.Thcse are best e gitudinally (generally alon lar to the first), and one along average normal anisotropy (the sum of the four values of R divided by 4) and thepl anisotropy, A R (computed as R Since a h R of zero means that of the sheet, an ideal drawing AR. Unfortunately, the two va accompanied by directional variations within the plane of the sheet. Due to these v tions, disks being drawn in Sheet-forming properti ations in both telnperaturc a generally become stronger when deformed at higher speeds, while many alumin alloys weaken and become more prone to failure during the operation.

DESIGN AIDS FOR SHEET METAL FORMING

The majority of sheet metal failures occur due lo thinning or fracture, and both are th sult of cxcessive deformati ating the severity of deforma limit diagram. A pattern or grid. such as the one in Figure 17-57, is placcd on the s of a sheet by scribing. printing. or etching. The circles generally have diameters be 2.5 and 5 inm (0.1 to 0.2 in.) to enable detection of point-to-point variations in strain tribution. During deformation, the circles convert into ellipses, and the distorted p can be measured and evaluated. Regions where the enclosed area has expanded a cations of sheet thinning and possible failure. Regions where the area has contracted undergone sheet thickening and may be sites of possiblc buckling or wrinkles. Using the ellipses on the deformed grid, the major strains (strain in the dire of the largest radius or diam major) can be determined for a variety of locations.These values can then be plot a forming limit diagram such as the one shown in Figure 17-57.If both major and strains arc positive (right-ha i n g , and the sheet metal will definitely decrease in thickness. If the minor strainis ative, this contraction may partially or major direction.The combination of tension and compressio~i is known as drawin the thickncss may decrease. increase, or stay the samc, dcpending on the relativ nitudes o l the two strains. the diagram, since its purpose is to re only occur in a tensile environment.
FIGURE 17-57 (Left) Typical pattern for sheet metal deformation analysis; (right) forming limit diagram used to determine whether a metal can be shaped without risk of fracture. Fracture is expected when strains fall above the lines

Minor strain (percent1

Those strains that fall above the forming lunit line indicate regions of probable fracture. Possible correchve actions ~nclnde modification of the lubricant,change in the die design, or variation in the clamping or hold-down pressure. Stram analysis can also be used to determine the best orientation of blanks relative to the rolhng direction, assist in the design of dies for complex-shaped products, or compare the effectiveness of various lubricants. Because of the large surface-to-volume raho in sheet material, lubrication can play a key role 1n process success or fallure.

ALTERNATIVE METHODS OF PRODUCING SHEET-TYPE PRODUCTS

ELECTROFORMING Several manufacturing processes have been developed to produce sheet-type products by directly depositing metal onto preshaped forms or mandrels. In a process known as electroformmng, the metal 1 sdepos~ted by plating. Nickel,iron, copper, or silver can be depos~ted in thicknesses up to 16 mm in ).When the desired thlckness has been attained, plating is stopped and the product is strrpped from the mandrel A wide vanety of sizes and shapes can be made by electroforming, and the fabricanon of a product requires only a single pattern or mandrel. Low production quantities can be made in an economical fash~on, with the principal limitat~on being the need to strip the product from the mandrel. Rephcation of the contact surface and profile are extremely good, but the uniformity of thickness and external profile may present problems. For applications like the product~on ot mult~ple molds from a single master pattern, the Interlor surface is the critical one, and the wall thlckness serves only to prov ~ d the e necessary strength. Exterior irregular~ties are not critical, and various types of backup material can be employed to provlde additional support. For applications where the exterior dimensions are also important, uniform depos~tion 1 s required

(i

SPRAY FORMING Similar parts can also be formed by spray deposrhon. One approach is to inject powdered matenal into a plasma torch (a stream of hot ioruzed gas with temperatures up to ll,OOOC or 20,00OT).The particles melt and are propelled onto a shaped form or mandrel. Upon unpact, the droplets flatten and undergo rap~d sohdificatlon to produce a dense, finegrained product. Multiple layers can be deposited to build up a desired size, shape, and thickness Because of the high adhesion, the mandrel or form is often removed by machintng or chemical etching.Most applications of plasma spray forming involve the fabrication of specialized products from difficult-to-form or u1h.a-high-melttng-point maternls. The Osprey proass, descrtbed m Chapter 18 (Section 18-l0),can also be adapted to produce thm, spray-formed products. Here, molten metal flows through a nozzle, where it is atomzed and carried by high-velocity nitrogen jets The semisolid part~cles are propelled toward a target, where they impact and complete theirsohdification.Tubes, plates, and slmple forms can be produced from a variety of matenals. Layered structures can also be produced by sequenced deposition.

Large quantities of steelplpe are made by two procmses that use the hot forming of steel stnp coupled with deformation welding of the free edges Both of these processes, butt weldsng of plpe and lap weldmg of plpe,uhlize steel in the form of kelp-long strips w~th specified width, thickness, and edge configuration. Because the skelp was produced by hot rokng and the weld~ng process produces further deformation and recrystallization, pipe welded by these processes tends to be very uniform in structure and properties

BUTT-WELDED PIPE
In the manufacture of butt-welded pipe, steel skelp is heated to a specified hot-working temperature by passlng it through a tunnel-type furnace. Upon exiting the furnace, the skelp is pulled through forming rolls that shape it into a cylinder and bnng the free ends mto contact.The pressure exerted between the opposite edges of the skelp is suffichent

to upset the metal and produce a welded seam.Additiona1 sets of rollers then s u shape the pipe, and it IS cut to standard, preset lengths. Product diameters range 3 mm ~n.)to 75 mm (3 m.), and speeds can approach 150 d m l n (500 ftlmin).

(k

LAP-WELDED PIPE

CLASSIFICATION OF PRESSES

ment When selecting a press for a given application, coss~deration should be the capacity required, the type of power (manual, mechanical, or hydraulic), her of sbdes or dnves, the type of drrve, the stroke length for each dnve, th of presses and groups them by the type of drive. Manually operated presses such as foot-operated or krck presses are genes for very light work such as shearing small sheets and thm material. Movrng toward larger equipment, we find that mechanrcal drrves tend top last mohon and positive control of daplacement. Once burlt, however, the flexib a mechanical press is hmited, since the length of the stroke 1s set by the design

die operations.Typim1 capacrtles range up to about 9000 metric tons.

pled with an action that resembles a drap hammer (but slower and with less impa this reason, screw presses have become quite popular in the forging mdustry. In contrast to the mechanical presses, hydraulzc presses produce rnotlon as sult of p~ston movement, and longer or variable-length strokes can be progra

Crank

Eccentr~c Friction disk

Knuckle joint

Toggle

Screw

Hydraul~c

pressures are more accurately controlled, and full pressure is availahle throughout the entire stroke. Speeds can be programmed to vary or remain constant during an operation. Since position is varied through fluid displacement, the reproducibility of position will have greater variation than a mechanical press Hydraulic presses are available in capacities exceeding 50,000 metric tons and are preferred for operations requiring a steady pressure throughout a substantial stroke (such as deep drawing), operations requiring wide variation in stroke length, and operations requiring high or widely variable forces. In general, hydraulic presses tend to be slower than the mechanical variety, but some are available that can provide up to 600 strokes per minute in a high-speed blanking operation. By using multiple hydraulic cylinder$ programmed loads can be applied to the main ram, while a separate force and timing are used on the blank holder.

TYPES OF PRESS FRAME

Pved
&&SS

ifnini

c.
1

@.Fd
S h e

I'

As shown in Table 17-3,presses should also be selected with consideration for the type of frame. Frame design often imposes limitations on the size and type of work that can he accommodated, how that work is Fed and unloaded, the overall stiffness of the machine, and the time required to change dies. Presses that have their frames in the shape of an arch (arch-frame presses) are seldom used today, except with screw drives for coining operations. Gap-framepresses, where the frames have the shape of the letter C, are among the most versatile and commonly preferred presses. They provide unobstructed access to the dies from three directions and permit large workpieces to he fed into the press. Gap-frame presses are available in a wide range of sizes, from small bench types of about 1 metric ton up to 300 metric tons or more.

. . '5 . . ? , .'. ..... ..

'

. .,.

. .. :

.. ..' ~ ~ ~ i k c a t k i nPrews ' o.f . According to Type of Frame

Gap
Foot Bench Vertical Inclinable Inclinablr Open back Horn ?urret

Arch
Crank or eccentric Percussion

Straight Sided
Many variations. but all with straight-sidcd framcs

Popular design features include open back, inclinab~llty, adjustable bed, and ing bolster. Open-back presses allow for the eject~on of products or scrap throug opening ~nthe back of the press frame.Inclinuble press69 can be tilted,so that eje can be assisted by gravity or compressed a u jets. As a result of these features, openinclinable (OBI) presses are the most common form of gap-frame press The addl of an udjustable bed allows the base of the machine to raise or lower to accomm different workpieces. A slzdzng bolsfer permlts a second d ~ to e be set up on the while another IS in operatlon. Die changeover then requires only a few minutest clamp the punch segment of the active die, move the second die set into position, the new punch to the press ram, and resume operation Figure 17-59 shows an back tnclinable gap-frame press with a sl~dlng bolster A horn press is a special type of gap-frame press where a heavy cyhndric or "horn," appears in place of the usual bed Curved or cyl~ndrical workpiece placed over the horn for such operations as qeaming, punching, and riveting, trated in F~gure 17-60. On some presses, both a horn and a bed are provided,wit vision for swinging the horn aside when not needed. FlGURE 17-59 Inclinable Turretpresses are especrally useful in the production of sheet metal parts w gap-frame press w~th sl~dlng merous holes or slots that vary in size and shape.They usually employ a mQ bolster to accommodate two d ~ e sesfor rap~d change of tool~ng. framestructure and add upper and lower turrets that carry a number of punche (Courtesy of Niayara Machine & The two turrets are geared together so that any desired tool set can be quickly r Tool Works, Buffalo, NK) into position. Strarght-s~dedpresses have frames that consist of a crown, two up rights,^ b bed, and one or more movlng slides.Accessibility is generally from the front an but openings are often prov~ded m the side uprights to permit feeding and unlo workpieces. Straight-sidedpresses are available in a ulde vanety of sizes and desi are the preferred design for most hydraulic, large-capdc~ty, or special~zed mech dr~ve presses. As an added benefrt, elastic deflect~ons tend to be uniform acr workmg surface, as opposed to the angular deflechons that are typical of the g design. Figure 17-61shows a typical straight-sided press.

SPECIAL TYPES OF PRESSES

FIGURE 17-60 Mak~ng a seam on a horn press. Note the protrud~ng "horn" that replaces the lower press bed (Courtesy of N~ayara Machine & Tool Works, Buffalo,NY)

FIGURE 17-61 A 200-ton (1 800-kN) stra~ght-s~ded press

(Courtesy of Rousselle Corporation, West Chicago, IL.)

Presses have also been destgned to perform specific types of operations. Transfer1 have a long moving slide that enables multiple operations to be performedsimx ously in a single machine. Multrple die sets are mouilted s ~ d by e side along tht After the completion of each stroke, a continuous strip is advanced or indi workpreces are transferred to the next station by a mechanism like the one sht Figure 17-62.Transfer presses can be used to perform blanking, piercing, formint ming, drawing, flanging,embossing, and coining. Figure 17-63 ~llustrates the prod of a part that rncorporates a variety of these operations By usmg a single machine to perform multiple operations, transfer presse high production rates, high flexibility, and reduced costs (attributed to the rr labor, floor space, energy, and maintenance). Since production 1s usually hetwe and 1500 parts per hour, these machines are usually restricted to operations 4000 or more den tical parts are requ~red daily, each rnvolv~ng three or more se operations A total production run of 30,000 or more identlcalparts is generally ( betweenmajor changes in too1ing.A~ aresult,transfer presses are usedpnma dustries such as automotive and apphances, where large numbers of identic ucts are being produced. Four-clide or multlslrde machines like the one shown in Figure 17-64 are ly versatile presses that are designed to produce small, intricately shaped pa c o n h n ~ o ~fed ~ly wlre or coil strip. The basic machme has four power-driven sli motions) set 90" apart. The attached tooling is controlled by cams and designed erate m a progressive cycle. In the sheet metal vanation, strip stock is fed mto t chine,where it is stra~ghtened and progressively pierced, notched, bent, and cut of various slide stations. Flgure 17-65 presents the operating mechanism of one su chme. As the material moves from right to left. it undergoes a straightenu

'arming Processes
Top or

motions

FIGURE 17-66 Example of the piercing, blanking, and forming operations performed on a multislide machine. (Courtesy of U.S. Baird Corporation, Stratford, CT)

pierced, blanked, and formed into a folded sheet metal product.The strip stock up to 7.5 mm (3 in.) wide and 2 5 mm (& in.) thick. Wires up to about 3 mm (g i ameter are also commonly processed. Products such as hinges, links, clips, an blades can be formed at very high rates on these machines. Setup times are long,s production runs are preferred.

PRESS-FEEDING DEVICES

move them after forming. See Chapter 36 for an expanded discussion of manufa automation.

lli Key Words

flanging flow turning forming forming limit diagram forward extrusion Gueriu process high-energy-rate forming hold-down force hydraulic press independent die set ironing kick press lancing mechanical press minimum bend radius modular tooling multislide press neutral axis nibbling normal anisotropy notching Osprey process parallel-plate hydrofonning peening perforating piercing pillow forming pipe welding planar anisotropy,AR plasticstrain ratio, R pop-rivet press bending press brake progressive die punch rake angle redrawing roll bending roll forming roll straightening roller burnishing rubber tooling seaming shallow drawing shaving shear shear forming shearing sheet sheet hydroforming skelp slitting spinning spray forming springhack squaring shears steel-rule die strain analysis strain-hardening exponent, n stretch forming stretcher leveling stretching strip stripper plate subpress die superplastic forming swaging thread rolling transfer die transfer press trimming tube bending tube hydroforming turret-type punch press

IB Review Questions
pt dishngnishes sheet Eormingfrom bulb tormmg7

at is a definition of shearmg? are sheared or blanked edgm generally not smooth?

might a lo~lgsheanng cut he made in aprogressire fashion? LSa shtting operabn?

n ~ and h dre be m prop-

a complex die set from

d~esand transferdres? do wmpound dtes diffw from progressive dies? 1 s the attracilve feature of a turret-type punch press? re the stressstates on theexterior surface and interior

22. Why does a metal usually become thlnner m the region of a bend7 23. What 1s springhack, and why is it a concern dunng bendmg? 24. What types of oparations can be performed o n a press brake? 25. What factors determine the minimum bend radius for a mateTi-?ll . . . . . 26. If a r~ght-angle bend 1s to be made m a oold-rolled sheet, should it be made with the bend lying dong or perpendicular to the direction of previous rolling? 27. From a manufacruringvlewpoint, why is it desirable for all bends in a pradud (or component) to have the same radlus? 28. What is the rfifference between arr-bend and bottomrng dtes? Which IS more flexible? Which produces more reproduwble bends? 29. Whai is the pnmary benefit ofmcorporatlng a coinmg action in bottom bznd~ng? The primary negatlve feature? 30 What type ofprbduots are produced by roll bending? 31 What is the role of the form Moek in draw bendmg and cornpressim bending? 32. How can we prevent flattening or wstnkhng when benbng a tube? 33. What type of product geometry can be produwd by cold-roll form1ng7ISthe process appropriate for mahngshort lengths of spec~ahzed products7 34. What are sornemethods for stmightenmg or flattening rod or sheet9 35. What two d~snOctly different metalforming processes use the term drawzng? 36 Why a the toohng cost for a spmning operation relatively low?

37. How is shear forming different from spinning? 38. For what types of products would stretch forming b e an appropriate manufacturing technique? 39. What is the distinction between shallow drawing and deep drawing? 40. What is the function of the pressure ring o r hold-down in a deep-drawing operation? 41. Explain why thin material may be difficult to draw into a defect-free cup. 42. H o w can rediaw operations be used to produce a taller,smaller diameter cup than can be produced in a single deep-drawing operation? 43. What are draw beads, and what function do they perform? 44. Why is a trimming operation often included in adeep-drawing manufacturing sequence? 45. How does the Guerin process reduce the cost of tooling in a drawing operation? 46. How can fluid pressure or rubber tooling be used to perform bulging? 47. What is sheet hydroforming? 48. What explanation can be given for the greater formability observed during sheet hydroforming? How is this feature being used in the automotive induslry'? 49. What is the purpose of the inward movement ol the end plugs during tube hydroforming? 50. What is the difference between low-pressure tube hydroforming and high-pressure tube hydroformingin terms of both pressures and the nature of the deformations produced? 51. What are some of the basic methods that have been used to achieve the high-energy-release rates needed in the H E R F processes? 52. Why isspringback rather minimal in high-energy-rate forming?

53. What are some well-known products that have by processes including ironing? By embossing 54. What material and process conditions are r superplastic forming? 55. What is the major limitation of the supcrpla sheet metal? What are some of the more altr; 56. What properties from a uniaxial tensile test ca sess sheet metal formability? 57. How is the formability in biarial tension differ uniaxial tension? 58. What techniques can be used to assess "norn and "planar anisotropy" in sheet material? 59. How can strain analysis he used lo determi possible defects or failure in sheet metal coml 60. What is a forminglimit diagram? 61. Explain the key difference between the right sections of a forming limit diagram.lhese sect1 to stretching and drawing. 62. Describe two alternative methods of prod1 shaped thin products without requiring sheet tion techniques. 63. What two hot-forming operations can he used from steel strip? 64. What are the primary assets and limitations press drives? Of hydraulic drives? 65. What are some of the common types of press 66. What is the purpose of inclining o r tilting a pr 67. Describe how multiple operations are perf011 ously in a transfer press. 68. What types of products are produced on multislide machine?

Problems
1. The maximum punch force in blanking can be eslimated by the equation: force = S t L where: S = the material shear strength 6 = the sheet thickness L = the total length of sheared edge (circumference or perimeter) a. How would this number change if the rake angle is equal to a 11change across the width or diameter of the part being sheared? b. How would this number change if the rake were increased 1 change across the width or diameter? to a 3 2. Consider the various means of producing tul such as extrusion, seam welding, butt welding, piercing, and the various drawing operations, 1 vantages,limitations, and typical applications 3. Tube and sheet hydroforming have been un growth. Investigate current uses for these prt motive-and other fields. 4. What are some of the techniques for minimiz of springback in sheet-forming operations? 5. Select a forming process from either Chapter I vestigate the residual stresses that typically ac sult from that proccss.If they are considered to how could these residual stresses bereduced 01 out damaging or deteriorating the product?

1 1 Fabrication of a One-Piece Brass Flashlight Case

268, also known as yellow brass (66% copper, 34% zinc). As shown, the process lnvolves a blanklng operation, followed by tnlee deep dmws (the third of whlch IS only a partial), and then several operatlons to shape the upper segment of the flashl~ght. The interior surface will be left as-fabricated, and the brass s ~ d e walls of the case become part of the completed electrrcal clrcu~t dur~ng operation. A decorative chrome plat~ng is subsequently to the exterlor appl~ea The Copper Development Assoc~at~on's Standards Handbook cbtes a number ofelectrrcal apphcatmns for this altoy and particularly cites flashl~ghtshells, so it would appear that a good mater~al choice has been made. Among the clteda'commonfabr~cat~on processes" are blanklng, drawlng, forming and bendlng, sp~nning,and stampmg. It would appear that the material is Indeed compatlbte with fhs proposed processing activ~t~es.The ~~capaclty for being cold is rated as excellent, as issuitablity for soldermg and braztng. In the annealed conditron, the y~eld strength vanes from 14 fo 22 ks~, depend~ng on gram SIB, whlte the companion tensile strength is 46 to 53 ksi and percent elongat~on

II

The sheet metal fabrication sequence for a one-piece brassflashlight case. and dcitails of time, temperature, cooling condit~ons, so on. Would some form of protective atmosphere be required? If so, what would you recommend? 5 Since this material is capabre of slgnlflcant cold-work strengthening, would it be advantageous to market your flashlrght in its cold-worked cond1tion7 What would be the advantages and disadvantages of this optlon? If you feel it is advantageous, what would be your desired final condition? 6 Yellow brass IS qulte susceptible to stress-corroslon cracking. What could you do to mlnlmlze thls mode of faliure? 7. alternative maferlals have been proposed for this appl~catlon Prov~de a brief evaluation of each, cons~der~ng both fabncatron and use a. 5000 serles aluminum sheet (the matenal often used in car bodies)-does not respondto age harden~ng b. 6061 aluminum sheet (the materlal used ln canm skmsf--can be age hardened c. 1008 steel sheet that has been copper plated on both sides (ins& for eleetr~cal c~nductivity and outside as a base for chrome plate)

Yield Tenslle Strength % Elongation Strengm(ks0 fksl) 54.0 43 Quarter hard 40 0

1. What properttes make this an appropriate material for this appl~catlon? 2 What features would you want to specrfy when pur~hawng your startlng material? Cons~der temper, grain w e , surface condltton, edge conditron, and so on 3. Discuss any signlflcant features or concerns with regard to the btanklng, flet deep draw, rdraw, and second partial redraw operatlons 4 It Is 11kely that some form of intermediate (and posslbly final) anneal will be requlred Prescr~be a su~tabte procedure for this material, rncludlng the necessary

POWDER METALLURGY
18.9 H~T-ISOSTATIC P ~ ~ s s ~ G 18.15 POWDER ~TALLURGY 18.1 INTRODUCTION 1%.1,0 OTHER TECHNIQUESTD PRODUCTS 18.2 THEBASIC PROCESS 18.3 POWDER MANUFACTURE , ,, , PRQ~~UCE ~~~~~DENSITY Ply 18.16 ADVANTAGES AND 18.4 RAPIDLY SOLIDIFLE~ POWDER ~ , pRmum ~ M ~ ~ R ~ [~ ~ .* 1 8 ~. : 1 1 :~ 'Mmm ~ IA NJE , ~ I,O MOLDINO N ~ ~ ~ ~

18.5 POWDERTESTING AND EVALUATION

18.7
~CCIMPACTING
,,
,

18.8 SINTERING

18.1 INTRODUCTION

duced. Major areas of application tend to be those for which the PIM process hass

for any part where the geometry and production quantity are appropriate.

lions, platinum ingots, lead printing type, and tungsten wires, the primary material for

A period o f rapid technological development occurred alter World War 11, primarily on automotive applications, and il-on and slecl replaced copper as the nant PIM material. Aerospace and nuclear developments created accelerated de ior refractory and reactive metals. materials for which powder processing is attractive. Full-density products emcrged in the 1960s. and high-performance su loy components. such as aircraft turbinc engine parts, wcre a highlight of the 1 Developmenls in the 1980s and 199Os includcd the commercialization of rapidly ified and amorphous powders as wcll as PIM injection molding icchnology. Reccnt years havc been ones of rapid growth lor the PIM industry. From 1 1980; the consumption of iron powder increased 10-fold.A similar increase oc bctween 1980 and 1990, and the exponential growth continued through the 1990s. most products are still under 50 mm (2 in.) in size, some have been produced with w up to 45 kg (I00 lb) with linear dimensions up to 500 mm ('20 in.).

SecrroN 18.3 Powder Manufacture

461

Automotwe applications now account for nearly 70% of the powder metallurgy market. In 1990,the average U.S automobile contained about 10 kg (21 lb) of PIM parts. By 1995, the amount had Increased to over 13.6 kg (30 1b)-then to 16.3 kg (36 lb) by 2000 and to 19.5 kg (nearly 45 lb) by 2005. Some Ford,Chrysler,and General Motors V-8 engines contam between 27 and 33 pounds of PIM parts. A Chrysler V-6 engine contains 88 P/M parts with a total weight of 27 pounds. A high percentage of connecting rods are now produced by powder metallurgy. Automatic transmissions often contaln between 15 and 25 pou~lds of PIM components Other aleas where powder metallurgy products are used extensively include household appliances, recreational equpment, hand and power tools, hardware items, offlce equipment, industrial motors, and hydraulics. Areas of rapid growth Include aerospace applications, advanced composites, electronic components, magnetic materials, metalworking tools, and a variety of biomedical and dental applications Iron and low-alloy copper and copper-based powders steels now account for 85% of all P/M usage, w ~ t h comprising about 7%. Stainless steel, high-strength and h~gh-alloy stcels, and aluminum and alumlnum alloys are other high-volume matenals.T~tanium, magneslum,refractory metals, particulate composlles, and intermetall~cs are seeing increased use.

The powder metallurgy process generally conslsts of four basic steps: (1) powder manufacture, (2) m k n g or blending, (3) compacting, and (4) sintering Compact~on is generaUy performed at room temperature, and the elevated-temperature process of 5lntenng is usually conducted at atmospher~c pressure. Optlonal seconda~y processing often follows to obta~n special properties or enhanced precision. Figure 18-1 presents a simpllfied flow chart of the conventional die-compaction P/M process

JFACTURE
The properhes of powder metallurgy products are h~ghly dependent on the characteristics of the starting powders. Some important properties and characteristics include chemistry and purzty, particle size, scze distrcbuhon,partlcle shape, and the surface texture of the particles. Several processes can be used to produce powdered material, with each imparting distinct properties and characteristics to the powder and hence to the final product. Over 80% of all commercial powder is produced by some form of melt atomizutron, where liquid material is fragmented into small droplets that cool and sohdify into particles. Various methods have been used to form the droplets, several of which are Elemental or alloy metal powders
Addtt~ves
(lubricants or blndecsl
I

Q
Blending

+
Compaction Optlonal secondary
manufacturing

Optjonal secondary

FIGURE 18-1 Sirnpl~fied flow chart of the basic powder metallurgy process.

462

CHAITFR 18 Powder Metallurgy

Atomizing gas or

FIGURE 18-2 Two methods for producing metal powders. (a) melt atomizat~on and (b) atomization from a rotating consumable electrode.

tures as the velocity and media of the atom~zingjets or the speed of electrode r the starting temperature of the hquid (which affects the time that surface tension on the individual droplets pnor to solidiiicat~on), and the environment provi cooling. When wohng is slow (such as in gas atomization) and surface tension smooth-surface spheres can form before solidification.Wlth the more rapid co water atomization, irregular shapes tend to be produced. Other methods of powder manufacture include: 1 . Chemical reduction of particulate compounds (generally crushed 0x1 A large amount of iron powder IS produced by reducing iron ore or roll The resulmg powders are usually irregular in shape and spongy in texture. 2. Electrolytic deposztzon from solutions or fused salts with process condit~ons the production of a powdery deposit that does not adhere to the cathode. 3. Pulverization or grinding of brittle materials (comminution). 4. Thermal de~omposltion ofparficulate hydnde~ or carbonyls. Iron and nickel are produced by carbonyl decomposit~on, resulting in small, spherical parti 5. Preciprtationfrom solution. 6. Condensatzun ofmetal vapor&.

methods. Some methods can produce only elemental powder (often of high

be produced that are not possible with conventional melhng.All of the powders undergo furtber operations,such as drying or heat treatment, prior to further pr

SECIION 18.6

Powder M ~ x ~and n g Blend~ng

463

,4 RAPIDLY SOLIDIFIED POWDER AND AMORPHOUS) ~MICROCRYSTALL~NE

Increasing the cooling rate of an atomized liquld can result in the formation of an ultra-fine or microcrystalline grain slze i n these materials, a large percentage of the atoms are located in grain boundary reglons, glving unusual properties (such as h g h diffuslvity), expanded alloy possibilities, and good formability. If the coohng rate approaches or exceeds 106oCisec,metals can soliddy without becoming crystalline.These amorphous or glassy metals can also exbibit unusual or unique properties, wh~ch include high strength, improved corrosion resistance, and reduced energy to induce and reverse a magnetization. Amorphous metal transformer cores lose 60 to 70% less energy in magnetization than conventional silicon steels. As a result, it 1s estimated that over half of all new power distribution transformers purchased m the United States wiil utihze amorphous metal cores. Production of amorphous material, however, requires immensely high coohng rates and hence ultra-small dimensions Atomization with rapid cooling and the "splat quenchmg" of a metal stream onto a cool surface to produce a continuous ribbon are two prominent methods Slnce much of the rlbbon material is further fragmented Into powder, powder metallurgy is the primary means of fabricating useful products from amorphous material.

.5 POWDER TESTING AND EVALUATION

Key propertles of powdered material mclude bulk chemistry, surface chemistry, particle size and s u e distribution, particle shape, surface texture, and internal structure. In addit~on, powders should also be evaluated for their suitability for further processing Flow rate measures the ease by whlch powder can be fed and dwtributed into a die.Poor flow characteristics can result in nonuniform d ~ filhng e as well as nonuniform dens~ty and properties in a final product. Associated with the flow characteristics 1s the upparent densrty, a measure of a powder's ability to fill available space without the application of external pre6sure.A low apparent denstty means that there is a large fractlon of unfilled space in the loose-fill powder. Compresszbzhty tests evaluate the effectiveness of applied pressure in raising the density of the powder, and green strength is used to descr~be the strength of the pressed powder immediately after compactmg. It is well established that hlgher product density correlates w t h superior mechanical properties, such as strength and fracture resistance. Good green strength 1s required to maintaln smooth surfaces, sharp comers, and intricate details during ejectlon from the compacting d ~ or e tooling and the subsequent transfer to the slntering operation. The overall objective is often to achleve a useful balance of the key propertles. The smooth-surface spheres produced by gas atomizahon,for example, tend to pour and flow well, but the compacts have extremely low green strength, disintegratmg easily durmg handling.The uregular particles of water-atomued powder have better compress~bility and green strength but poorer flow characteristics.The sponge iron powders produced by chemical reduction of iron oxide are extremely porous and have highly irregular, extremely rough surfaces.They have poor flow cha~acter~stlcs and low compacted density, but green strength is quite high.Thus, the same material can have wldely different performance characteristics, depending upon the specifics of powder manufacture

It is rare that a single powder will possess all of the characteristics des~red in a

given process and prodnct.Most likely, the startlng material will be a mlxtureof various grades or sizes of powder, or powders of different composit1on5,along wlth additions of lubricants or bmders.

464

detallurgy CHAPTER 18 Powder 1

In powder products, the final chemistry is often obtained by combining pure or nonmetal powders, rather than starting with prealloyed mater form chemistry and stsucture in the final product, therefore, sufficient diffusion occur during the sintering operation. Unique composites can als the distribution of an immiscible reinforcement material in a mat of metals and nonmetals in a single product such as a tungsten carbide-cobalt ma ting tool for high-temperature service. Some powders such as graphite, can even play a dual role, during compaction and a source of carbon as it alloys with iron during sintering t duce steel. Lubricants such as graphite or stearic acid improve the flow charact and compressibility a t the expense of reduced green strength. Binders produce the verse effect. Since most lubricant8 or binders are not wanted in the final product, th are removed (volatilized or burned off) in the early stages of sintering, leaving hoi that are reduced in size or closed during subsequent heating. Blending or mix& operations can be done either dry or wet, where water or o solvent is used to enhance particle mobility, reduce dust formation, and lessen explo hazards. Large lots of powder can be homogenized with respect to both chemistry distribution of components,sizeg and shapes. Quantities up to 16,000 kg (35,000 lb)h been blended in single lots t o ensure uniform behavior during processing and production of a consistent product.

One of the most critical steps in the PIM process is compuction. pressed and densified into a shape known as a green compact, usually at room te ature. High product density and the uniformity of that density throughout the co are generally desired characteristics. In addition, the mechanical interlocking an welding of the particles should provide sufficient green strength for in-process han and transport to the sintering furnace. Most compacting is done with mechanical presses and rigid t and hybrid (combinations of mechanical, hydraulic, and pneumatic) presses canal used. Figure 18-3 shows a typical mechanical press for compacting powders and movable set of compaction tooling. The removable die sets allow the time-consu alignment and synchronization of tool movements to be set up while the pr producing parts with another die set. Compacting pressures ge 3 and 120 tons/in2. depending on material and application (see Ta range of 10 to 50 tons/inz.being the most common. While most PIM presses have capacities of less than 100 tons, increasing numbers are being purchased with hi capacity. Because of pressures and press capacity, powder metallnrg limited to pressing areas of less than 10 square inches, but larger parts have bec more common. Some P/M presses now have capacities up to 3000 tons and are cap of pressing areas up to 100 square inches. When even larger produ paction can he ptrlormed by dynamic methods, such as use of an shock wave. Metalforming processes, such as rolling, forging, cxt have also been adapted for powder compaction.

Compaction Applicalion
Porous metals and filtcrs Rcfracli,~ymclals and carhidcs

Pressures

Porous bearings Machine parls (medium-density imn & r e e l )

High-density copper and alulninuln parts High-dcnhity iron and slccl parts

3-5 5-15 10-25 20-50 18-20 5&120

70-2011 14h-350 275-690 250-275

Figure 18-4 shows the typical compaction sequence for a mechanical press. W~th the bottom punch in its fully raised position, a feed shoe moves into position over the die.The feed shoe is an inverted container filled with powder, connected to a largepowder container by a flexible feed tube. Wlth the feed shoe in position, the bottom punch descends to a preset fill depth, and the shoe retracts, with its edges leveling the powder. The upper punch then descends and compacts the powder as i t penetrates the die.The upper punch retracts and the bottom punch then rises to eject the green c0mpact.A~ the die shoe advances for the next cycle,its forward edge clears the compacted product from the press, and the cycle repeats During uniaxial or one-direction compaction, the powder particles move pnmarily in the direction of the applied force. Since the loose-fill dimensions are 2 to 2.5 times the pressed dimens~ons, the amount of parhcle travelin the pressmg hrection can be substantial.The amount of lateral flow, however,is quite hmlted. In fact, it is rare to find a particle in the compacted product that has moved more than three particle diameters off of its original axls of pressmg.Thus, the powder does not flow hke a hquid; it simply

2 Charg~ng
(filling) die wlth powder

Compaction beg~ns

4 Compaction

5 Election

6 Recharging

completed

of part

d~e

compaction sequence for a s~ngle-level part, showlng the functions o f the feed shoe, die, J lower punches Loose powder 1 s shaded, compacted powder 1 s sol~d black.

FIGURE 18-5 Compactton w~th a stngle movtng punch, showtng the resultant (shaded), nonuniform dens~ty highest where particle movement is the greatest

FIGURE 18-6 Density

dtstribut~on obtalned with a double-acting press and two movlng punches. Note the Increased un~formity compared to Figure 18-5.Thicker parts can be effectively compacted.

an equal and opposing force is created.This opposlng force ispr compresses unt~l a combmation of (1) resistance by the bottom punch and (2) friction betwe particles and the dle surfaces. Densificatlon occurs by particle movement as plastic deformation of the individual particles. As illustrated in Figure 18-5, when the pressure is appl~ed by only one punch lmum dens~ty occurs below the punch and decreases as one moves down the col is very difficult to transmit umform pressures and produce uniform density thro a compact, especially when the thickness is 1arge.B~ use ot a double-action press, produ pressmg movements occur from both top and bottom (Figure 18-6), th~cker be compacted to a more unlform density. Since s~de-wall frlct~on is a key factor I pactlon, the resultmg density shows a strong dependence on both the thicknes width of the part being pressed. For uniform compaction, the ratio of th~cknes should be kept below 2.0 whenever possible When the ratlo exceeds 2.0, thepr tend to exhibit considerable variation in density. As shown in Figure 18-7, the average density of the compact depends amount of pressure that 1s applied, with the specific response being strongly dep on the characteristics of the powder being compressed (its slze, shape, surface t mechamcal properties, etc.) The final density may be reported as e ~ t h e an r density m units such as grams per cub~ccentimeter or as a percentage of thep or theoretical density.The difference between this percentage and 100% corres the amount of void space rema~ning within the compact, F~gure 18-8 shows that a s~ngle displacement will produce different de therefore, for compactlon in d~fferent thcknesses of powder It 1s imposs~ble, punch to produce uniform dens~ty in a multithickness part. When more than one ness is requrred, more complicated presses or compaction methods must be em F~gure 18-9 ~llustrates two methods of compacting a dual-thickness part. By pr different amaunts of motion to the various punches and syncbro~nz~ng these mo to provide s~multaneous compaclion, a uniform-dens~ty product can be produce Since the complexity of the part d~ctates the complex~tyof equipment, metallurgy components have been grouped into classes. Class 1components are plest and easlcst to compact.They are thin, smgle-level parts that can be presse force from one direction.The thickness is generallv less than inch (6.35 mm).

Pressure, MPa

FIGURE 18-7 Effect of

compacting pressure on green denstty (the density after compaction but before sintering). Separate curves are for several commercial powders.

Pressure, ksi

18-8 Compact~on of a

314 original volume or 1-113 x original

density pper punch. Note the ~fference In compacted Initial conditions

(a)

112 original volume or 2 X original density

After Compaction

(b)

FIGURE 18-9 Two methods of compacting a double-thickness part to near-un~form dens~tv.Both involve the

Press Actions
Single Double Double Double o r multiple

paction. Productic several hundred p

Warin camp,
preheating the pov better to the a p p l i ~

Class 1

Class 3

Class 2

class4

468

CHAPTER 18 Powder Metallurgy

Compaction can also be enhanced by increasing the amount of lubricant powder.This reduces the friction between the powder and the die wall, as we proving the transmission of pressure through the powder. If too much lubricant however, the green strength may be reduced to the point where it is insufficient ejectionand handling. or the final properties may become unacceptable. While pressing rates vary widely, small mechanical presses can typically e u p to 100 pieces per minute. By means of bulk movement of particles, deform individual particles, and particle fracture or fragmentation, mechanical compac raise the density of loose powder to about 80% of an equivalent cast or forge Sufficient strength can be imparted to retain the shape and permit a reasonable a of careful handling. In.addition, the compaction process sets both the nature and bution of the porosity remaining in the product. Because powder particles tend to be somewhat abrasive and high pressures volved during compaction, wear of the tool components is a major concern. quently, compaction tools are usually made of hardened tool steel. For parti abrasive powders, or for high-volume production, cemented carbides may be em Die surfaces should be highly polished and the dies should be heavy enough towit the high pressing pressures. Lubricants are also used to reduce die wear.

In the sintering operation. the pressed-powder compacts are heated in a contr mosphere to a temperature below the melting point but high enough to permit s diiiusion, and held lor sufficient time to permit bonding of the particles. Most met sintered at temperatures of 70 to 80% of their melting point. while certain refra materials may rcquirc tcmperatures near 90%. Table 18-3 presents a summary ofs common sintering temperatures.When the product is composed or more than one mat the sintering temperature may be above the melting temperature of one or more ponents:fie lower-melting-point materials then melt and flow into the voids betwee remaining particles, and the process becomes liyuid-phase sintering. Most sintering operations involve three stages, and Illany sintcring furnaces three corresponding zones. The firs/ operalion. the burn-uj~orpnrge, is designed t bust any air, volatilize and remove lubricants or binders that would interfere wit bonding. and slowly raise the temperature of the compacts in a controlled manner. heating would produce high internal pressure from air entrapped in closed po volatilizing lubricants. and would result in swclling or fracture oc the compacts.Whe compacts contain appreciable quantities of volatile materials, their removal create ditional porosity and permeability within the pressed shape. Thc manufacture of p ucts such as metal filters is designed to take advantage of this feature. When the prod are load-hearing components, however, high amounts of porosity are undesirable the amount of volatilizing lubricalit is kcpt to an optimized minimum. The secon high-iem~iernture,vtage is where the desired solid-state diffusion and bonding be the powder particles take p1ace.A~ the material seeks to lower its surface energy,a move toward the points of contact between the parlicles.The areas of contact bec larger. and the part becomes a solid mass with small pores of various sizes and sh The mechanical bonds of compaction become true metallurgical bonds.The timein stage must be sufficient to produce the desired density and final properties,usuallyv ing from 10 minutes to several hours. Finally, a cooling period is required to lower temperature of the products while retaining them in a controlled atmosphere.This turc serves to prevent oxidation that would occur upon direct discharge into air as as possible thermal shock from rapid cooling. Both batch and continuous furnaces used for sintel-ing. All thrcc stages o l sintering must be conducted in the oxygen-frcc conditions vnculrm or protective atmo.~phere.This is critical because the compacted shapes ty ly have 10 to 25% residual porosity, and some of the internal voids are connected posed surfaces. At elevated tcmperatures, rapid oxidation would occur and signific impair the quality of interparticle bonding. Reducing ntmospheres, com~nonly base

S~ntemng Temperature Metal

Alum~num alloys Brass Copper Ironistee1 Stamless steel Cemented carb~des Molybdenum Tungsten Var~ous cerarmcs
"C
590-620 850-950 750-1000 1100-1200 1200-1280 1350-1450 1600-1700 220&2300 14UO-2100

"F
1095-1 150 1550-1750 1400-1850 2000-2200 2200-2350 245i1-2650 2900-3100 40004200 2550-3800

hydrogen, d~ssociated ammonia, or cracked hydrocarbons, are preferred since they can reduce any ox~de already present on the particle surfaces and combust harmful gases that are liberated during the sinterlng. Inertgases cannot reduce exlsting oxides but WIU prevent the formation of any additional contaminants. Vacuum sinterlng 1s frequently employed with stainless steel, titanium, and the refractory metals Nttrogen atmospheres are also common. Durlng the smtering operation, a number of changes occur m the compact Metallurgical bondsfom between the powder particles as a result of solid-state atomic diffusion, and strength, ductility, toughness, and electrical and thermal conductivities all Increase. If different chemistry powders were blended, interdiffus~on promotes the formatron of alloys or lntermetall~c phases As the pores reduce in size, there will be a concurrent increase in denslty and contraction in product dimensions. To meet final tolerances, the dimensional shrinkage wll have to be compensated through the des~gn of oversized compaction d~es. During smtering, not ail of the porosity is removed, however. Conventional pressed-and-smtered PIM products generally contain between 5 and 25% residual porosity. Slnter brazzug IS a process m wh~ch two or more separate pieces are jo~ned by brazing while they are also belng smtered. The individual pieces are compacted separately, and are assembled wlth the braze metal posltloned so it wlll flow into the joint. When the assembly is heated for sintering, the braze metal melts and flows between the joint surfaces to create the bond. As sintering continues, much of the braze metal diffuses into the surrounding metal, producing a final joint that is often stronger than the materials being joined.

In conventional press-and-sinterpowder metallurgy, the pressing or Wmpachon is usually performed at room temperature and the sintering, at atmospheric pressure. Hot-isostahc pressing (HIP) combmes powder compaction and slntering into a single operation that involves gas-pressure squeezing at elevated temperature. While t h ~ may s seem to be an ~mprovement over the two-step approach, it should be noted that heated powders may need to be "protected or isolated from harmful envrronments, and the pressurizing m e d ~ a must be prevented from entering the voids between the particles. One approach to hot-isostatic pressing begins by seahng the powder in a flexible, airtight, evacuated contamer, which is then subjected to a hlgh-temperature, high-pressure environment. Conditions for processing irons and steels involve pressures around 10,000 to 15,000psi (70 to 100 MPa) coupled with temperatures in the ne~ghborhood of 1250C (2300F) For the n~ckel-based superalloys, refractory metals, and ceramic powders, the equipment must he capable o f 45,000 p s ~ (310 MPa) and 1500C (2750F). Multlple pleces, totaling up to several tons, can now be processed in a single cycle that typ~cally lasts several hours.

After processing, the products emerge at full density with unifo properties that are often superior to those of other processes Near-n possible, thereby reducing material waste and costly machining operations the powder is totally isolated and compaction and sintering occur simultaneous process is attractive for reactive or brittle materials, such as berylli zirconium, and titanium. Since die compaction is not required, large parts are sible, and shapes can be produced that would be impossible to eject from r' pacbon dies. Hot-isostatic pressing has also been employed to densify (such as those that have been conventionally pressed and sintered), heal inte ity in castings, and seal internal cracks in a variety of products. The elimina ducbon of defects yields startling improvements in strength, toughness, fatigue resis and creep life. Several aspects of the HIP process make it expensive and unattractive for volume production. The first is the high cost of cunnzng the powder in a ing medium that can reslst the subsequent temperatures and press removlng this materlal from the product (decanning). Sheet metal common, but glass and even ceramic molds have been used. The seco valves the relatively long tlme for the HIP cycle. While process advances have cycle times from 24 hours to 6 to 8hour&production is still l m t e d to several loa and the number of parts per load is limited by the ability to produce and ma uniform temperature throughout the pressure chamber. The smter-HIP process and pressure-assisted slntercng are techniques tha been developed to produce full-density powder products without the ning and decanning. Conventionally compacted PIM parts are placed in a pressu chamber and sintered (heated) under vacuum for a time that is sufficient to s surface and isolate all internal porosity. (Note:This generally requires achieving sity greater than 92 to 95%.) Whlle mamtaining the elevated temperature, thev is broken and h g h pressure is then applied for the remainder of the process.The surface produced during the vacuum slnterlng acts as an isolating can pressure stage. Since these processes start with as-compacted pow and cooling cycle that would be required inate the additional heat~ng sintered in the conventional manner and then sub~ected to the HIP process for , densification.

TECHNIQUES TO PRODUCE HIGH-DENSITY

RODUCTS
The high-temperature metal deformation processes can also be used to prod density PIMparts. Sheets of sintered powder (produced by roll c ing) can be reduced in thickness and further densified by hot r depicted in Figure 18-11. Rods, wires, and small billets can be produc trusion of encapsulated powder or pressed-and-sintered slugs.Forging can be a form complex shapes from canned powder or simple-shaped sintered preforms powdered material, these processes offer the comblned benefits of p and the respective forming process, such as the production of fabricated shap uniform fine grain size,uniform chemistry, or unusual alloy composition

Hot

sheet

i '

produc~ng continuous she from powdered feedstock.

SEGTION 18.11

Metal Injection Mold~ng (MIM) or Powder Injection Moldlng (PIM)

471

The Ceracon process is another method of raising the density of conventional pressed-and-smtered PIM products wthout requrmg encapsulation or canning.A heated preform 1s surrounded by hot granular material, usually smooth-surface ceramlc particles. When the assembly is then compacted in a convent~onal hydraulic press, the granular material transmits a somewhat uniform pressure. Encapsulation is not requred slnce the pressuriz~ng medium 1s not capable of entenug pores m the material.When the pressure cycle 1s complete,the part and the pressurizing mediumseparatefreely, and the pressure-transmithng granules are reheated and reused. Yet another means of producing a hi&-density shape from fine particles is in situ compactron 01 spray formlng (also known as the Osprey proces~). Consider an atomuer similar to that of Figure 18-2a,in whch lets of inert or harmless gas (nitrogen or carbon droxide) propel molten droplets down into a collectmg container. If the droplets solidify before impact, the container fills with loose powder. IL the droplets remain liquid during their flight, the container fills with molten metal, which then solidifies into a convent~onal casting. However, if the cooling of the droplets IS controlled so that they are semisolid (and computers can provide the necessary process control), they act as "slush balls" and flatten upon impact.The remaining freezing occurs quickly, and the resultant product 1s a uniform chemislry, fine grain size, high-density (in excess of 98%) solid. Depending on the shape of the collect~ng conlainer, the spray-formed product can be a finished part, a strip or plate, a deposited coating, or a preform for subsequenf operations, such as forging. Both ferrous and nonferrous products can be produced with deposit~on rates as h ~ g h as 200 k~lograms (400 pounds) per minute. Un~que composites can be produced by the simultaneous deposition of two or more material$ rnjecting secondary particles into the stream and promoting m-stream reactions.

.11 METAL INJECTION MOLDING (MIM) OR POWDER INJECTION MOLDING (PIM)

For many years, injection molding has been used to produce small, complex-shaped compofients from p1astic.A thermoplastic resin a heated to Impart the necessary degree of fluidity and is then pressure ln~ected into a die, where it cools and hardens Die castIng is a sim~lar process for metals but is restricted to alloys with relatively low melting temperature, such as lead-, zmc-, alummum-,and copper-based materials. Small,complexshaped products of the higher-melting-point metals are generally made by more costly processes, which include investment casting, machining directly from metal stock, or convent~onal powder metallurgy. Metal mnjecuon molding (MZM), also calledpowder injection moldrng (PZM), is a rather recent extemion of convent~onal powder metallurgy that combines the shape-forming capablhty of plastics, the precision of die casting, and the materials flexibility of powder metallurgy Since powdered material does not flow like a fluid, complex shapes are produced by first combining ultra-fine (usually in the range of 3 to 20pm) spherical-shaped metal, ceramic, or carbide powder with a low-molecular-weight thermoplastic or wax matena1 in a mix that is typically 60% powder by volume This mixture IS frequently produced in the form of pellets or granules, which become the feedstock for the ~njection process. After heatmg to a pastelike consistency (about 260C or 500F), the mater~al is injected into a heated mold cavity under sufficient pressure (about 10,000 ps~) to ensure die f11ling.Aftercooling and ejection, the binder materlal a removed by one of a variety of processes that ~nclude solvent extraction, controlled heating to above the volatilizat~on temperature, or heattng in the presence of a catalyst that breaks the bmder down into removable products. Removing the binder is currently the most expenslve and timeconsuming part of the process. Heatmg rates. temperatures, and debtnding tlmes must be carefully controlled and adjusted for part thickness.The parts then undergo conventional mntenng, where any remaining binder 1s first removed, and the diffusion processes then set the final properties of the product. During sintering, MIM parts typically shrink 15 to 25%, and the density increases from about 60% up to as much as 99% of ideal. (Note. Since MIM parts are molded w~thout density variations, the subsequent shrinkage tends to be both uniform and repeatable.) Secondary processes may take the

F~ne powder
I

B~nder
I

e
M~xlng and pellet~zjng

Optional

FIGURE 18-12 Flow chart the metal inlectron moldmg process (MIM) used to produ small, intricate-shaped parts1 metal powder
I

form of surface cleaning or finishing,plating, machining, or heat treatlng.The high density enable3 the secondary processes to be conducted in the same manner aE wrought products Flgure 18-12 summarizes the full sequence of activities, and Table: provides a summary comparison of conventmnal powder metallurgy and MIM While the size of conventional PIM products is generally llmited by press cat ty, the slze of PIM injection moldings is more limited by economics (cost of tha: powders) and blnder removal. The best candidates for PIM injection ma complex-shaped parts with thicknesses of less than inch, weights unde (20 grams), and made from a metal that cannot be economically dle cast. compete with and frequently replace machined components or Investment cas ' shapes are generally too complex to compact by conventional powder metaU the injection moldrng can often reduce or eliminate costly mach~ningSection nesses as small as 0.010 Inch are possible because o f the fineness of the powder. era1 rule, the smaller the part and the greater the complexrty, the more likely be an attractive alternative to machining, castlng, stamping, cold forming, or a1 powder metallurgy F~gure 18-13 shows a variety of MIM products Medium to large production volumes (more than 2000 to 5000 identical generally required to justify the cost of dle design and manufacture. The r high final density (95 to 99% compared to 75 to 90% for conventional P M p a uniformity of that density, the close tolerances (0.3 t o 0.5%), and excellent finish (about 125 p In.) all combine to make the process attractive for man tions Parts can be made from a wide selection of metal alloys, ~ncludmg s less steel, tool steel, brass, copper, titanium, tungsten, nickel-based s ceramics, and many spec~alty materials. The final properties are superior to conventional powder metallurgy and are generally close to those of wrough equivalents.

Fartide size Particle respOlise PnfosRy nonmetal) Amount of binderliubricant Homo~eneilp of green part Final sintercd density

20-250 pm Deforms plasticdly

10-20% 0.54%

a l lpm

Wnileformed
3MO%

h'onhomog~eous

<92%

30-40% Hamogeneous >96%

.
,

. . , -. , - ~,..
I..

SECTTON 18.12 Secondary Operations

473

tction
>arts.

lARY

OPERATIONS
Powder metallurgy products are often ready to use when they emerge from the sintermg furnace. Many P/M products, however, utilize one or more secondary operations to provide enhanced precision, improved properties, or special characteristics. During sinteriug,product dimensionsshrink due to denshation. In addition, warpingor dsstortion may occur duringnonuniform cool-down from elevated temperatures. As a mult,a second pressing operation, known as repuesscng, cointng,or srzing,may be required to restore or improve dimensional precision.The part is placed in a die and subjected to pressures equal to or greater than tbe initial presstng pressure.A small amount of plastic flow takes place, resulting in high dimensional accuracy, sharp detail, and Improved surface finish. The associated cold working and Increase in part density may combine to increase part strength by25 to 50%. (Note:Bdcause of the shrmkage that occurs during sintering, repressing cannot be performed with the Same set of tooling that was used for the original powder compaction.) If massive metal deformation takes place in the second pressing, the operation is known as P/M furg~ng. Conventronal powder metallurgy is used to produce a preform, which is one forging operation removed from the finished shape.The normal forging sequence of billet or bloom production, shearing, reheating and sequential deformation is replaced by the manufacture of a comparatively simple-shaped powder metallurgy preform followed by a single hot-forgingoperation.The forging stage produces a more complex shape, adds precisian, provides the benefits of metal flow, and increases the density (ofren up to 99%).The increase in density is acmmpamed by a significant improvement in mechanical properties. While proteetive atmospheres or coating$ are required to prevent oxidation of the powder preform during heat~ng and hot forging, the PIM p r o m s can often provide a sigrificant reduction in scrap or waste. (By controlling preform weight to withim 0.5%,flash-free forgmg can otten be performed ) Forged products can benefit from the improved properties of powder metallurgy, such as the absence of segregation, the uniform fine grain size, and the use of novel alloys or unique

474

CHAPILK 18 Powder Metallurgy composites. The conventional powder metallurgy process can be expanded to size products w ~ t h increased complexity. The tolerance requirements 01 cams, and gears can often be met without subsequent machining Figure 18-14 ~llustr reduction in scrap by comparing the same part made by conventional forging a PIM forge approach. PIM forged connecting rods, l ~ k those e shown in Figure 18-1 rently account for more than 60% of all connecting rods used in North Ameri are typical of the high-volume steel parts currently bang produced. Impregnation and infiltration are secondary processes that utillze the m n t nected porosity or permeability of low-density PIN products Impregnation refer forcing of oil or other liquid,snch as a polymeric resin,into the porous network. be done by immersmg the part in a bath and apply~ng pressure or by a combinat uum-pressure process.The most common application is that of oil-~mpregnated b After imprrgnat~on, the bearing material will contam from 10 to 40% oil by which wll provide lubr~cation over an extended hfetime of operation. In a simil ner,PIM parts can be impregnated with fluorocarbon resin (such asTeflon) top products offering a combinat~on of high strength and low trict~on. When the presence of pores is undeslrable,PlM products may be subjected t mnfiltratron. In this process a molten rnetal or alloy with a meltingpomt lower than conshtuent flows into the mterconnected pores of the product under pressure lary action. Steel parts are often infiltrated mth copper,for example After infi engineering properbes such as strength and toughness are lmproved to a level w are generally comparable to those of solid meld1 products Infiltration canals to seal pores prior to plating, improve machinability or corrosion resistance, or components gas- or liquid-tight.Additional heat~ng after infiltration can cause inl slon between the infiltrant and base metal, further enhancing mechan~cal propert'

Comparison of conventional forging and the forging of a powder metallurgy preform to produce a gear blank (or gear). Moving left to right, the top sequence shows the sheared stock, upset section, forged blank, and exterior and interior scrap associated with conventional forging. The finished gear is generally machined from the blank with additional generation of scrap. The bottom pieces are the powder metallurgy preform and forged gear produced entirely without scrap by P/M forging. (Courtesy of CKN Sinter Metals, Auburn HiNs, MI.)
FIGURE 18-14

FIGURE 18-15

PIM forged connecting rods have been produced by the m~ll~ons. (Courtesy of Metal Powder Industries Federabon,

Princeton, N / )

Powder metallurgy products can also be subjected to the conventional finishing operations of heat treatment, machining, andsurface treatment. If the part is of high density ( ~ 1 0 % porosity) or has been metal impregnated,conventional processing can often be employed. Special precautions must be taken, however, when processing low-denslty PIM products. During heat treatment, protective atmospheres must again be used and certain liquid quenchants should be avoided Speeds and feeds must be adjusted when machining, and care should be taken to avoid pickup of lubricant or coolant In general, PIM products shouldbe machined using sharp tools,light cuts, and hlgh feed rates. When a large amount of machining is required, special maclunability-enhancing additions may be incorporated into the initial powder Mend. Nearly all common methods of surface finishing can be applied to PIM products, including platings and coatings,d~ffusion treatments, surface hardening, and steam treatment (whlch is used to produce a hard corrosion-resistant oxide on ferrous parts).As with the other secondary processes,some process modifications may be required if the part has a reasonable amount of porosity or permeability. Since most parts are small and are produced in large quantity, barrel tumbling is another common means of cleaning, deburring, and surface modification.

.13 PROPERTIES OF P/M PRODUCTS

Because the properties of powder metallurgy products depend on so many variablestype and size of powder, amount and type of lubricant, pressing pressure,sintering temperature and t~me, finishing treatments, and so on-it is difficult to provide generahzed information. Products can range all the way from low-density, highly porous parts \nth tensile strengths as low as 10 ksi (70 MPa) up to hlgh-denslty pleces with tenslle strengths of 180 ksi (1250 MPa) or greater. As shown in Figure 18-16, most mechanical properties exhibit a strong dependence on product density, with the fracture-limited properties of toughness, ductility, and fatigue life being more sensitive than strength and hardness.The volds m the PIM part act as stress concentrators and assist in starting and propagating fractures.The yield strength of PIM products made from the weaker metals is often equivalent to the same material m wrought form. If higher-strength materials are used or the fracture-related tensile strength ts specified, the properties of the PIM product tend to fall below those of wrought equivalents by varying but usually substantial amounts. Table 18-5 shows the properties of a few powder metallurgy materials compared with those of wrought material of s ~ m ~ lcomposition. ar When larger presses or processes such as PIM forging or hot-rsostatic pressing are used to produce higher density, the strength of the PIM products approaches that of the wrought material. If the processing results in full density

Tensile strength

/
x

FN 208 Fe 2% Ni- 8C

6.6

7.0

7.2

- 0

6.6

7.0

7.2

- 0

6.6

7.0

7.2

Density g/crn3
18-16

Density g/crn3

Density glcrn3

erlytn, ten5ilr strenyth, Cnarpy ' i ~ ~ r ~enerqy dct (snown in ft- os), and percrlit e ongar011ill

Mecnanical propertle, versJr as-cinlereu nensiry lor rwo iron-nasco pouvoerr. Prupenies neplcrea ncluuc d 1 - n. gaqc cngth.

476

C H A ~ 18 R Powder Metallurgy

Materiala
Iron

Form and Composition

Wrm~bt P / M 4 9 % Fe mm PM-99% Ee min Wrought AlSI 1025 P!M-0 25% C. 99.7546 Fe Wroughl type 303 PIM iype 3U3 Wrought 2014 PIW 201 AB Wrought 6061 P ! M 6Q1 AB Wrought OFHC PIM capper

Cond~tton~
HR As smtered As slntered HR As stntered
Annealed As sintered T6
T6

Percent of lT%eoret~cal Densty

Tensile Strength

Elongatio 2 14. (Oh)

30

89 94
-

9
15 25 2

Stcel

84

Sta~nlers
sleel

50
2 20
2

Alumlnum

T6 T6 Annealed AS s~ntered

15
2 50 8

Represed

Wrought 260
PIM 70% Cu3n% Zn
%qunalent wroughl metal shown for cornpanson

Annealed As smtered
h H hor ~ roUed,Tb,age hwdened

89

37

255

18 65 26

with fine grain m e , PIM parts can actually have properties that exceed the wrou cast equivalents. Slnce the mechanical properties of powder metallurgy products dependent upon density, it is zmportanf that productr be desrgned and materials se so that the finalproperties will be achleved wzth the anticipated amount offinalp Physical properties can also be affected by poros~ty. Corrosion resistance be reduced due to the presence of entrapment pockets and fissures. Electrical, t and magnetic properties all vary with density, usually decreasing with the pres pores. Porosity actually increases the abil~ty to damp both sound and vibration,ho and many PIM parts have been des~gned to take advantage of this feature.

I B ! 18.14 DESIGN OF POWDER METALLURGY PARTS

Powder metallurgy is a manufacturing system whose ultimate objective is to ec cally produce products for specific engineering apphcations. Success begins wit des~gn and follows with good material and proper processing, In desigmng pa are to be made by powder metallurgy, it must be remembered that PIM is a specia ufacturing process and promcon should be made for its unique factors Products converted from other manufacturing processes w~thout modification m desig perform as well as parts designed spec~fically for manufacture by powder met Some basic rules for the design of PIM parts are as follows:

1. The shape of the part must permit ejection from the die. Side-wall surfaces sho
parallel to the direct~on of pressmg. Holes or recesses should have uniform cro tion with axes and side walls parallel to the direction of punch travel. The shape of the part should be such that powder is not requ~red to flow into cavities such as thin walls. narrow splines, or sharp corners. The shape of the part should permit the construction of strong tooling. The thickness of the part should be within the range for which P N parts c adequately compacted. The part should he designed with as few changes in section thickness as possi

2.

3 . 4.

5.

6. Parts can be designed to take advantage of the fact that certain forms and properties can be produced by powder metallurgy that are impossible, impractical. or uneconomical to obtain by any other method. 7. The design should be consistent with available equipment. Pressing areas should match press capability, and the number of thicknesses should be consistent with the number of available press actions. 8. Consideration should also be made for product tolerances. Higher precision and repeatability are observed for dimensions in the radial direction (set by the die) than for those in the axial or pressing direction (set by punch movement). 9. Finally, design should consider and compensate for the dimensional changes that will occur after pressing, such as the shrinkage that occurs during sintering.

The ideal powder metallurgy part, therefore, has a uniform cross section and a single thickness that is small compared to the cross-sectional width or diameter. More complex shapes are indeed possible,but it should be remembered that uniform strength and properties require uniform density. Holes that are parallel to the direction of pressing are easily accommodated. Holes at angles to this direction, however,must be made by secondary processing. Multiple-stepped diameters, reentrant holes. grooves, and undercuts should be eliminated whenever possible. Abrupt changes in section, narrow deep flutes, and internal angles without generous fillets should also be avoided. Straight serrations can be readily molded, but diamond knurls cannot. Punches should be designed to eliminate sharp points or thin sections that could easily wear or fracture. Figure 18-17 illustrates some of these design recommendations and restrictions.

No

Yes

Yes

Sharp

Fillet radius ,

Yes

No

Yes

Better
No

Yes

No

Yes

Best

(avoids sharp-cornered
punch)

features for powder metallurgy products. FIGURE 18-17 Examples of poor and good des~yn Recommendat~ons are based on ease of pressmy, des~gn of tooling, un~formlty of properties, and ultimate performance.

POWDER METALLURGY PRODUCTS

The products that are commonly produced by powder metallurgy can generally be sified Into six groups.
1. Porous or permeable productq ~ u c as h bearing\, filters, and pressure orflow r tort Oil-~mpregnated bearings, made from either iron or copper alloys, const' large volume of PIM products.They are widely used m home appliance and motive applications since they require no lubrication or maintenance during service Life. PIM filters can be made with pores of almost any size, some as sm 0.0025 mm (0 0001 tn ). Unl~ke many alternative filters, powder metallurgy filte withstand the conditions of elevated temperature, high applied stresses, and cor environments. 2. Products of complex shapes that would requzre considerable machining when accuracy and fine surface fimsh otherprocesses. Because of the dimens~onal characteristic of the PIM process, many parts require no further processing,an ers require only a small amount of finish machmmg.Tolerances can generally b to within 0.1 mm (0.005 m.). Large numbers of small gears are currently being by the powder metallurgy process. Other complex shape&such as pawls, ca small activating levers, can be made quite economically. 3. Products made from materials that are d i f f ~ u l t tmachine o ormaterials wrth cngpornts. Some of the first modern uses of powder metallurgy were the pr of tungsten lamp filaments and tungsten carbide cuttmg tools 4. Products where the combinedproperties of two or more metals (or metals a metals) are desired. This unique capability of the powder metallurgy pr applied to a number of products. In the eleetrlcal industry, copper and graphi frequently combined in applications like motor or generator brushes where c prov~des the current-carrying capaclty and graphite provides 1ubncation.Be have been made of graphite combined with iron or copper or from mixtures metals, such as tln and copper, where the harder material provides wear re and the softer material deforms in a way that better distributes the load El contacts often combine copper or silver wlth tungsten, nickel, or molybd Here, the copper or silver provides high conductivity, while the htgh meltin perature materialprovldes resistance to fus~on when the contacts experience and subsequent closure. 5. Products where the powder metallurgy process produces clearly superiorpr The development of processe? that produce full density has resulted in PI ucts that are clearly superior to those produced by cumpetlug techniques.1 of critical importance such as aerospace applicat~ons, the additional cos processing may be justified by the enhancement of properties. As another e conslder the production of PIM magnets A magnetic field can be used particles prior to sintering, resulting in a product with extremely hlgh flux 6. Products where the powder metallurgy process offers definite economic adv The process advantages described in the next section often make powder me gy the most economicalamong competing ways to produce a part.

Figure 18-18shows an array of typical powder metallurgy products.

ADVANTAGES AND DISADVANTAGES OF POWDER METALLURGY

Like all other manufacturing processes,powder metallurgy has distinct advanta disadvantages that should be considered if the techmque is to be employed econ and successfully.Among the important advantages are these: 1. Elimnut~on or reduction of machinig. The drmensional accuracy and surfac PIM products are such that subsequent machining operations can be totally e

18-16 Typical parts

for many applications. If unusual dimensional accuracy is required, simple coining or sizing operations can often give accuracies equivalent to those of most production machining. Reduced machining is especially attractive for difficult-to-machine materials. 2. High production rates.Al1 steps in the PIM process are simple and readily automated. Labor requirements are low, and product uniformity and reproducibility are among the highest in manufacturing. 3. Complex shapes can be produced. Subject to the limitations discussed previously, complex shapes can be produced,such as combination gears,cams, and internal keys. It is often possible to produce parts by powder metallurgy that cannot be economically machined or cast. 4. Wide variations in compositions are possible. Parts of very high purity can be produced. Metals and ceramics can be intimately mixed. lmmiscible materials can be combined, and solubility limits can be exceeded. Compositions are available that are virtually impossible with any other process. In most cases the chemical homogeneity of the product exceeds that of all competing techniques. 5. Wide variations in properties are available. Products can range from low-density parts with controlled permeability to high-density parts with properties that equal or exceed those of equivalent wrought counterparts. Damping of noise and vibration can be tailored into a PIM product. Magnetic properties, wear properties, friction characteristics, and others can all be designed to match the needs of a specific application. 6. Scrap is eliminuted or reduced. Powder metallurgy is the only common manufacturing process in which no material is wasted. In casting, machining, and press forming, the scrap can often exceed 50% of the starting material.This is particularly important where expensive materials are involved, and powder metallurgy may make it possible to use more costly materials without increasing the overall cost of the product. An example of such a product would be the rare earth magnets. The major disadvantages of the powder metallurgy process are these:

1. Inferior strength properties. Because of the residual porosity, powder metallurgy parts
generally have mechanical properties that are inferior to wrought or cast products of the same material. Their use may be limited when high stresses are involved. The required strength and fracture resistance, however, can often be obtained by using different materials or by employing alternate or secondary processing techniques that are unique to powder metallurgy. 2. Relurively high tooling cost. Because of the high pressures and severe abrasion involved in the process, the PIM dies must be made of expensive materials and be

wder Metallurgy

relatively massive Because of the need for part-speclfic tooling, production qu ties of less than 10,000 identical parts are normally not practical 3. High rrzateraal c o ~ tOn . a unit weight bas~s, powdered metals are considerably expensive than wrought or cast stock. However, the absence of scrap and thee nation of machining can often offset the higher cost of the starting material In tion, powder metallurgy is usually employed for rather small parts where the ma cost per part 1s not very great. 4. Site and shape 1mzttations.Thepowder metallurgy process IS simply not feasibl many shapes Parts must be able to be ejected from the die The thicknessid~a (or th~cknesslwidth) rabo islimited Thln vertical sections are difficult,and th all slze must be within the capacity of available presses Few parts exceed (25 in2.) In pressing area. 5 . Damens,ons change dunng ~interang. While the actual amount depends on a v of factors, includ~ng as-pressed density, slntermg temperature, and sinter~ng dimensional change can often be predicted and controlled. 6. Densaty vanat~onsproducepropertyvarratzons.Any nonuniform product denslt is produced durlng compacting generally results m property variabons through part, For some products, these variatrons may be unacceptable. 7. Health and safety hazards Many metals,such as alumnum, titanium, magnesl iron, are pyrophor~c-they can ignite or explode when in particle form w l surface/volume ratios. Fine particles can also remain airborne for long time be Inhaled by workers. To mm~mize the health and safety hazards, the ha metal powders frequently requires the use of inert atmospheres dry boxes, an as well as specla1 cleanliness of the work~ng environment.

SUMMARY
For many years, powder metallurgy products carrted the stigma of "low streng "inferior mechanical propertles "This label was largely the result of comparisons d den tical" parts were made of the same material, but by various methods of ma ture In essence, the size,shape, and material were all specihed. In such a cornparis product with 10 to 25% residual porosity would naturally be mferior to a fully product made by casting,forming, or machining processes.Unfortunately,it is th' of compartson that 1s frequently made when one considers converting an existing or existing part to PIM manufacture. A far more valid comparison can be obtained by specifying sue, shape, a srred ~ncchanacalproperherEach process can then be optimized by the selection materlal and process conditions. Powder metallurgy can use its unique materi as iron-copper blends for which there are no cast or wrought equivalents.ThePI ucts can be designed to provide the targeted properties while containmg the amounts of resldual poros~ty. Since all products will then possess the targeted ical properties,process comparison can then be based on economic factors, suc production cost. On this basis, powder metallurgy has emerged as a signlGcant facturing process, and its products no longer carry the stigma of "inferior mech properties " Table 18-6 summarizes some of the important manufacturing features powder processing methods Note the variations in product wze, production production quantity, mechanical propertles, and cost.

Review Questions

481

Conventional Press and S~nter lntemed~ate <5 pounds Good Excellent >so00 Excellent ?O 001 ~n 11n Fau 80-90% of wmught Low $0 50-5 OOIlb

Metal Lnjection Mold~ng (MIM) Smallest cli4 pounds Excellent Good >5000 Good +O 003 m hn Very good 9W95% of wrought Intermediate $1 OW10 OOllb

Shape cornplenty Productlon rate Productlon quantrty Dimens~onai

prcclsmn Density

Mechanical propertces cost

Hot-Isostatic Pressing (HIP) Largest I-IOOOpounds Very good Poor 1-1000 Poor +o 020 la lrn Excellent Greater than wrought
Bgh

PIM Forg~ng lntermedrate <5 pounds Good Excellent >10,000 Very good +O 0015 m 11n Excellent Equal to wrought Somewhat low $1.00-5 001lb

>$I00 0011b

i . Key Words
gas atomization green strength hot-isostatic pressing (HIP) impregnation infiltration isostatic compaction liquid-phase sintering lubricant metal injection molding (MIMI mixing particle shape particle size permeablhty PIM forging poroslty powder injection molding (PIMI ~, powder metallurgy prealloyed powder pressure-assisted sintering protective atmosphere rapidly solidified powder repressing secondary operations sinter brazing sinter-HIP sintering size distribution sizing spray forming (Osprey process) surface texture vacuum sintering warm compaction water atomization

I I Review Questions
oduct would be considered to he aprospect for 13. What is apparent density, and how is it related to the final density of a PIM product? 14. What is green strength, and why is it important to the manufacture of high-quality PIM products? 15. How do the various powder properties relate to the method of powder manufacture? 16. What are some of the objectives of powder mixing or blending? 17. How does the addition of a lubricant affect compressibility? Green strength? 18. How might the use of a graphite lubricant be fundamentally different from the use of wax or stearates? 19. What types of composite materials can be produced through powder metallurgy? 20. What are some of the objectives of the compaction operation? 21. What limits the cross-sectional area of most PIM parts to several square inches or less? 22. Describe the movement of powder particles during compaction. What feature is responsible for the fact that powder does not flow and transmit pressure like a liquid? 23. For what conditions might a double-action pressing be more attractive than compaction with a single moving punch?

at are the four basic steps that are usually involved in makproducts by powder metallurgy?

What is the most common method of producing metal tare some of the other techniques that can be employed oduce particulate material'? ich of the powder manufacturing processes arelikely to be estricted to the production of elemental (unalloyed) metal at are some of the unique properties of amorphous metals? y is powder metallurgy a key process inproducingproducts m amorphous or rapidly solidified material? y is flow rate an important powder characterization

24. How is the density of a PIM product typically reported? 25. Why is it more difficult to compact a multiple-thickness part? 26. Describe the four classes of conventional powder metallurgy products. 27. What is isostatic compaction? For what product shapes might it he preferred? 28. What i$ the benefit of warm compaction? 29. What is a reasonable compacted density? How much residual porosity is still present? 30. What types of materials are used in compaction tooling? 31. How do the common sintering temperatures compare to material melting points? 32. What are the three stages associated with most PIM sintering operations? 33. Why is it necessary lo raise the temperature of PIM compacts slowly to the temperature of sintering? 34. Why is a protective atmosphere required during sintering? During the cool-down period? 35. What types of atmospheres are used during sintering? 36. What are some of the changes that occur to thecompact during sintering? 37. What is the purpose of the sinter brazing process? 38. The combined heating and pressing of powder would seem to b e an improvement over separate operations. What features act as deterrents to thls approach? 39. What are some of the atlractive properties of hot-isostatic pressed products? 40. What is the attractive feature of the sinter-HIP and pressureassisted sinkring processes? 41. What are some of the other methods that can produce highdensity PIM products? 42. Describe the spray-formingprocess and the uniquefeaturethai enables production of high-density, fine grain size products. 43. How is the injection molding of powdered material similar to the injection molding of plastic or polymeric products? 44. I n the MIM process, what is done to enable metal powder to flow like a fluid under pressure? 45. How is, the metal powder used in metal injection molding (MIM) different from the metal powder used in a conventional press-andsinter production? 46. What are some of the ways that the binder can be removed from metal injectionmolded parts?

47 Why are MIM productsmjection molded to saes tha slderably larger than the desired product? 48 For what types of parts a PIM injection molding a n . manufacturing process7 49. How does the final density of a MIM product corn press-and-sinter PIM part? 50 What is the purpose of repressmg,coinmng.orsmng op 51. Why can't we use the origlnal compaction t o o h g tc regressing7 52. What 1s the major difference between repressmg forging? 53. What is the difference between impregnatlon andmf How are they slmllar? 54 Why mlght d~fferent conditions be required for the h men(, machuung,or surface treatment of apowderm product? 55 The propertles of PIM products are strongly tied tc Whlch propertles show the strongest dependence? 56 How do the physical properties of PIM products v density? 57. What advice would you want toglve to aperson wh, nlng to convert the manufacture of a component from mg to powder metallurgy? 58. What is the shape of an "ideal" powder metallurgy 59 What are some P M products that have been intentio s~gned to usc the porosity or pemeabdity features of the 60. Give an example of a product where two or more mat, mixed to produce a composite PIMproduct m t h a ui of properties. 61 What are the most cited assets of tho powder mf method of parts manufacture? 62 Why 1s hnish machining such an expensive componen manufacture? 63. Describe some of the maternal5 that can be made I parts that could not be used for processes such as car forming 64. Why is PIM not attractive for parts with low prt quantities? 65. What features of the PIM process often eompensatr hlgher cost of the starting material? 66 How mlght you respond to the crit~clsm that PIM pa lnfenor propertiesq

Problems
~n:itc~ 1'11 lor :it\ling prdiciics. thc pr1m:~ry ~ . ~ r i ; ~ arc h l . chc~ni\tr\.:~ntl .~ ~ ~ I I.. I I \ . . : itrucl~lr\IIV . al features of the starting material will be erased by the melting. For forming processes, the material remains in the solid state, so the principal concerns relnting to the starting material are chemistry and purity, ductility, yield strength, strainhardening characteristics, grain size,and so on. What are some of the characteristics that should be specified for the starting powder to assure the success of apowder metallurgyprocess? In what ways are these similar to or different from those mentioned for casting and forming processes? 2. In conventional powder metallurgy manufacture, the material is compacted with applied pressure at room temperature
I . \\'hen spccit) ln!,, I h u \t:,rl~n::

:rnd ih-n 111tcrc.lh! .let ltcJ tt.lnpcr,tllr,. at ~ t m l ~>re\\urc.\\'tth P hl 1101 ~ ? r ~ . \ i ilhc ~ l sI .C I ~ I ~~)o\vdcr C 13 ed to pressure w h ~ l e it 1s also a t elevated temper, would appear, therefore, that hot pressmg could pr fin~shed part in a smgleoperotion and would be a m nomical and attractwe manufacturing process What have been overlooked in this argument that would favor the press-and slnter sequence for convt manufacture? 3. Investigate the method(s) ured to produce tung candescent lamp filaments How does thernethodus compare to the method developed by Coolidge in IXOOs?

Case study

483

,.npeller

fot

an Automobile Water Pump

controlled height The manufacturer is Interested in
improving quality and lower~ng cost

he component plctured in the f~gure 1 s the impeller of a water pump used by a majorautomot~ve manufacturer. The outer diameter 01 the component e 2.75 in., arnl the tal he~ght of the SIX curved vanes IS 0 75 in. (wlth a lerance of D 005 in.) The diameter of the center hole e ,625 In , and the flat base is 0 187 In. thick, V~bratron and alance conslderat~ons require accurate pasltlonlng and nitorm thickness d the six curved vanes. A relat~vely mooth sutface fin~sh 1 s deerable for good fluld flow. The maxlmum operating temperature has been stlmated at 30Oa? and thecontact flufd should be a ~ateriantrfreeze m~xture wrth eorrosron-resistant dd~t~ves.The designer has provlded a target tenslle trangth of 30,000 PSI and notes that a mrntmurn amount f fracture reststance 1 s abo desirable Since there hould be no direct metal-to-metal rubbing, enhanced ar res~stance does not appear to be necessary.This 9 a h~gh-volume component, however, so low total cost nater~al plus manufactur~ng) would appear to be a prlme bjectlve S~mllar components have been sand cast from cast ron, with a grinding operation be~ng required to malntaln

1. Is a ferrous mater~al needed to prov~de the des~red properties, or might a nonferrous metal be acceptable? 2 . What processes would you want to ~onsrder to massproduce such a shape? Are there more attractwe casting processes?Is th~s a candidate for metal forming, and r f so, which process or processes? Is powder metallurgy a possrblfltyfor thls pmduct7 If so, an the desired pmperties be ach~eved wlth the denslhes that are common for a tradluonal press-and-sinter operation? 3. Investigate the vanous materlabprocess combinations that would be candidates for product~on. Select and defend your "best cho~ce:' 4. Mtght thrs part be a cand~date for manufacturefrom a nonmetal, such as molded nylon, some other polymer, or even some form of reinforced composite material? How would you suggest pr~duclng the desired shape lf one or more of these materials were cons~dered? Would you have to compromise on any of the pesformance requ~rements?

An automobile water pump impeller.

ELECTRONIC ELECTROCHEMICAL CHEMICAL AND THERMAL MACHINING PROCESSES

19.1 INTRODUCTION 19.2 CHEMICAL MACHINING PROCESSES Chemical Machining Chemical-Mechanical Polis Photochemical Machining for Electronics How ICs Are Made

. i19.1 INTRODUCTION

as follows: Complex geometries beyond simple planar or cylindrical features can be ma Parts with extreme surface-finish and tight tolerance requirements can be obt Delicate components that cannot withstand large cutting forces can be machin Parts can be machined without producing burrs or inducing residual stresses. Brittle materials or materials with very high hardness can be easily machined. Microelectronic or integrated circuits are possible to mass-produce. NTM processes can often be divided into four groups based upon the material re mechanism: See Table 19-1.
1. Chemical. Chemical reaction between a liquid reagent and the workpiece r in etching. 2. Electrachemicul.An electrolytic reaction at the workpieee surface is responsib material removal. 3. Thermal. High temperatures in very localized regions evaporate materials. 4. Mechanical. High-velocity abrasives or liquids remove material (see Chapter 2

Machining processes that involve chip formation have a number of inherent limit Large amounts of energy are expended to produce unwanted chips that must

Finally, some geometries are too delicate to machine, while others are too co

SECTION19.2

Chemical Machin~ng Processes

485

When examining these processes, be aware that convent~onalend milhng (see Chapter 25) has these typical machining parameters. Feed rate-25 to 5000 mmlnnn (5 to 200 in.1min) Surface finish-1.5 to 3.75pm (60 to 150 pin) AA Dimensional accuracy4.025 to 0.05 mm (0.001 to 0.002 in.) Workplecelfeature size-61 cm X 61 cm (25 in. X 24 in.); 2.5 cm (1in.) deep In cornpalson, NTM processes typically have lower feed rates and requlre more power consumption when compared to machining. However, some processes permit batch processing, whlch increases the overall throughput of these processes and enables them to compete with machining. A major advantage of some NTM processes is that feed rate is independent of the material being pr0cessed.A~ a result, these processes are often used for difficult-to-machine materials.NTM processes typically have better accuracy and surface finish, with the ability of some processes to machine larger feature sizes at lower capital costs. In most applications, NTM requires part-specific tooling, while generalpurpose cuttlng and workholding tools make machining very flemble.There are numerous hybrid f o m s of all these processes, developed for special appllcabons, but only the main NTM processes are described here due to space hmitations.

.2 CHEMICAL MACHINING PROCESSES

CHEMICAL MACHINING Chemical machiilng (CHM) IS the simplest and oldest of the chipless machiningprocess=.The use of CHM dates back 4500 years to the Egyptians, who used it to etch jewelry. In modern practice,lt 1s applied to parts ranging from very small microelectronic circutts to very large engravings up to 15 m (50 ft) 1ong.TypicaIly metals are chemicallymachmed, although methods do exist for etching ceramics and even glass. In CHM, material is removed from a workpiece by selectively exposing it to a chemical reagent or etchant.The mechanism for metal removal is the chemical reaction between the etchant and the workpiece, resulting in dissolution of the workpiece. One means for accompl~shing CHM is called gel mrllzng, where the etchant is applied to the workpiece in gel form. However, the most common method of CHM involves covering selected areas of the workpiece w t h a maskant (or etch resist) and imparting the remaining exposed surfaces of the workplece to the etchant. The general material removal steps for CHlM are.

1 . Clean~ngContaminants on the surface of the workplece are removed to prepare for apphcation of the maskant and permit uniform etching This may include degreasing, rinsing, and/or pickling. 2. Masklng. If selective etching 1s desired, an etch-resistant maskant 1s applied and selected areas of the workpiece are exposed through the maskant in preparation for etching. 3. Etchmg.The part is either immersed in an etchant or an etchant 1s continuously sprayed onto the surface of the workpiece.The chemical reaction is halted by nnsing 4. Strzpping. The maskant is removed from the workpiece and the surface is cleaned and desmutted as necessary.
Lateral dimens~ons in CHM are controlled in large part by the patterned maskant. Maskvng can be performed in one of several ways depending on the level of prec~sion required In CHM The simplest method of applying a maskant is the cut-ad-peelmethod. In this procedure the maskant material, typically neoprene, polyvinyl chloride, or polyethylene, 1 s applied to the entire surface of the workpiece by dipping or spraying. Once the coating dries, it is then selectively removed in those areas where etching n deslred by scnbing the maskant with a knife and peeling away the unwanted portions. When volume permits, s c r ~ b ~ n templates g may be used to lmprove accuracy. Cut-and-peel

Typical
Typical Penetration or Feed Rate, Process Typical

Mmlm (ipm)

Surface,Finish M , p m (pin.)

Qpical Accuracy, Mm (in.)

Workpiece or F e a t u r e Size, cm (in.)

Comments

up to 1.27 (0.5) thick Photochem~cd machining as above

as above

0.025 to 0.05 (0.001 to 0.002)

30 X 30 Limited to thin material; burr-free (12 X 12); up to blanking of brittle m a t e d ; 0.15 (0.06) thick tooling cost low; used in microelectronics

Electrochemical machining (ECM)

2.5 to 12.7 (014.5)

0.4 to 1.6 (1M3)

O.OU to 0.13 [O.OOOW,OOS); 0.05 (0.002) in cavities

30 X 30 (12 X 12): 5 (2) deep

Stress-free, burr-free metal removal in hard-to-machine metals; tool design expensive; disposal of waste a problem;MRR independent of hardness; deep cuts will have tapered walls Charged high-velocity stream of eIectrolyte; hole diameters down t o 0.127 mm (0.005 in.); 40.1-hole aspect ratios possible

Electrosiream drilling

1.5 t o 3 (0.0M.12)

0.25 to 1.6 (l0-63)

0.025 (0.001) or 5% of hole dia,

up to 0.5 (0.2) thick

Silped-mbe
electrolytic machining (STEM)

as above

0.8 i o 3.1 (32-125)

0.025 to ( L I B
(0,QOI-.W5)

tourinely up t o
127 (5)

Special formof ECM using

conductive tube with insulated

machining (EBM) Laser-beam machining

(12-60)
100 to 2500 (4-100)

0.8 to 6.35 (32-250)

0.005 to 0025 (0.OW2-0.001)

0.025 to 0.63 Capable of micromachining thin mat (0.014.25) thick hole sizes down to 0.05 mm (0.002 in.); 100:1 hole aspect ratios; requires high vac

0.8 to 6.35 (32-250)

0.013 to 0.13 (0.00054l.005)

up to 2.5 (1) thick

Capable of drilling holes down to 0.127 mm (0.005 in.) at 20:i aspeci ratio in seconds; has heat-affected zone and recast layers which may require removal Clean rapid cuts and profiles in almost all plates; 5 ' to 10'' taper; cheaper capital equipment Special form of PAC limited to thin sheets of material, straighter, smaller kerf

Plasma arc cutting (PAC) Precision PAC

250 la 5000 (1&200) as above

0.6 to 12.7
(25-500)

0.5 to 3.2 (0.024.125) 0.25 (0.01)

up to 15 (6,) thick

as above

up to 1.5 (0.625) thick

Wire EDM

100 to 250 (4-10)

(0.25 in.] thick glass Abrasive waterlet machining

15 to 450 (0.6-18)

2.0 to 6.35 (80-250)

0.13 to 0.38 (0.0054.015)

up to 20 (8) thick

Use in glass, titanium, composite$ nonmetals, and heat-sensitive or brittlc materials; produces tapered walls in deep cuts; no bums

v - i L - :' ' +@qg~gg;~~~:&;z~~~ru.r**-r nggr I.lrimru4rrxtai~tii",filiii~Z:' a

b

.- -

R, > 40, tool wear and taper h m ~ t hole aspect ratlo at 2 5 1

(0 005-0 015)

th~ck

coatings are thlck, rangtng from 0.025 to 0 13 mm (0.001 to 0.005 m.). Because of thls th~ckness, the maskant can withstand exposure to the etchant for the extended periods of time necessary to remove large volumes of material This technique 1s generally preferred when the workpiece is not flat or is very large or for low-volume work where the development of screens or phototools necessary for other masking methods 1s not justified. The scribe-and-peel method for ~tepped machlnrng 1s shown in Ftggure 19-1. Another method used to apply maskants 1s screenprinnng, mvolvlng the use of traditional silk-screening technology.The method apphes the maskant through a mask made from a fine silk mesh or stainless steel screen. Masks are typically formed by on top of the application, exposure, and development of a light-sensltlve emuls~on screen.The screen is pressed agalnst the surface of the workplece and the maskant 1s rolled on. Screen prmting is good for high-volume, low-precis~on applications with tolerances typ~cally m the 0.05 to 0.18 mm (0.002 to 0.007 m.) range. Etch depth is 11m1ted to about 1.5 mm (0.06 in.) by the thickness of the maskant, typically on the order of 0.05 mm (0.002 in.).

Part to be

mllled

Parl w ~ t h mask appl~ed

Mask scr~bed and stripped

part with mask removed F~n~shed

area of mask Add~tional scrlbed and removed

19-1 Steps requrred ce a stepped contour by

488

CHAPTER 19

Electronic Electrochem~cal Chemical and Thermal Machining Processes

Mask

FIGURE 19-2 Typical chemical

verhang

Mask
1

milling defects: (a) overhang: deep cuts with improper agitation; (b) islands: isolated high spots from dirt, residual maskant, or work material inhomogeneity; (c) dishina: thinnina in center due idimproper agitation or stacking of parts in tank.

Work

Work

Work

'9

la)

Etch rates in CHM are very slow compared wlth other nontrad~tlonal ma processes. However. etching proceeds on all exposed surfaces s~mnltaneously, whic nificantly increases the overall material removal rate on large parts The etch rat CHM is dlrectly proportional to the etchant concentration d~rectly adjacent to the being machmed. For parts machmed by immersion, the uruformity of the etchant centration within the bath can be ~mproved by ag~tation. If the bath is not agitate erly, several defect conditions can resull, as shown in Flgure 19-2. Islands, or high spots, can be the result of improper agitation on large parts. Islands can formed due to inadequate cleaning or inhomogeneity with the work material. Smgle-sided, blind etching of the part is called chemical milling or, when the toreslst method of apply~ng maskants IS used,photochemrcal mtlltng Chemical m is so named because its earliest use was for replacing mechanical millmg on lar ponents. Chemical mllling is often used to remove weight on axcraft com shown III F~gure 19-3. Through-etching of the workplece 1s called chemzca chernzcal) blanking. The process competes with blanking, laser cutting, and ele d~scharge machining (EDM) for through-cutttng of thin material sheets. Chemica mg is typically performed uslng double-sided elch~ng to Increase productio and min~mize taper on theetched walls of the featu1e.A key requirement for ch blanking 1s registration of top-s~deand bottom-side screens or phototools masking. Because of the precision required, chemical blanking 1s not perform the cut-and-peel fiethod of masking.

lconel 71 8 aircraft engine parts. These sheet metal parts for a jet fighter engine are chemically rn to remove weight. (Left) As-formed workpiece; (middle left) workpiece coated with liquid rubber, fiberglassscrib' template in place; (middle right) scribed workpiece; (right) finished part. About 0.035 in. of stock is removed fro the 0.070-in.-thick workpieces. Tolerances are held to i0.004 in.
FIGURE 19-3

SLCTION 19.2 Chem~cal Machin~ng Processes

489

Advantages and Disadvantages of Chemical Machining. Chemical machining has a number of dislinct advantages when compared with other machining and forming processes. Except for the preparation of the artwork and phototool, screen or scribing template, the process is relatively simple, does not require highly skilled labor, induces no stress or cold working in the metal. and can be applied to almost any metalaluminum, magnesium, titanium, and steel being the most common. Large areas can be machined tanks for parts up to 12-by-50 ft and spray lines up to 10 ft wide are available. Machining can be done on parts of virtually any shape. Thim sections, such as honeycomb, can be machined because there are no mechanical forces involved. The tolerances expected with CHM range from 10.0005 in. on small etch depths up to 50.004 in. in routine production involving substantial depths. Tolerances in chemical milling increase with the depth of the cut and with faster etch rates and vary for different materials. The surface finish is generally good to excellent for chemical polishmg. In using CHM, some disadvantages and limitations should be kept in mind. CHM requll-es the handling of dangerous chemicals and the disposal of potentially harmful byproducts, although some recycling of chemicals may be possible.The metal removal rate is slow in terms of the unit area exposed, being about 0.2 to 0.04 lhlmin per square foot exposed in the case of steel. However, because large areas can be exposed all at once, the overall removal rate may compare favorably with other metal removal processegpartitularly when the work material is not machinable or the workpiece is thin and fragile, unable to sustain large cutting forces. Photochemical Machining. Figure 19-4 shows the specific steps that are involved when photochemical machining (PCM) is performed with the use of photoresists.These are as follows:

1 . Clean the workpiece. 2. Coat the workpiece with a photoresist, usually by hot-roller lamination of dry-film photoresists, on both sides, although liquid photoresists may also be applied by dipping, flowing. rolling, or electrophoresis (i.e., migration of charged molecules in the presence of an electric field). For liquid photoresists, the coating is heated in an oven to remove solvents. 3. Prepare the artw0rk.A drawing of the workpiece is made on a computer-aided design (CAD) system.

Clean workplece

Start

Dip or spray workpiece to apply the photoresist

coating

appl~ed to surface (Apply master pattern 1 Both sides of resist exposed.

Negatives

Photoresist Unexposed

developed. areas o f resist removed.

Etched by reagent

Mask removed.
F~nish

FIGURE 19-4

Basic steps in photochemical machining (PCM).

4. Develop thephototool.The GAD file is used to derlve a photograph~c negative workpiece. Several methods may be used.Typically,the CAD drawing is down1 to a laser-imaging system that exposes the desired Image directly onto photograp (e.g., silver halide) film. In the past, oversized artwork was used to Increase the curacy of the phototool through photograph~c reduction of the artwork. 5. Expose the photoresist. Brlng the phototool in contact with the workpiece, a vacuum frame to ensure good contact, and expose the workpiece to Intense violet (UV) light 6. Develop the photoresist.Exposure of the photoresist to intense UV light alters the ch istry of the photoresist, making it more resistant to dissolution 1n certaln solvents. placing the exposed maskant in the proper solvent,the unexposed areas of the resist removed, exposing the underlying material for etching. All resldue is rinsed away 7. Spray the workplece with (or immerse it in) the reagent 8. Remove the remaming maskant. PCM has been widely used for the production of small, complex parts, suc printed circuit boards, and very thln parts that are too small or too thin to be bla or m~lled by ordinary sheet metal forming or machlmng operations, respectiv Refinements to the PCM process are used in the microeletronics fabrication.

Design Factaft in U~emical Machining. When designing parts that are to be

by chemical machinmg, several unique factors related to the process must be ke mind. First, if artwork is used, dimensional vanations can occur through size ch in the artwork or phototool film due to temperature and humidity changes.These c controlled or el~minated by putting the artwork on thicker polyester films or gla controlling the temperature or humid~ty in the artwork and phototool production or by using a hrect-wrlte laser-imaging system. The second item that must be cons~dered is the etch factor, sometimes refers as etch radrus, wh~ch describes the undercutting of the maskant The etchant acts iso ically on whatever surface is exposed Areas tkat are exposed longer wlll have metal removed from them. Consequently, as the depth of the etch increases, ther tendency to undercut or etch under the maskant (Figure 19-5). The etch factor, chemical machining is d e f i e d as

E=U
d

where d is the depth of cut and U is the undercut as defined in Figure 19-5 In chemical machmning, the term ansotropy (sometimes referred to as etch factor) to describe the directionality of the cut. Anisotropy,A, of a material-etchant intera in photochemical machinmg is defined as A = d

which is the inverse of equation 19-1. In many electrical and electronic pro anisotroples much greater than 1 are desirable in order to permit greater densi electrical and electronic components and wires An allowance for the etch factor must be taken i t o account in designing th and the artwork or scribing template. In the case of chem~cal milling, the width opening in the maskaut must be reduced by an amount sufficient to compensatef undercut under both sides of the maskant:

W ,

Wf - ( E - d )

where Wf is the final desired wdth of the cut. This allowance is considered mum allowance;it has been found that results will vary based on etching conditions, tual etch allowances w~ll have to be somewhat greater and adapted to specificcon In double-sided chemical blanking, a sharp edge remains along the line a breakthrough occurs Because such an edge 1s usually objectionable, etching or

FIGURE 19-5

Undercutting of the mask or resist is defined by the etch factor, which must be accounted for in designing the part using the artwork on the scribing template. is continued to produce nearly straight side walls, as shown in 19-5g.In order to achieve nearly straight side walls, it is typical to allow the process to continue for the amount of time necessary to do a through-cut from one side. The anisotropy of the cut defines the maximum limit for aspect ratio (i.e., ratio of depth to width) of the cut, which is ameasure of how deep and narrow a cut can be made. In chemical blanking, by using a double-sided etch,the n~aximum aspect ratio of the cut may be effectivelydoubled if the process is stopped at breakthrough (Figure 19-5f).This is the effective maximum aspect ratio that may be achieved in CHM. Consequently, it is difficult to produce deep, narrow cuts in materials using CHM.Table 19-2 shows some etch rates and etch factors for common metal and etchant combinations in CHM.

Preferred Metal
Aluminum Copper Nickel allays Phosphor-bronze Silver Titanium Tool steel

Etchant
FeCI,
P~CI, FeCb Chromic acid FeN03 HF HNO3

Penetration Rate (mmimm)

0.025 0.05 0.018 0.013 0.02 0.25 0.018

Etch Factor E
1.7 :1 2.7 :1 2.0 : 1 2.0 :1 1.5:l 2.0 :1 1.5:1

~nditions, at which rdinarily

Source: G. E Benedict. Nontraditional Manufocluriag Processes, Marcel Dckker. New York, 1987, p. ZOO.

C H A 19 ~ Electronii:Electrochemical Chemical and Thermal Machining Processes

The soundness and homogeneity of the metal are very important. Wrough rials should be uniformly heat treated and stress relieved prior to processing.Al chemical machining induces no stresses, it may release existing residual stresses metal and thus cause warpage. Castings can be chemically machined provided t are not porous and have uniform grain size. Lack of the latter can cause non etching rates,producing islands. Because of the different grain structures that exi welds, weldments usually are not suitable for chemical machining. Preferential e due to intergranular attack can also be an issue in CHM.

CHEMICAL-MECHANICAL POLISHING Chemical-mechanical polishing (CMP), or chemo-mechanical polishing, uses th ergy of chemistry and mechanical grinding to obtain flatness on the order of CMP is used in the fabrication of integrated circuits (ICs) to obtain planar s aiier dielectric and metal depositions during interconnection of circuit compo As shown in Figure 19-6, the process involves rotation of the wafer as it is p against a slurry-filled abrasive pad that rotates counter to the wafer.The slurry co both a particle abrasive as well as an etchant and is deposited directly onto the sive pad by a mechanical arm.The wafer is held in a carrier by a backing pad d to distribute the mechanical force evenly over the surface of the wafer.Fuseds a weak KOH solution is a common slurry for oxide polishing. Ferr' phate with silica or alumina is used to polish tungsten. Raised features a faster than flat regions, since mechanical pressure concentrated on the ralsed enhances etching. For dielectric polishing, typical etch rates are on the orde nmlminute.

Wafer

SIDE VIEW

/
Holder

1 Abrasive

"*

Slurry feed

FIGURE 19-6 Schematic of chemical-mechanical polishing (CMP).

TOP VIEW

S E ~ O19.2 N

Chemical Mach~n~ng Processes

493

PHOTOCHEMICAL MACHINING FOR ELECTRONICS The most common and most preclse method for creatlng maskants involves the use of UV light-sens~tivemulsions,calledphotoresisn. In thls method, photoresists are applied to the surface of the workp~ece and selectively exposed to an Intense ray of UV I~ght through a photographic negatlve of the image to be patterned The use of photores~sts, calledphotochem~cal machmcn~,is widely used m the manufacture of integrated circuits in electronics. The entire sequence of operations for the manuIacture o f electromcs 1s given in Frgure 1-13 in Chapter 1. Integrated circults use sem~conductor materials-such as sllicon, galllum arsen~de, and germanium-that can be made to be e ~ t h e electrically r conductmg or electr~calIy Insulatlng by changing the type and concentration of impur~ty atoms found within the mater~al.Like metals, all semiconductors have crystalline m~crostructuresexhlblting long-range order in the form of a Lattice. However, unllke metals,semiconductor atoms are characterized as having half-filled valence shells, and so, when placed into a lattice, the semiconductor atoms form covalent bonds A t room temperature (25"C), s~licon permits a small amount of electr~cal condudlv~ty that 1s too small for most electronic app11calions.The electrical conduct~vity of semiconductors can be altered by inserting lmpurlty atoms Into the semiconductor lattice. The process of mod~fylng the electrical properties of semiconductors by in2roducing Impurity atoms is commonly referred to as dopmg. The first level of electromc manufacturlng involves the manufacture of the ICs or chips.This is a complex process involving many steps, the sequence of wh~ch depends on the particular electrical device. The ~nitlal steps of doping by d ~ f f u s ~ o or nIon nnplantation and oxldation are performed in large mach~nes that manipulate the wafers m and out of varlous vacuum chambers in the correct sequence and duration. Doping can be accompl~shed in bulk by alloying at the time of crystal format~on. However, selective doplng is required for ICproduct~on. Selectwe dopmgm most early IC devices involved thermal dlffuslon, more recently, as dev~ce dimensions have continued to shrink, ion ~mplantat~on has become more sultable to better control the depth and concentration of the dopant atoms in the sillcon wafer. Slllcon is the most w~dely used sennconductor material. It is plentiful and can be readily produced in single-crystal form; see Figure 19-7for detads. Also, the native oxidc, silicon dioxide, can be used as both a dieletric layer and a diffusion mask during
processing.

FIGURE 19-7 How the silicon wafer is made. One of the key reasons that single crystal silicon is the most widely used semiconductor material is that it can be refined and grown economically in single-crystal form. Here's how it is done. Under equilibrium conditions, molten silicon, when cooled, produces a polycrystalline structure. However, under controlled conditions, silicon can be grown from a single seed in a large single-crystal ingot called a boule. The technique used most often for growing single-crystal silicon Is called the Cmchralski method. In the Czochralski method, a small seed crystal is lowered into molten silicon and raised slowly, allowing the crystal to grow from the seed. The size of the seed crystal is about 0.5 cm diameter by 10 cm long. Its crystallographic orientation is critical because it defines the crystallographic orientation of the boule which controls the electrical properties within the boule. The melt consists of electronic grade (99.999999999% pure) polycrystaliine silicon (poiysilicon). If desirable, dopant may be added to the melt, although alloying complicates the crystal growth process. The silicon is melted in a fused silica crucible within a furnace chamber backfilled with an inert gas such as argon. The crucible is heated to approximately 1500C and maintained at slightly above the melting point with a graphite resistance heater. Once grown, the boule is characterized for resistivity and crystallographic defects. The unusable end portions of the boule are cut off, and the outside of the body is ground into a cylindrical ingot. For diameters below 300 rnm, fiats are ground along the length of the ingot to identify the crystallographic orientation and the boule is sliced into wafers using wire or diamond saws. The wafers are lapped, chemical etched, and polished, ready for IC production.

494

C H A ~19 R Electron~c Electrochem~cal Chem~cal and Thermal Mach~ning Proceses

3. Mask oxrde and photolithography

5. Mask removal

and (gate ox~dei fresh ox~dat~on

6. Deposited polys~llwn

9. Photolithography

11, Oxide

deposition

12

Photolrthograp

13. Etch

14. Metallization

16. Etch

FIGURE 19-8 The sequence of processes or steps requ~red to manufacture a slmple (MOS) f~eld effecttransistor dev~ce 1 s shown here. (Source. Texas oxfde-sem~conductor
17. Flnal overcaat

Engineering Extensron Service, Semrcondvctor Processing Ovewrew, College Station, Texos

A&M University System, 1996.)

HOW ICs ARE MADE The ability to selectively modify the electrical properties of semiconductors is bone of microelectronic manufacturing. As shown in Figure 19-8,the manufactu~gfabrication sequence for ma1 ple bipolar diode has many steps, beginning with the production of a silicon v a predoped, single-crystalingot (bor~le) that is cut into wafers, lapped, and polisl duce silicon wafers.The wafers are placed in vacuum chambers, where an oxi grown on the surface of the wafer to act as a mask during subsequent doping I strate. The oxide layer is patterned using photolithography in combination w i (see 13gure 19-9).Photolithography is used to produce a polymeric mask ove~ layer, which will allow only select areas of the oxide layer to be etched.Aft (chemical machining), the polymericmask is removed from the silicon dioxide the silicon is doped (by diffusion) with boron. After doping, the silicon dioxit removed, and a second silicon dioxide layer is grown and patterned to establisl in the silicon dioxide layer above the duped regions. Next, a thin metal film is on top of the silicon dioxide to provide an electrical pathway allowing the pr

UV l~ght

layer Sll~con 1 rtlng wafer wfth layer ~epatterned

Resrst coatlng

12) Coat with photoresist

using

spln coatlng

(3) Bake the resfstto set its

d~ssolut~on properties

141 Ex~ose resist bv sh~ntno " light through a photomask

imerse exposed wafer developer

16) Hard

bake

17) Etch the film

18) Strip ~ f f t h e resist

Wafer

a a
4

Lens 1

Mask

Lens 2

Contact printing

Proximity printing

Wafer Projection prlntlng

S10,

FIGURE 19-9 (a) The photolithography process itself has multiple steps, and each step can have variation, as shown in (b), the exposure step (4) in part (a).

negative electron regions of the diode to be connected to an external power supply. Photolithography and etching are used once again to pattern the thin film into leads and contact pads large enough for biasing the device.To protect the linal integrated device from mechanical damage and moisture, a final passivation coating is added. This example shows the production of a single IC component.Typically, multiple components and, further, multiple circuits are produced in parallel during IC fabrication. Over the years IC fabrication has evolved from the original small-scale integration (SSI) architecture of the 1960s, with 2 to 50 electronic components per circuit, to the ultra-large-scale integration (ULSI) architectures of today, with tens of millions of components per circuit.The classification of ICs by scale of integration represents the successive advancement of semiconductor processing technologies to provide lowercost, higher-performance ICs. Each increase in the number of components represented a breakthrough in miniaturization technology (e.g., photolithography and clean rooms)

CHAPIER 19 Electroni c Electrochemical Chemical and Thermal Machining Processes that permitted the fabrication of smaller 1C components with improved perfo To achieve lower cost, manufacturing processing technology breakthroughs w e ed to make miniaturization technologies possiblc and economical. Today lhis tr seeking higher performance at lower cost continues. The photolithography process involves the sequence of steps shown in Fi First. a liquid photoresist is applied to the surface of the silicon oxide layer ov wafer.Typically,this is done with a process known as .spin corrling. In spin coatino forces are used to produce a photoresist layer of uniform thickness. Next, ti is soft baked on a hot plate or in an oven. In this step, solvents used to reduce of the photoresist during spin coating are evaporated, and adhesion between the w the photoresist is improved. After soft bake, the photoresist is e.ryosed, using a ph to transmit a pattern of electromagnetic radiation onto the surface of the step is performed using a machine called a stepper, because thc litho device is indexed or stepped across the wafer, subjecting it to repeat each chip being made. Once the resist has been exposed, the wafer is deve ical solvent.Development chemically machines unwanled resist materials, derlying material to be etched. Next. the resist is hurd baked to remove any r solvents after development and to further toughen the remaining resist ag etching or implantation processes. Hard bakes generally take longer and ly higher temperatures than soft bakes. Once the downstream etching or implantat made use of the resist, a photoresist stripping step is necessary for removal of the re Obviously, the most important requirement of the photoresist is that it res downstream etching or implantation process. Other requirements important to th tion of resists are their resolution and sensitivity. Resolution refers to the smallestline that can be reproduced repeatably by the resist.The resolution of the resist is str function of the source of ionizing radiation or the exposure machine tool used.S ty refers to the amount of ionizing energy required to modify the solu

IC MANUFACTURING AND ECONOMICS By the start of the year 2000, ICs with over 10 million transistors had been produce it is true that small circuits are inexpensive, the cost of packaging, testing, and ass the completed circuits into an electronic system must be taken into acc arc separated into individual chips, each chip must be handled individually.Thus pa ing and testing costs often dominate the other production costs in the fabricationof One way to improve the economics of microelectronic manufact wafer sizes.The key benefit from processing larger wafers is an incre age of usable area. Larger wafers have a smaller proportion of the area being by edge losses and wafer dicing. Since the mid-1980s, waler diamet threefold from 100 mm to 300 mm, which rcquired the developn~ent of new equi throughout the setniconductor manufacturing pr0cess.A second strategy for imp semiconductor economics involved increasing the number of chips per wa decreasing IC dimensions. IC dimensions have decreased more than 50-fold in the 30 years.The smallest feature size in 1971 was 10 microns. By 2001. transistors with features as small as 0.18 micron were made.Again, the catalyst for this improvemen an investment in the process technology-in particular, photolithography. Die yicld improvement is another more desirable way to improve econd without making large capital investments. The die yield depends (the fraction of silicon wafers that started versus those that finished thc process), involves the processing yield (the fraction of good dies per wa (the fraction of dies that are packaged), and the burn-in yicld (th dies that survive wafer testing). A single, submicron dust particle tr photoresist and reticle in a photolithographic step can cause a point de result in the malfunction of an entire 1C. As a result, all microe is conducted in clean rooms, where special clothing must be wor nation of wafers by dust particles). The air is continuously filtere high-efficiency particulate-arresting filters to keep the dust level at a mi are commonly processed in Class 100 clean rooms.

Sprocket

drive holes

Inner leads
bonded to IC

of polymer
support rlng after exclslng
by

W~ndow i n tape to fac~litate and expose outer excis~ng leads beyond polymer support rlng for outer lead bondlng

custom die

IC PACKAGING Several levels of packaging and assembly are necessary to integrate the IC chip with other electronic devices t o make it part of a fully functional commercial or military product. IC packaging serves to distribute electronic signals and power as well as provide mechanical interfacing to test equipment and printed circuit boards (PCBs). In addition to this interconnection role, IC packages protect the delicare circuitry from mechanical stresses and electrostatic discharge during handling and in corrosive environments during its operational life.Finally, because of the high density of the integrated circuits, dissipation of heat generated in the circuits has hecome more critical. The first step in IC packaging is to attach the chip to the die using various techniques, including wire bonding, tape-aulomated bonding (TAB), and flip-chip technology. In wire bonding, also known as chip-and-wire aftachmenl,the chip is attached to the tape with an adhesive, and wires are attached to bonding pads o n the chip. See Figure 19-10. Gold wire as thin as 25 microns and aluminum wire as thin as 50 microns can be attached to the chip, and the other ends of the wires then become the leads on a die, so the die can not get packaged, as shown in Figure 19-11. IC chips are mounted on a variety of packages made from a variety of materials. Figure 19-11 shows a cutaway view of the most well-known IC chip package; the dual in-line package (DIP) refers t o the two sets of in-line pins that go into holes in the PCB.
Plastic DIP Molded plast~c bodyNo cavity between body and c h ~ p

FIGURE 19-11 The dual in-line

Lead pitch

package (DIP) has a leadframe and package body. The leads on the chips are connected to the pins on the DIP. (Source: Seraphim, D. R, Lasky, R. C., and Li, C-%, Principles of Electronic Packaging, New York: McCraw-Hill, 1989.)

The DIP, like all other IC packages, is made up of a leadframe and a package Typically composed of a copper alloy (sometimes w ~ t h an aluminum coating), the frame provides electrical rntetiace between the IC and the PCB.The DIP body is from a low-cost epoxy. which facihtates mass producbon. In high-rehability applica ' (e.g,, mrlitary), where hermetic (airtight) sealmg of the package is important. cer package bodies are used. Generally, IC packages are grouped mainly based on the arrangement, s and quantity of leads. Lead pitch refers to the center-to-center distance between on an IC package. In conformance to standa~d-sett~ng bodies, such as the Electr Industries Association (EIA) in the United States and EIA Japan, lead pitchesa 20 mils (0.02 in.) are measured in inches. Below 20 mils, lead pitches are measur mrllimetets There are two methods by which components are connected to the cir the PCB.The DIP is the leadrng example of through-hole (TH) technology, also as pin-in-hole (PIH) technology, where IC packages and dlscrete components serted Into metal-plated holes in the PCB and soldered from the underside PCB. In surface rnounr (SM) technology, electronic components are placed on der paste pads that have been dispensed onto the surface of the PCB. Figure shows the cross seclion of solder joints for typlcal SM- and TH-packaged comp on a PCB. SM packages are more cost-effective m electronic assembly, and t h ~ SM s tec ogy has replaced a lot of the TH technology, but not entrrely, since not all elec components can be purchased in an SM package. SM packages are designed for mated productron and allow tor hrgher crrcuit board deus~ty than TH compo The manufacturing challenges associated with SM technology include weaker strength and solderabilty Issues relating to lower m process lead temperatures Als components have only one lead geometry, whereas SM components have many ent designs The key packagng famihes for TH technology are dual in-line pa (DIPS) and pLn grzd urrays (PGAs). In SM technology, IC packages cannot be discussed separately from lead g try. Lead geometnes affect the electrical performance, sEe constramts on the PC ease of assembly of the IC package.The most basic form of SM lead is the butt le J-lead. (See F~gure 19-12.) Butt leads are normally formed by clipplng the leads components.This technique is sometimes used to convert an existing TH compon an SM component. Consequently, butt-leaded components do not typically sav space on the PCB. However, they can reduce costs by eliminating the need tope TH soldermg of the PCB after SM soldering. Butt-lead components tend to re the lowest solder jomt strengths, and therefore relrabihty is an issue. Gull wzngleads bend down and out, whereas J-leads bend down and In. G leads allow for thrnner package slzes and smaller leads. which is Important for co applications such as laptop computers i n addition, packages with gull mng le compatible with most reflow soldering processes and have the ability to self-ali lng reflow if they are slightly msoriented. Gull wing leads are compatible wi p ~ t c hpackages, but inspectron of solder joints is difficult rn the final so configuration. Gull wing Leads are also suscepbbile to lead damage and devlatlon Lead coplanarity. J-leads are sturdier than gull wings and stand up better in h The solder joint of J-leads face out, making inspection easier. J leads have a profile than gull wings, which can be a disadvantage for compact applications. same time, this higher standoff makes postsolder cleaning easler. J-leads can be for packages with between 20 and 84 leads Solder bulls are increasingly bemg used to provide SM interconnection thr 19-12shows a BGA package. BGAs provide hrgh lea grrd arrays (BGAs).F~gure because the solder balls are arrayed across the enfire bottom surface of the package. counts on BGAs can go as h ~ g h as 2400, with most in the 200 to 500 lead range Be of their arrayed nature,BGAs do not need as fine a p~tch (40 to 50 mils) as quad ages (QFPs), which can help in electronic assembly ylelds To further boost yiel der balls on BGAs have excellent self-alignmg capability during reflow and req

SEC.IION 19.2

Chemical Machining Processes

499

Slngle slde

SIP (angle in-line package)

ZIP (zig-zag in-llne package)

Through-hole mount Dual slde-DIP (dual in-l~ne package)

Full surface-PGA ( p ~ gridarray) n

TSOP (thin small-outllne padage) Surface mount

a
Bun leads

SOJ (small-outline J-lead package)

QFP (quad flat package)

QFJ (quad flat J-lead package1 Quadruple s~de

111 surface-

BGA (ball grld array)

LCC (leadless chlp carrier)

SM
Saldei
ball

LCC SOJ (leaded chip catrier, small-out llne J-lead package)

Packages can be throughholemount orsufface mount type, and be made out of plastic or ceramrc. The pins can be on one s~de (SIP), two s~des (DIP), or four sides lquad) of t b e package, or distributed over the surface of the package IPGAs or BGAs)

"CURE 19-12 Here is a summary of the varlous types of packaging used for ICs (Source: Manzfone, L. T , astic Packaging of M~croelectron~c Devices, New York. Van Nostrand Reinhold, 1990.)

500

CHnrl~n 19

Electron~c Electrochem~cal Chem~cal and Thermal Machlnlng Processes

FIGURE 19-13 How printed coplanarity (6 to 8 mils) than other 1eads.The downs~de of BGAs is the difficulty ass clrcu~t boards are made ated with cleaning, inspection, and rework of solder joints and the lack of compaub The pnnted cncu't wlth some reflow methods, since jolnts are out of s~ght beneath the package or prlnted wlrlng board, The next level of electronic manufacturing involves populating the PCBs w~th connects the 1C wlth other 1 and other devices usmg the surface mount or through-hole techniques. See F~gure components to produce a funct~onal clrcult Spec~f~cally, a for a brief description of the PCB fabrication process. PCB is a lammated set of PCBs can be slngle sided, double sided, or multilayer. Single-sided PCBs sl layers O r laminates of have metall~cclrcu~ts on one side o f t bulk sheet materials that have insertion holes that extend through metallic clrcultsthat are used to may be inserted into the board (Fi interconnect the varlous packaged components. A s mounted onto the pads on the sam shown on the left, each PCB Double-sided PCBs are used in cases where clrcuits must "jump," or cross over, and laminate has a another. In this case, vza h o l (or ~ simply vias) are needed to route the cncults over pads The base mater~al must be another. Was are essentially meta electrically insulatingto pmv,de nect a c~rcuit on one side to the o support to all components maklng up thec~rcu~t Pads on that are also used as insert~on h the laminate are connected by number of packaged camponents on the board increases, the complexity of the cir conduct~ve tracks or traces Increases, givlng rlse t o (usually copper) that have been double-sided boards are laminat depos,ted onto the surface of ermost track on one slde of the b for traces the base. ~h~ and pads IS depos~ted by through vias. Vias wlthm a lam electroless plat~ng and burred vrus.V~as that come out on only one side of a mult~layer PCB are called bh Surface mount pa~t~ally burred vras. Multilayer (SM) components are connected more to the PCB at pads (lands) or, i n the case of throughhole (TH) technology, at Insertion holes ELECTRONIC ASSEMBLY Typical base materials used The term eleclronic ascembly is generally reserved for the third level of electronin may be epoxy-~mpregnated ufacturing rnvolving the soldering of packaged ICs and other discrete components f~berglass, poly~m~de, or ceramic. PCBs using e ~ t h ethrough-hole r andlor surface mount. As explained in the sectlon fiberglass pdckaging,TH technology refers to the insertion of packaged leads into plated thr the cheapest substrate for holes in the PCB and soldering of the terminals from the backside. SM techn lnterconnectlng leaded packages. Frberglass 1 s used to tnvolves temporary attachme Increase the mechanical contaming solder paste, whic of the dev~ce for handling, while smaller and have much d epoxy resln Imparts better nations rather than leads ductlllty. The f~berglass 1 s Impregnated on a continuous reduced Inductance and capacitance losses. lhne where resin ~nfiltrates the The sequence of o fiberglass mat I n a dip and Insertlon can he perform the soaked fabric passes through insertion, leads are gen a set of rollers to control thickness and a n oven where the between jointsduring sol resln is partially cured. The erally, soldering of TH components is performed automatl~ally through a process result~ngglass resln sheet is as wave soldenng. Wave soldering involves the conveyance of a preheated and pt called prepreg Multiple prepregs are then pressed together between electmformed copper foll under preclse heat and pressure cond~t~ons to form a copper-clad lammate or PCB Heat -----t

Heat

lnsert~on hole
PCB process to

Impregnate f~berglass wlth epoxy resln

Hole through board

Component Component lead

1111

1 ; 11

1Land

Base material

Hole through

Component

PCBs can be single sided, double sided, or multilayer. (Source: judd, M., and Brindley, K., Soldering in Electronics Assembly, Boston: Reed International Books, 7 992.)
FIGURE 19-14

PCB over a standing wave of solder created by pumplng actson The comb~nat~on of capillary achon and pumplng action permlts flow of the solder from the underside of the board into the joint Cleanliness of the PCB is critlcal for wetting of the lead and PTH. A high-pressure asr jet is used to blow off excess solder from the underside of the board to prevent solder b r ~ d g ~ n Postsolder g cleanmg of the boaxd ~ncludes degreas~ng and deflnwing. One key considerahon for TH solder joints is joint strength The trade-off is the clearance between the insertion lead and insert~on hole. As the clearance decreases, jolnt strength increases However, with smaller clearances it is more d~fficult to insert plns in holes. Clearances on the order of 0.25 mm are typical. Another factor affecting joint strength ~nvolves the cluzchmg of leads. Clinched ledd joints are muoh stronger than unchnched join(s.Because the mechanical strength of TH jolnts is generally superior to SM joints, large, heavy components are generally attached wlth TH technology. SM assembly involves application of solder paste to the lands on the surface of the PCB, placement of SM components on top of this paste, and reflow of the solder paste within an oven. Solder paste consists of small spherical particles of bolder less than a tenth of a millimeter in diameter together with flux and solvents used to dissolve the flux (imparting tackiness) and tlucken the paste. At the time of applsation, the paste has the consistency of peanut butter and is applied by screentng, stenciling, or dispensmg. In screening and stenc~ling, a solder paste printer is used to apply solder paste through a mask (screen or stencil) by running a squeegee over the surface of the mask.The mask

Electronic Electrochemical Chemical and Thermal Machining Processes

a TH PCB (above).The steps for SM asse are given below, along with the steps for mixed technologies-both T H and SM. (Source: Hoskord, M. R., Electronic Circuit Cards and SurfaceMount Technology: A Guide to Their Design, Assembly, Application, New York: Prentice Hall, 1992.)
FIGURE 19-15 The assembly process steps for making

is pressed against the PCB, allowing contact between the paste and the lands board. After the squeegee has passed, the mask snaps back from the surface,lea island of solder paste on the PCB lands. Stencils are typically metal sheets or wire that have been chemically etched using a lithographic process. Screens are

which can lead to other problems, including solder paste slurrip (spreading out solder pastes after application).

S E C T I O N 19.2 Chemical Machining Processes Squeegee

I

503

Solder paste

Squeegee

16 Schemat~c for der paste on a squeegee in a screen cess in SM (Source Prasad, R , Int Technology: New York. id Pract~ce, Hall, 1997, p. 493 )

Substrate/ holder

Substrate

Paste deposit

Snap-off distance Once the solder paste is positioned on the board, a component placement machine, also known as apick-and-place machine, is used to place the components onto the solder paste pads.The flux in the solder pasteis tacky and holds positioned components in place until oven soldering. Components are fed to a robotic manipulator that has a vacuum chuck or a mechanical chuck, or both. Component feeders deliver components to the manipulator. Several types of feeders exist, including tape (or reel), hulk, tube (or stick), and waffle pack. The feeder system must he carefully selected based on the desired quantity per feeder, availability, part identification, component cost, inventory turns, and potential for damage during shipping and haudling.Tape feeders are widely utilized and are most desirable for high-volume placement. Tube feeders are useful for smaller-volume assemblers, even though costs per component are higher. Waffle packs are flat-machined plates with inset pockets to hold various chips. In general, waffle packs increase the cost of assembly. However, some IC packages, like the bumperless, fmepitch QFPs, require a high level of protection during handling to minimizelead damage, so this component requires the tape-feeding mechanism. Bulk feeding of I C components, through the use of a vibratory bow1,may be useful for prototyping environments. The economics of SM technology are driven by component placement equipment, which determines the throughput of the SM line and is the source (at least partially) of most defects requiring rework. Further, placement equipment strongly influences startup costs because it may involve as much as 50% of the capital equipment cost in setting up a line. Key criteria in the selection of placement equipment include placement accuracy, placement rate, maximum PCB size, types and sizes of components, and maximum number of feeders, among others. In general, placement equipment has been classified into four discrete types: (1) high throughput, (2) high flexibility, (3) high flexibility and high throughput, and (4) low cost and low throughput with high flexibility. High-throughput placement machines are called chlp shooters. Chip shooters are typically dedicated to the placement of passive (resistors, capacitors, etc.) and small active (IC) components and can place components at rates up to 60,000 components per hour with linear repeatability aronnd0.05 to 0.1 mm and rotational accuracy of 0.2" to 0.5" over an area 350 by 450 mm. area. After components are placed, the PCB is placed in a reflow oven where the solder paste melts, causing a fluxing action that permits the melted solder to wet the leads and the PCB lands. To achieve this, the PCB must be exposed to an appropriate thermal profile, or time-temperature curve, as it passes through the oven.At a minimum, the thermal profile must include at least four zoues.The firs1 zone, called preheating, is used to drive off any nonflux volatiles within the paste.The second zone, the soak zone, is used to bring the entire assembly up to just below the reflow temperature of the paste.The third (reflow) zone quickly raises the temperature of the solder paste above the reflow temperature,

C H A I R19 Electronic Electrochemical Chemical and Thermal Machining Processes

FIGURE 19-18 Relationship of penetration rate, current dens~ty, and rnachin~ng gap in electrochemical machining.

"0

004

008

0.12

016

020

Mach~nlng gap, mm

hardness or toughness of the work material.This makes ECM advantageous forth chining of certaln high-strengthmaterials with poor machinabihty. Penetrationra to 0.1 m.imin are obtalned rouhnely III WaspaIloy, a very hard metal alloy. Pulsed-current ECM (PECM) has recently shown the potential to improve raaes and surface finish in traditional ECM. In PECM, high-current densities ( Mcm2) are pulsed on for durations on the order of 1ms and pulsed off for intern the order of 10 ms The relaxation (pulse off) interval permts reaction by-products removed from the interelectrode gap at low electrolyte flow rates without elect1 deposition on the ECM tool. As a result, high-current densities can be used at sm terelectrode distances, improving both removal rates and precision. Because PEC lows lower electrolyte flow rates, n has been proposed to remove recast layers fro surface of dies produced by electrical discharge machinmg (EDM) Some effor being made to Integrate this technology into EDM platforms Pulsed currents have also been used in elrctrochrmccul micromachinzng (E In one variation of the process, photolithographic masks are used to concentrate rial removal in selective areas of thin films. In addition to providing smoother su (on the order of electropolished surfaces) than those found in traditional E C E i high-current density (- 100 A/cm2)results in less taper in etching profiles and mar form material removal across the workpiece, whlch are Important for shrlnkingfr slzei in mlcromachining Electrochemzcul polishing is a modification of the ECM process that opt essentially the same as ECM, but with a much slower penetration rate. Current d is lowered, whlch greatly reduces the material removal rate and produces a fine on the order of 10 min. Electrochemical polishing must be differentiated electropol~shzng, which operates without the uye of a part-speclfic hard tool. Electrochemical Hole Machining. Several electrochemrcal hole-drdhng pro1 have been developed for drilhng small holes wlth high aspect (depth-to-diametc tios in difficult-to-machine materials. In elechostream drzllmg, large numbers o f (over 50) can be s~multaneously gang drilled in nickel and cobalt alloys w~th dian down to 0.005 in. at aspect ratios of 50:l Machining is performed by a high-ve stream of charged, acidlc electrolyte ejected from the capillary end of a drawn tube (the tool). An internal electrode (e.g., a small titanium wire) 1 s fed into the end of the tube and placed close to the capdlary. The electrode 1s used to char! electrolyte, which 1s pumped through the tube.An acidic electrolyte is used s t the sludge by-product goes Into solution Instead of clogging flow. Voltages used process are 10 to 20 times higher than that of typical ECM processes. This tech was originally developed to drill high-mcident-angle,cooling holes in turbine blad jet engines.

A second process, known as the shaped-tube electrolylrc rnachrning (STEM) process, was also created m response to unique challenges presented in the jet engine industry. L k e the electrostream process, STEM is also capable of gang drilhng small holes in difficult-to-machinematerials However, the STEM procevs is generally not capable of drilling holes smaller than about 0.02 in. STEM 1 s capable of making shaped holes with aspect ratios as h ~ g h as 300:l.Holes up to 24 in. m depth have been dr~lled. Llke the electrostream process,it uses an acidlc electrolyte to min~m~ze clogg~ng due to sludge buildup. The major differences between the STEM process and the electrostream process are the reduced voltage levels (5 to 10 V dc) and the specla1 electrodes, which are long, straight, metallic tubes coated with an ~nsulator (Figure 19-19).The ~nsulator helps to elinnuate taper by constrainmg the electrolytic action between the bottom of the tool and the workpiece. Titanlum is often used for its abihty to resist acids.The electrolyte 1s pressure-fed through the tube and returns through the gap (0.001to 0.002 m.) between the insulated tube wall and the hole wall. Electrolyte concennations may include up to 10% sulfunc acid. Lower concentrations may be used to Increase tool l i e
Electrochemical Grinding. Electrochem~calgnnding, commonly designated ECG, is a low-voltage,high-current variant of ECM in which thc tool cathode is a rotating, metalbonded, d~arnond gnt grind~ng wheel.The setup shown in Figure 19-20 for grind~ng a cutt~ng toolis typica1.A~ the electrlc current flobvs through the electrolyte between the workpiece and the wheel, some surface metal n removed electrochemically and some is changed to a metal oxide, which is ground away by the abrasives As the oxide film is
Metal tube electrode

STEM electrode

FIGURE 19-19 The shaped-tube electrolytic rnachlning (STEM) cell

ECM technique for drilling small holes usrng a process 1 s a special~zed metal tube electrode or metal tube electrode wlth dielectr~c coatlng.

508

CHAITER 19

Electronic Electrochemical Chemical and Thermal Machin~ng Processes Electrolyte-sodlum nltrate

I -

Electrode, conduct~ve wheel

Current pick-up brushes
\

Tank

I
FIGURE 19-20 Equipment

1
Filter
Pump

setup and electrical circuit for electrochemical grinding.

removed, new surface metal is exposed to electrolytic action. Most of the material removal is electrochemical, with only about 10% of the material being removed by grinding. The metal removal rate (MRR) is dependent on many variables. Table 19-4 gives some typical values for the MRR. The wheels used in ECG must be electrically conductive abrasive wheels. For most metals, resin-bonded aluminum oxide wheels are recommended. The resin bond is loaded with copper to provide for negligible electrical resistance.The wheels are dressable, using a variety of wheel-dressing measures. Once dressed, the wheels can then be used for precision form-grinding operations. The abrasive particles are hard,nonconducting materials such as aluminum oxide, diamond, or borazon (cubic boron nitride). In addition to increasing the efficiency of the process, the abrasives act as an insulating spacer,maintaining a separation of from 0.0005 to 0.003 in. (0.012 to 0.05 mm) between the electrodes. A dead short would result if the insulating particles were absent. The particles also serve to grind away any passivating oxide that should form during electrolysis.The process is used for shaping and sharpening carbide cutting tools, which cause high wear rates on expensive diamond wheels in normal grinding. E C G greatly reduces wheel wear.Fragile parts (e.g., honeycomb structures), surgical needles, and tips of assembled turbine blades have also been ECG-processed successfully.The lack of heat damage, burrs, and residual stresses is very beneficial, particularly when coupled with MRRs that are competitive with conventional grinding and far less wheel wear.
Nectrochemical Deburring. Electrochemical dehurring is a deburring process which works on the principle that electrolysis is accelerated in areas with small interelectrode gaps and prevented in areas with insulation between electrodes. The cathodic tool in electrochemical deburring is stationary and generally shaped as a negative of the workpiece to focus the electrolysis on the region of the workpiece where burrs are to be removed. Portions of the tool not used for deburring are coated with insulation to prevent the electrolytic reaction.The process does not require a feed mechanism. A fixture made of insulating material is used to position the workpiece with respect to the cathodic tool. Because of the small amount ofmaterial removed, electrochemical dehurring generally requires short cycle times under one minute. Design Factors in Electrochemical Machining. In general, current densities tend to concentrate at sharp edges or features and, therefore, produce rounded corners.Therefore, ECM processes should not be used to create sharp corners or pockets in the workpiece, although sharp corners are possible on through-features. Due to the need for an interelectrode gap, the actual feature size will be larger than the tool size.The overcut on the side of the tool is normally on the order of 0.005 in. Taper should be expected in all features due to electrolytic reaction between the side of the tool and the workpiece.

or Various Metals
Metal
Alum~num

Valency
3

Density Ib/rn3 gicm"

Metnl Removal Rate at 1000A

lblhr

rnT/mm

cm3/mm

Berylhum Chrom~um
Cobalt

2
2

3
6

Magnesium

Mangilnere

2
4

Nickel

Titmum
'Mgsten Uran~um

Vanadium
Z~no 2

0258

713

269

0174

2 85

Suzrm 1985 SCTE Cnnjrreme Placpdm#S,ASM,MetaLF Park, OH 1986

Depending on the tool deslgn, taper can be held to 0.001 m./in. For micromachin~ng applications, insulator material may be placed on the side of the tool to prevent the taper effects. Control of the electrolyte flow can be diff~cult in parts with irregular shapes. Changes in electrolyte concentration due to varying flow patterns can change the local resistance across the interelectrode gap, resulting in local variations in removal rates and tolerances. In addition, high electrolytic flow rates can cause eroslon of workpiece features. Therefore, complex geometries requiring tortuous interelectrode flow paths are discomged. . (;ener;jll!, no derfimcntiil cftccr, 10 the prop2rtir.c or' the workpiccr. rn;ilefi:~l arc expcctcd \\.h:n usin!; I-.('Xl. IIo\\c\'cr,thc.lack of con~prcwi\r. rcsitlu;\l >trcs\chi n ~ p a r ~ e ~ l tothe surface of the workpiece during operatlon can have a negative impad on the fatigue resistance of the part when compared with mechanically machined parts. Further, if parameters are set improperly, the process may favor electrolysis at the grain boundaries, resulting in intergranular attack, which may also have a negalive effect on the fatigue resistance of the part If powhle, shot peening or some simikar process may be used to impart compressive stresses to the surface of the workpiece, thereby improving its fatigueresistance,

disposal of environmentally harmful by-products.

CAL

DISCHARGE MACHINING
As early as the 1700s, Benjamin Franklin wrote of witnessing the removal of me electrical spark?. However, it was not until the 1940s that development of the ele discharge machining ( E D M ) process, also known as electric or electro disc machining, began in earnest.Today,it is one of the most widely used of the nontrad' machining processes.

workpiece, the high temperatures created by the spark also melt or vaporize th creating tool wear. The dielectric fluid is pumped through the interelectrode ga Servo to
control

FIGURE 19-21 EDM or spark erosion machining of metal, using high-frequency spark discha a dielectric, between the shaped tool (cathode) and the work (anode). The table can make X-Y

movements.

R E 19-22 Schematic m of equipment for wire using a moving wire

flushes out the a l p s and bubbles while confining the sparks. Once the highest point on the workpiece is removed, a subsequent spark is created between tho tool and the next highest point, and so the process proceeds into the workpiece. Literally hundreds of thousands of sparks may be generated per second.Thismaterial removal mechanism is described as spark erosion. Two different types of EDM exist based on the shape of the tool electrode used. In rum EDM, also known as sznker EDM or simply EDM, the electrode is a die in the shape ofthe negative of the cavity to be produced in a bulk material. By feeding the die into the workpiece, the shape of the die ismachined Into the workpiece. In wzre EDM, also known as electvicnl discharge wire curtmg,the electrode is a wire used for cutting through-cut featuresdriving the workpiece with a wmputer numerlcal controlled (CNC) table (Figure 19-22). W i r e ED M uses a continuousiy moving conductive wire a5 the tool eleBrode.The tensioned wire of copper, brass, tungsten, or molybdenum is used only once, traveling from a take-off spool to a take-up $pool qhile being "guided to produce a straight, Barrow kerf ~nplates up to 3 in thick. The wire diameler ranges from 0.002 to 0.01 in.,with positioning accuracy up to +0.00002 in. in machines with numerical control (NC) or tracer control,The dielectricis usually de~onized water because of its 1ow viscosity.This process is widely used for the manufacture of punches, dies, and stripper plates with modern machines capable of routinely cuttingdiereliet intricate openings, t~ght radius contours, andcomers See Figure 19-23for some examples. EDM processes are slow compared to more convenbonal methods of machining, and they produwe a matte surface finish composed of many small craErs. While feed rates in EDM are slow, EDMprw~zwscan stdl compete with convent~onal machining m producing cbmplex geometrie&particularlyin hardened tool materia1s.A~ a result, one of the biggest applications of EDM processes is tool and die making Another drawback of EDM is the formation of a recast or remelt layer on the surface and a heat-affected zone below the surface of the wqrkpiece. Figure 19-24 shows a scanning electron micrograph of a recast Layer on top of a ground surface. Note the small spheres in the lower-right corner attached to the surface, representing chips that did not escape the surface. Below the recast layer is a heat-affected zone on the order of 0.001 m. thick.The effect of the recast layer and heat-affected zone is poor surface finish as well as poor surface integrity and poor fatigue strength. MRR and surface fini8h are both controlled by the spark energy. hmodern EDM equipment, the spark energy is controlled by a DC power supply. The power supply works by pulsing the current on and off at certain frequencies [between 10 and 500 kHz).The on-time as a percentage of the rota1 cycfe time (inverse oP the frequency) o control the pulse voltage, is called the duty cycle. EDM power supplies must be ahLe t current, duration, duty cycle, frequency, and electrode polarity.The power supply controls the spark energy mainly by two parameters:current on-time and discharge current.

512

CHAPTER 19 Electronic Electrochemical Chemical and Thermal Mach~ning Processes

I

FlGURE 19-23 Examples of made on wire EDM workp~eces [\IC mach~ne ha tach^).

FIGURE 19-24 S E M

m~crographof EDM surface (r~ght) on top of a ground surface in steel The spheroidal nature of debr~s on the surface is in evidence around the craters
(300X).

F~gure 19-25 shows the effect of current on-time and discharge current on crater Larger craters are good for hrgh MRRs Conversely,small craters are good for operations.Therefore, generally, higher duty cycles and lower frequencles are us maxitruze MRR Further, higher frequencles and lower discharge currents are usel improve surface finish while reducing the MRR Higher frequencies generally ~ncrea~e tool d wear.

ci

SECTION 19.4 Electrical Discharge Machining

513

w i h tnc a l n o d r l t per soark

Mcta renioval rates llicrensc of energy

Pulse energy

The crater size decreases with

current on-tlme. D
2 2

19-25 The principle

100

50

05

removal for EDM.

n
2

Time in

&see

Pulse energy An estlmate for the MRR in ram EDM for a given workpiece mater~al has been given in Weller: (19-9) T:jZ3 C is a proportionality where MRR is the material removal rate in i ~ ~ . ~ / r(cm'lmin); nin constant equal to 5.08 in British standard units (39.86 in SI units); I is the discharge current in amps; and T, is the melting temperature of the workpiece material, "F ("C). Melting points of selected metals are shown inTable 19-5.While this equation shows that the MRR is based mainly on the discharge current, it is recognized that MRR is more a function of current density.For graphite electrodes a generally accepted rule of thumb for the maximum current density is 65 amps per square centimeter of surface area. Given these guidelines, an expression for the maximum MRR, MRR,,,, can be derived: MRR,,,
=

MRR

=-

C.I

Cmax.A,
~ 2 2 3

where C,, is a proportionality constant equal to 51.2 in British standard units (2591 in SI units) and is the bottom-facing surface area of the electrode in in.' (cm2).Thisformula suggests that the MRR in EDM is also a function of the tooling geometry.The larger and more highly contoured the electrode surface,the slower the MRR at the same discharge current.

514

CHAPTER 19

Electrontc Electrochem~cal Chemical and Thermal Mach~n~ng Processes

Alum~num Cabon steel cobalt Copper Mauganosc Molybdenum Nickel T~tnnlum TungGLm

The size of the cavlty cut by the EDM tool will be I than the tool. That is, the distance between the surface of electrode and thesurface of the workpiece represents the cut and is constrained by the mlnlmum interelectrode d ~ s Temperature, Metal "F (OC) necessary for a spark, which is essentially constant over all 1200 (660) of the electrode, regardless of sue or shape.Typlcal overcut zsoo (1371) ues range from 0 0005 to 0.02 m. Overcut depends on the 2696 (1480) voltage plus the chip slze, wh~ch varies with the amperage E 1980 (1082) equipment manufacturers publish overcut charts for the d 2300 (1260) ent power supplies for their machines, and these values c 4757 (2625) used by tool designers to detennlne the appropnate h e n s 2651 (1455) of the electrode.The dimens~ons of the tool are basically e 3308 (1820) to the desired d~menslons of the part less the overcut valu 6098 (3370) While different materials are used for the tool electro graphite is the most widely used, representing approxlrn 85% of electrodes.The choice of electrode mater~al dep on its machinab~lity and cost as well as the des~red MRR, surface finish.and tool w Equations 19-9 and 19-10 show that the most important mater~al characteristic for 1s meltmg temperature.This relationship also applie? to tool wear.The higher the Ing temperature of the electrode, the less tool wear (i.e.,material removed). In add1 to its good machinab~lity, graphite has a very high sublimat~on temperature (3500 which is good for minimizing tool wear. In addit~on to melting temperature, mate w~th high densities and high speclfic heats tend toward less tool wear Cheaper rn hc electrode materials with lower melting temperatures can be used in cases wher temperaturemetaIs are to bemachmed. And in some cases requring good ~urface a case can be made for metallic electrodes, since spark energies are generally lowe suiting in reduced temperatures m the interelectrode gap. Copper, brass, copper-t sten,aluminum, 70211-30Sn, and other alloys have all been employed as electrode ( materials for d~fferent reasons In addition to tool material selection, tool wear m a m~nimized by using the proper polarity across the electrode? To minimlre tool we most electrode materials, the tool should be kept posltive and the workpiece neg although larger MRRs are posblble with reversed polarity. The dielectric fluid has four main functions: electrical msulahon between th and workpiece; spark conductor: flushing medium; and coolant.The fluid must Ion prov~de a channel for the spark and deiomze quickly to become an insulator. Per the most unportant factor in EDM IS flushing of the interelectrode gap to remover ual materials and gas. Filters in the fluidic circuit are used to remove these wastes the dielectnc flu~d. In addition, the fluid must carry away the heat produce process. A gross temperature change in the dielectric f l u ~ d s~gnificantly the properties of the fluid Therefore, a heat exchanger 1s added to the fluidic a r remove heat from the dielectric fluid. Common d~electric flu~ds lnclude pa kerosene, and silicon-based dielectric oil. Polar compounds, such as glycerine (90:lO) w ~ t h triethylene oil as an add~t~ve, have been shown to lmprove the MR decrease the tool wear when compared with traditional dielectnc fluids, such as kero The d~electric materials must be safe for inhalation and skln contact because oper ark m constant In contact with the fluid.
Advantages andDisad~antages of EDM. EDM is applicable to all materials fh fairly good electr~cal conductors, including metals. alloys, and most carbides The ness, toughness, or brittleness of the material imposes no limitations EDM prov relat~vely simple method for making holes and pockets of any desired cross se ' materials that are too hard or too brittle to be machined by most other meth process leaves no bvrrs on the edges,About 80 to 90% of the EDM work perf the world is in the manufacture of tool and die sets for injection molding, forgmng,~ ing, and extrusions The absence of almost all mechanical forces makes it possi EDM fragile or delicate parts w~thout distortion EDM has been used in microm ing to make feature sues as small as 0.0004 in. (0.01 mm).

On most materials, the process produces a thm, hard recast surface, which may he an advantage or a disadvantage, depending on the use. When the workpiece material is one that tends to be brittle at room temperature, the surface may contain f i e cracks caused by the thermally induced stresses. Consequently, some other finishing process is often used subsequent to EDM to remove the thln recast and heat-affected layers,particularly if the product will be fatigued. Fumes, resulting from the bubbles produced during spark erosion, are given off during the EDM process. Fumes can be toxic when electrical discharge machm~ng boron carbide, tltanlum horide, and berylhum, posing a sign~ficant safety issue, although machining of these mater~als is hazardous m many other processes as well.

ELECTRON AND ION MACHINING

As a metals-processmg tool, the electron beam 1s used mainly for weldmng, to some extent tor surface hardening, and occasionally for cutting (mainly dnlling). Electron-beam machlnrng (EBM) is a thermal process that uses a beam of high-energy electrons focused on the workpiece to melt and vaporize metal.This process shown in F~gure 19-26 is performed m a vacuum chamber.The electron beam 1s produced in the electron gun (also under vaecum) by thermionic emlaslon. In its simplest form, a Cilament (tungsten or lanthanum-hexaboride) 1s heated to temperatures in excess of 200PC, where a stream (beam) of electrons (more than 1 blllion per second) 1s emitted from the hp of the filament. Electrostatic optics are used to focus and direct the beam.The desired beam path can be programmed wtth a computer to produce any deslred pattern in the workplece. The diameter of the beam is on the order of 0 0005 lo 0 001 in., and holes or narrow slits with depth-to-width ratlos of 100:l c a n be "machined" with greal prectslon m any material. The interaction of the beam with the surface produces dangerous X-rays; therefore, electromagnetic shielding o f the process is necessary.The layer of recast matenal and the depth of the heat damage are very small. For micromach~ning applications, MRRs can exceed that of EDM or ECM.Typica1 tolerance7 are about 10% of the hole diameter or slot width.These machrnes require high voltages (50 to 200 kV) lo accelerate the electrons to speeds of 0.5 to 0.8 the speed of light and should be operated by fully trained personnel. Ion-beam machcnlng (IBM) is a nano-scale (lo-') machining technology used m the m~croelectronlcs industry to cleave defectlve wafers for characterization and failure analysis. IBM uses a focused Ion beam created by thermionic em~ssion slmilar to EBM to mach~ne features as small as 50 um.The ion beam may be focused down to a 50-nm diameter and is focused and positioned by an electrostatic optics column Current densities up to 5 A em2 and voltages between 4 and 150 kV provide ion energies up to th~ck can be processed 300 A cmZkeV.Target substrates as large as 7" X 7" X i"

H~gh-voltage cable

I
Optical

I Anode
Valve

Electron-beam g u n

\
chamber

vlewlng system Electron beam Magnet~c lens to focus beam Deflectlon tolls

Vlewlng port
19-26 Electron-beam ig uses a high-energy beam (1O9 Wlin.?.

High-

Vacuum chamber

,,>#.,,,,m

Worktable

11

516

CHA~E 19 R Electronic Electrochem~cal Chemical and Thermal Machining Processes

Monochromic coherent

FIGURE 19-27

Schematic diagram of a laser-beam machine, a thermal N T M process thatcat micromachine any material.

LASER-BEAM MACHINING
Laser-beam machining ( L B M ) uses an intensely focused, coherent stream of laser) to vaporize or chemically ablate materials. A schematic of the LBM p shown in Figure 19-27. Lasers are also used for joining (welding, crazing, sol heat treating materials (see Chapters 5 and 36 for a discussion), and rapid prot (see Chapter 14). Power density and interaction time are the basic parameters processing, as shown in Figure 19-28.Drilling requires higher power densities interaction times compared to most other applications. The material removal mechanism in LBM is dependent on the of the laser used. At UV wavelengths (i.e., between about 200 and 400 material removal mechanism in polymers (for example) is generally ther poration. Below 400 nm, polymeric material is typically removed by chemical In ablation, the chemical bonds between atoms are broken by the excess a of laser energy absorbed by the valence electrons in the material. The adva of chemical ablation is that since it is not a thermal process, it does not resu heat-affected zone. Laser light is produced within a laser cavity, which is a highly reflective c taining a laser rod and a high-intensity light source,or laser 1amp.The light source is "pump up" the laser rod, which includes atoms of a lasing media that is capable of a ing the particular wavelength of light produced by the light source.When an atom of

Shock

FIGURE 19-28
O

POW@

Interaction time ( s e c ) m

in laser processing vary the application.

S E ~ T I19.4 ~ N Electrical Discharge Machining

Comments
Snibing thin films Power low Often uneconomical

2XO l S peak

Low ( 5 to 10)

0.2-7 ms

2 x 106peak

Low (0.2)

0.5-10 ms

Continuous

1000

1-24.000 50-250 ns 0.5-7 ms -

Large material removal in one pulse, drilling diamond dies, spot welding Largematerial removal in one pulse Welding

Often uneconomical Compact, econom~cal at low powers Compact and economical Very bulky at high powers Bulky but econam~cal Bulky but econom~cal Short wavelength, h~gh energy, and h~gh average power

Repstitively Q-switched Continuous

3 X 10' peak; 30 average

15,000

Repetitively Q-switched

75,OM)peak 1.5 average

400

50-200 ns
1 0 0 p and up

Resistor trimming, eledro~c Spot weld, drill Cutting organic materials, oxygenassisted metal Resistor trimming Welding, hole production, cutting Micromachining, industrial materials processing, and laser annealing

0.248

Pulse

Up to several
kW

100-1000

15-45 ns

;I:

h
&cimeo

0.308

Pulse

200

300

40-50 ns

p:Modified from I. E Ready, Selecting a l a m tar material working. Laser Focus (March 1970).p40

aluminum garnet.

media is struck by a photon of light, it becomes energized.When a second photon strikes the energized atom, the atom gives off two photons of identical wavelength,moving in the same direction and with the same phase. This process is called stimulated emission. As the two photons now stimulate further emission from other energized atoms, a cascading of stimulated emission ensues.Toincrease the number of stimulated emission$ the laser rod has mirrors on both ends that are precisely parallel to one another. Only photons moving perpendicular to these two mirrors stay within the laser rod, causing additional stimulated emission. One of the mirrors is partially transmissive and permits some percent of the laser energy to escape the cavity.The energy leaving the laser rod is the laser beam. Table 19-6 lists some commercially available lasers for material processing. The most common industrial laser is the C 0 2 laser. The C 0 2 laser is a gas laser that uses a tube of helium and carbon dioxide as the laser rod. Output is in the far-infrared range (10.6 pm), and the power can be up to 10 kW. Nd:YAG lasers are called solid-state lasers. The laser rod in these lasers is a solid crystal of yttrium, aluminum, and garnet that has been doped with neodymium atoms (the lasing media).The output wavelength is in the near-infrared range (1064 nm),and power up to500 W is common.For micromachining applications, modifications can be made to a 50-W Nd:YAG laser to output at one-half (532 nm), one-third (355 nm), and one-fourth (266 nm) of this wavelength, with roughly an order of magnitude reduction in power from 1064 to 532 to 266 nm. More recently, gas lasers, called excimer (from excited dimer) lasers, have been developed with laser rods consisting of excited complex molecules (usually noble gas halides) called dimers. Excimer lasers are pulsed lasers that output in the near and deep UV range

iic Electrochem~cal Chemical and Thermal Machin~ng Processes at powers up to 100W. Exc~mer lasers are ~ignifmantly more expensive to purchas operate than CO, or Nd:YAG lasers. Lasers produce highly coll~mated, coherent (rn-phase) light, which,when fo to a sinall diameter, produces high-power densities that are good formachmmng.It I erally accepted that m order to evaporate materials. infrared poww densities m e of 10' W/mmz are needed. For CO, lasers, these levels are directly achievable.Ho er, in Nd:YAG lasers, these high power condition? ~ o u l s~grubcantly d decrease of the laser lamp.Therefore, Nd:YAG lasers make use of a Q-sw~tch, which bre the continuous light stream into a series of higher power pulses Pulsed Nd:YAG may have peak powers in excess of 30 kW at 1064nrn and peak power densities cess of 2 X lo7W/mm2,which are easily enough for metal sublimatron Wrfh this tutude of power density, the thermal effects of LBM are mrnrmal. Apphcatrons of LBM are widely varied. Most CO, Industrial laser CNC mg centers can focus the beam down to a diameter of about 0.005 in. Appllca i these systems range from cigarette paper cuttmg to dr~llmg microholes in turblne blades to cnttrng steel plate for cham saw blades. Fur printed cxrcuit hoard and chi packaging applicat~ons, Nd:YAT;. exclmer, and now pulsed lavers are be~ng used holes down lo 0.001 in in milli~econds in polylmide or polyester films. However drtlling is limited to rather thrn stock (0.01 in.), as the cutting ?peed drops off rapidly penetrahon into thematerlal Hole depth-to-diameter ratios of 103 are common hole geometry is irregular. Recast and heat-affected zones exlst adjacent to th Deep UV exclmer and Nd.YAG lasers are ideal tor m~cromachinlng applications. UV lasers use prrmarily a chem~cal ablation mechanism for material removal. In tlon to provid~ng the potential for eliminating thermal effects in machining, dee lasers (having lower wavelengths) may be focused to tighter diameters than lase higher wavelengths. While less powerful than Infrared lasers, deep UV lasers (b of chemlcal ablatron mechanisms) experience much greater energy efficiencies tmg. Holes as small as 60 pin. (1 5 pm) in diameter have been machlned in thm-fi tenals with excimer lasers In LBM, the wavelength used to process the material is marnly determined b optical characteristics (reflectivity, absorptivity, and transmirsiv~ty)of the wor Not all materials can be machined by all lasers. Proteclive eyeware is necessary working aroundlaser equipment because of the potentlal damage to eyesight from direct or scattered laser hght.

PLASMA ARC C U U l N C Plasma mc cutting (PAC) uses a superheated stream of electrically Ionized gas to and remove material (Figu~e 19-29).The 20,000 to 50,0000Fplasma is created water-cooled nozzle by electr~cally Ionizing a suitable gas such as nitrogen, hy of these gases The process can be used on almost any con argon, or mixt~~res metal.The plasma arc 1s a mixture of tree electrons, posrtively charged Ions, andn atoms. The arc is lnrhated 1n atenfined gas-filled chamber by a h~gh-frequency The high-voltage, DC power sustains the arc, which exits from the nozzle at nearvelocity The workp~ece is electr~callypositive. The high-velocity gases melt and away the molten metal "chips." Dual-flow torches use a secondary gas or waters to asslst m bl~wing the molten metal out of the kerf, glving a cleaner cut. The pr may be performed underwater,using a large tank to hold the plates bemg cut.The assrsts in confining the arc and reducing smoke. The marn advantage of PAC is Mrld steel: m. thick can be cut at 125 in imm. Speed decreases with thickness G nozzle life and faster cutting speeds accompany the use of water-injection-type to Control of nozzle standoff from the workplece is important. One electrode size c used to machine a wide variety of materials and thicknesses by suitable adjustme the power level, gas Lype, gas flow rate, traverse speed, and flame angle. PAC is times called plasma-beam machining. PAC can machtne exotic materials at hrgh rates. Profile cuttlng of metals,par ly of stainless steel and aluminum, has been the most prominent commercial However, mlld steel, alloy steel, tttanium, bron~e, and most metals can be cut

SrclrON 19.4

Electrical Discharge M a c h ~ n ~ n g

519

Constricted arc

Standoff

-1 ' II l 1 8 l' l 1
;

Plasma

S h ~ e l d g a s

(secondary1

- u\lur*p ece

Plasma jet (molten metal removed)

rapidly. Multiple-torch cuts are possible on programmed or tracer-controlled cutting tables on plates up to 6 in. thick in stainless steel. Smooth cuts free from contaminants arc a PAC advantage. Well-attached dross on the underside of the cut can be a problem, and there will be a heat-affected zone (HAZ).The depth of the HAZis a function of the metal, its thickness, and the cutting speed. Surface heat treatment and metal joining are beginning to use the plasma torch. See Chapter 31 for more discussion on plasma arcprocesses. Some of the drawbacks of the PAC process include poor tolerances, tapered cuts, and double arcing, leading to premature wear on the nozzle. Precision PAC, also called high-defuzition plasma and fine plasma cutting,uses a special nozzle, where either a highflow vortex or a magnetic field causes the plasma to spin rapidly and stabilizes the plasma pressure (Figure 19-30).The iast-spinning plasma results in a finely defined beam

Arc Current Density: 60.000 ampslsq. ~n Kerf Width. -0 0 4 I1 mm) on 118" ( 3-mm) m ~ l steel d

FIGURE 19-30 Preclslon plasma cutting nozzle

Electronic Electrochemical Chemical and Thermal Machining Processes

FIGURE 19-31 Thermochemical machining process for the removal of

burrs and fins.

THERMAL DEBURRINC

gases are formed by cornbusting (with a spark plug) explosive mixtures of oxygen fuel (e.g., hatural gas) in a chamber holding the workpieces. A thermal shock w moving at Mach 8 (2700 mls; 6000 mph) with temperatures up to 3300C (6000"F), porizes the burr in about 25 milliseconds. Because of the intense heat, the burrs fins are unable to dissipate the heat fast enough to the surrounding workpiece and, sequently, sublimate.The workpiece remains unaffected and relatively cool becau its low surface-to-mass ratio and the short exposure time. Small amounts of metal removed from all exposed surfaces, and this must be permissible if the process is t used. Consequently, while large burrs and fins can be removed with this procesi, parts with thin cross sections. Maximum burr thickness should be about the t feature on the workpiece. Thermal deburring will remove burrs or fins from a wide range of materials it is particularly egeedive with materials of low thermal conductivity,which easily

vantage of thermal deburring is that fine burrs are removed much more quickly cheaply than if they were removed by hand. Uniformity of results and greater qu assurance over hand deburring are also advantages of thermal debumng. One outc of thermal deburring is that the part is coated with a fine oxide dust that can be remo easily by solvents. Capital costs can be several hundred thousand dollars, and the imum workpiece dimensions are on the order of 250 mrn (10 in.) by 690 mm (27 in.). Chapter 36 for additional discussions on deburring.

. IKey Words
ablate ball gr~d arrays boule butt lead chemical blankingimilling chemical machining (CHM) chip
chip shooter

clinching

contact printing Czochralski method diamond sawing

Review Questions etching fl~p-chip gel milling gull wing heat-affected zone (HAZ) h~gh-definition plasma integrated nrcult (IC) ion-beam machmmng (IBM) J-leads laser-beam mach~nmng (LBM) maskant nontradltlonal machmmng (Nm) overcut overhang p-type sem~conductors photochemcal blank~ng photochem~cal machrning (PW) photomask photores~sts pin grld arrays pin-~n-hole plasma arc cuttmg (PAC) plasma etchmg precision PAC pr~nted circuit board ram E D M reactive Ion etehing screen printing seed crystal shaped-tube electrolytic machining (STEM) solder balls spark erosion stepped machining surface mount thermal deburring through-hole vapor-phase soldering wafer testing wafers wire EDM (WEDM) wire cutting

521

ctor

I I Review Questions
b o w do the MRRs for most NTM processes compare to conventional metal cutting? .Whatare the steps in chemical machining using photosensitive resists? Why is it preferable in chemical machining to apply the etchant by spraying instead of immersion? ,What are the advantages of chemical blanking over regular ;blanking using punch and die methods? .'Howare multiple depths of cut (steps) produced by chemical machining? ;Would it be feasible to produce a groove 2 mm wide and 3 mm deep by chemical machining? A drawing calls for making a groove 23 mm wide and 3 mm ,deepby chemical machining. What should be the width of the opening in the maskant? ,Could an ordinary steel weldment he chemically machined? Why or why not? .How would . vou . produce a tapered section by chem~cal machining? What is the principal application of thermochemical machining? ' I sECM related to chemical machining? ,:What effect does work materia1 hardness have on the metal removal rate in ECM? ,What is the principal cause of tool wear in ECM? .:Would electrochemical grinding be a suitable process for sharpening ceramic tools? Why or why not? $Jponwhat factors does the metal removal rate depend in ECM? ,Why is the tool insulated in thc ECM schematic? {iWhatis the nature of the surface obtained by electrodischarge !machininge? ,:What is the principal advantage of using a moving wire ielectrode in electrodischarge machining? ;What effect would increasing the voltage have on the metal ,:removal rate in electrodischarge machining? Why? .:If the metal from which a part is to be made is quite brittle and the part will he subjected to repeated tens~le loads, would
21. 22. 23. 24. you select ECM or electro discharge machining for making it? Why o r why not? If you had to make several holes in a large number of delicate parts,would you prefer ECM, EDM, EBM,or LBM? Why? What process would you recommend to make many small holes in a very hard alloy where the holes will be used for cooling and venting? Explain (using a little physics andmetallurgy) why the "chips" in a thermal process like EDM are often hollow spheres? What is a semiconductor? In general, what technological breakthroughs were necessary to advance to each successive level of integration? What is a silicon baule? What is the most complicated, expensive, and critical step in microelectronics manufacturing? List the phololithographic steps necessary to produce a resist mask on a silicon substrate. List four requirements of a photoresist. What b undercutting? What are some possible defects that can result from underetching? From overetching? What is meant by the term chip in electronics manufacturing versus EDM? What drives the increase in component density and die area within microelectronic manulacturing? Why areclean rooms so important to microelectronicproce~~ing? What two subcomponents make up an IC package? What are the advantages of surface mount technology versus through-hole (or pin-in-hole) technology for attachment of 1C packages and discrete electrical components to boards? Name the two key classes of TH packages. Name the four different types of SM lead geometries, and discuss the advantages of each. List the key steps involved in conventional IC packaging. What is a printed circuit board (PCB)?

25.

26.
27. 28.
29. 30. 31. 32.
33.

34.
35. 36.

37. 38.
39. 40.

522

C H A ~19 R Electronic Electrochemrcal Chemrcal and Thermal Machining Processes

s a materials engineer for the War Eagle Extinguisher Company, Carlos has recently been made aware of severa potent~ally hazardous failures that have occurred in h ~ company's s products. Bourdon tube pressure gages (F~gure CS-19) are used to monitor the ~nternal pressure of the sodlum b~carbonate (dry chemlcai) fire ext~ngu~shers In several extlngurshers, a longltudlnal crack has formed along the axis of the bourdon tube, a curved tube of ell~ptical cross section that has been fabricated from phosphor bronze tubing. These cracks are partlcularly d~sturb~ng for several reasons. Flrst, they allow the file extingusher to lose pressure and become inoperable. More S~gri~ficanTly, however, the cracks allow the tube to deflect elasticelly in such a manner that the gage st111 Indicates high internal pressure.Thus, wh~le the extingu~sher has actually lost all Internal pressure and IS useless, the readlng on the gage

would cause the owner to belleve that he stdl had an oper able firef~ghtlng device. Carlos is trylng to determine the cause of these failu and suggest appropriate corrective measures. Can you help him? 1. What addltlonal ~nformat~on would you like to have regardingthe failed components, their fabrication h~story, and the~r servlce hlstory7 Why? 2. What might be some of the possiblecausesof these fa~lures? What type of evldence would support each poss~billty~ What fypes of addltlonal tests or investlgations might you propose to Carlos 3. Could these fa~lures have occurred rn "normaP use, or isit l~kely that sane form of negligence, abuse, or mlsuse was involved? 4. What possible correctlve or preventatlve measures might you suggest toCarlos to prevent a recurrence?

(a) S~hemat~cof a bourdon.tube

pressure gage

(b)Crosssectlon of bourdon tube

RTHOGONAL MACHINING

20.1 INTRODUCTION
Machining is the process of removing unwanted material from a workpiece in the form of chips. If the workpiece is metal, the process is often called metal cutting or metal removal. U.S. industries annually spend well over $100 billion to perform metal removal operations because the vast majority of manufactured products require machining at some stage in their production, ranging from relatively rough or nonprecision work, such as cleanup of castings or forgings, to high-precision work involving tolerances of 0.0001 in. or less and high-quality finishes. Thus machining undoubtedly is the most important of the basic manufacturing processes. . W.Taylor at Midvale steel in the 1880s, the process Beginning with the work of F has been the object of considerable research and experimentation that have led to improved understanding of the nature of both the process itself and the surfaces produced by it. While this research effort led to marked improvements in machining productivity, the complexity of the process has resulted in slow progress in obtaining a complete theory of chip formation. What makes this process so unique and difficult to analyze?

Prior workhardening greatly affects the process. Different materials behave differently. The process is asymmetrical and unconstrained, bounded only by the cutting tool. The level of strain is very large. The strain rate is very high. The process is sensitive to variations in tool geometry, tool material, temperature, environment (cutting fluids), and process dynamics (chatter and vibration). The objective ofthis chapter is to put all this in perspective for the practicing engineer.

523

INPUTS
Machine tool selection Lathe Milling machine D r i l l press .Grinder -Saw Broach Machining center Cutting tool parameters Tool design geometry -Tool angles N o s e radius E d g e radius Cutting forces

Machining processes

-.

Single-point cutting Multiple-edge tools

Workpiece parameters Predeformation (work hardening prior to machining) Metal type BCC, FCC, HCP SFE Purity

Onset of shear direction &: Power Surface finish Tool wear, failufm. Deflections Temperatures Vibrations Part size

Cutting parameters Depth of cut Speed Feed Environment Oxygen -Lubricant .Temperature

II. Orthogonal (two-force1 model

Macroindustriai studies performed on plates and tubes Microstudies carried out in microscopes using high-speed photography

Determinations Specific horsepower,H Flow stress, 7, Chip ratios, r, Shear front directions; Velocities (chip, shear, and so onl Friction coefficients,p

Workholder Fixtures Jigs Chucks Collets

Heat in tool

FIGURE 20-1 The fundamental inputs and outputs t o machining processes.

FUNDAMENTALS
T h e process of m e t a l c u t t i n g i s c o m p l e x because i t has such a w i d e v a r i e t y of inp w h i c h a r e l i s t e d in F i g u r e 20-1.The variables are: T h e m a c h i n e tool selected to p e r f o r m t h e process T h e c u t t i n g t o o l selected (geometry a n d material) T h e p r o p e r t i e s a n d parameters of t h e w o r k p i e c e T h e c u t t i n g parameters selected (speed, feed, d e p t h of cut) T h e workpiece

holding devices o r f i x t u r e s or jigs

A s w e c a n see from F i g u r e 20.1, t h e w i d e v a r i e t y of i n p u t s m o s t of w h i c h a r e c r i t i c a l t o satisfactory p e r f o r m a n c e of t h e c o T h e r e a r e seven basic chip f o r m a t i o n processes (see Figu

drilling, sawing, broaching, shaping (pluning), and grinding machining), discussed in Chapters 22-24 a n d 26-28. C h a p t e r 25

(also c a l l e d abr describe workho

-*
Work rotates
I I

Saw blade
Work

Tool feeds

Turning

Sawing

Cutter rotates Work feeds Gr~nding

Work

feeds

Milling

Reciprocat~ng

Tool feeds

Shap~ng

Planing

Broach

I; ; Z T
statlonary
Broaching

:URE 20-2 The seven bas~c

ich~ningprocesses used In ~pformation.

devices and Chapter 21 will provide additional insights on cutting tools. Usually the workpiece material is determined by the design engineer to meet the functional requirements of the part in service. The manufacturing engineer will often have to select the cutting-tool materials and workholder parameters and then cutting parameters based on that work material decision. Let us begin with the assumption that the workpiece material has been selected. To make the component, you decided to use a high-speed steel cutting tool for a turning operation (see Figure 20-3). For all metal cutting processes, it is necessary to determine the parameters, speed, feed, and depth of cut. The turning process will be used to introduce these terms. in genera1,speed (V) is the primary cutting motion, which relates the velocity of the cutting tool relative to the workpiece. It is generally given in units of surface feet per minute (sfpm), inches per minute (in.lmin), meters per minute (mlm), or meters per second (mls). Speed (Vj is shown with the heavy dark arrow. Feed (f,) is the amount of material removed per revolution or per pass of the tool ovcr the workpiece. In turning, feed is in inches per revolution, and the tool feeds parallel to the rotational axis of the workpiece. Depending on the process, feed units are inches per revolution, inches per cycle, inches per minute, or inches per tooth. Feed is shown with dashed arrows. The depth of cut (DOC) represents the third dimension. In turning, it is the distance the tool is plunged into the surface. I t is half the difference in the initial diameter, D l , and the final diameter, D2:

DOC

Dl

D, = d

(20.1)

~m 20 Fundamentals of Machining/Orthogonal Machining

NOTE
The rpm of the rotatlng workpi iS N+ I t establishes the culung speed V, at the tool, accordmng t NS=12V/xD.

The depth of cut, d, 1 s equal to

tool travels parallel to theax1

FIGURE 20-3 Turnlng a cyltndr~cal workpiece on a lathe requlres you to select the cuttlng f cut. speed, feed, and depth o

The selection of the cutting speed V determines the surface speedof the ro part that is related to the outer diameter of the workpiece. -irD,Ns v = ---

12

where D, is in inches, Vls speed in surface feet per minure, and N, is the revol minute (rpm) of the workpiece.The input to the lathe will be in revolutionsper of the spindle. Figure 20-3 shows a typical machine tool for the turning process, a lathe. pieces are held in workholding devices. (See Chapter 25 for details on the workholders.) In this example, a three-jaw cbuck is used to hold the work rotate it against the tool.%e chuck is attached to the spindle, which is driven gears by the motor. Before the actual values for speed and feed are selected, he cutting-to al and geometry must be selected. The cutting tool is used to machine (l.e., cut) piece and 1s the most critical cornpent. The geometry of a single point (sing1 edge) of a typical high-speed steel tool used in turning a found in Chapter 21. tool geometry is usuany ground onto high speedsteel blanks, depending on what is being machined. &ure 20-4,taken from Metcut's Machinabdsty Data Handb the MfEiIE starting values for cutting speed (sfpm or mlmin) and feed (ipr or a given depth of cut, a given work material (hardness), and a given process Notice how speed decreases as DOC or feed increases, and cutting speeds in carbide and coated-carbide tool materials. To process different metals, parameters to the machine tools must be determined.For the lathe, the input p

SECTION 20.2

Fundamentals

527

2 %
185
82s 525

,I8

. a
,007 ,015 020

CPlO CPM CC-7 Cc-h CC-6

425
32.5

Quwched

375 ZM 145 115

.0211

CC-6

.I8 .A0 .50

CPlO CPM CP30

375

Quenched Tcmpercd

300

325

.OZU

CC-6

425

185 U0

.18 .40

CPl0 CP20

550

,020
.I8

CC-6 CP10 CPZO CP30

320 21s 170

.4n
.SU

125
to

Normalized, Annwled

,625

450
10

.020
.I8 .dU .50

CC-6 CPlO CP20 CP30

Annwlcd

PZO

260 170 135

RE 20-4 Examples of a table for selection of speed and feed for turning. (Source: Metcut's Machinability Data Handbook.)

CHAPTER ZO

kU

entals

logon

are uuc, me reen, and me rpm value of the spindle. The rpm value depends on selection of the cutting speed V Rewrihng equation for N,: 12V 3.8V -N , = -= .irDi Dl Cutt~ng speed, feed, and DOCselection depend on perience and experimentation are required to find to begin is by consulting tables af recommended tables are arranged according L o the process be the hardness, and the cutting-tool material. The for solving turning problemsin the book.For in listed at the end of the book or cutting-tool m This table is for turning processes only. The amount of metal renloved per determines the DOC. In practice, roughing of DOC and feed and are run at a lower su recommendations of V and f, in both English and metrlc units based on the DOC n ed to perform the job.Table values are usually conservative and should be consld starting points for determinmg the operational parametersfor a process Once cutting speed V has been selected, equation 20-3 IS used to determi spindle rpm, N, The speed and feed can be used with the DOC to estimate the removal rate for the process, or MRR. For turnmg, the MRR is MRR

-- 12Vf,d

This is an approximate equation for MRR. to 600 m?/min.The MRR can be used to estimate the horsepower needed top cut, as will be shown later. For most processes, the MRR equation can be view

t
d

Turning Speed, stated in surface feet per minute (sfpmi. IS the per~pheral speed at the cuttlng edge Feed per revolution in tumlng is a linear motion of the tool parallel to the rotating axis of the workplace The depth of cut reflectsthe third dlrnens~on

Ira

'1 Depth of cut

L= lengthof cut

Tm = L+A
fr&

Boring Enlarging hole of diameter Dl to D2.Borlng can be done w ~ t h rnultl tlng tools. Feed In Inches per rew

Machined surface

+idW

Tool

FIGURE 20-5 Relationship of speed, feed, and depth of cut i n turnlng, borlng, faclng, and cutoff operations typically done on a lathe.

Facing Tool feeds to center of workpiece su L = 012.The cuttlnn speed is decreasing as the tool approaches the center of the workptece. Grooving, paning, or cutoff Tool feed perpendlcuiar to the a tion The w ~ d t h of the tool produ depth of cut (DOC)

volume of metal removed divided by the time needed to remove it volume of cut MRR =

Tm

where T, 1s the cutting tune in mmutes.For turning, the cuttlng tlme depends upon the length of cut L d~vided by the rate of traverse of the cuttlng tool past the rotatlng workplecefSJ,, as shown in Rgure 20-5. Therefore, L + allowance T, =

fJ's

An allowance is usually added to the L term to allow for the tool to enter and exit the cut. Turning is an example of a single-point tool process, as is shaping. Milling and drilling are examples of multiple-point tool processes. Figures 20-5 through 20-9 show the basic process schematically. Speed (V) is shown in these figures with a dark heavy arrow.Feed ( f ) is the amount of material removed per pass of the tool over the workpiece and is shown as a dashed arrow. For many of the basic processes, the equations for T, and MRR are given. These equations are commonly referred to as shop equations and are as fundamental as the processes themselves, so the student should be as familiar with them as with the basic processes. If one keeps track of the units and visualizes the process, the equations are, for the most part, straightforward. See Table 20-1 for summary. In addition to turning, other operations can be performed on the lathe. For example, as shown in Figure 20-5, a flat surface on the rotating part can be produced by facing or a cutoff operation. Boring can produce an enlarged hole, and grooving puts a slot in the workpiece. The process of milling requires two figures because it takes different forms

Parameter
Ct%tting cpced, fpm Revoluhons per wnute, N. Fwd rat% ~ni m ~ n Feed perrev tooth pass, IU /rev Cutting tune, nun, T, Rate of metal remm d, m?/mimm Horsepower required at spindle Horsepower requued at motor Torque at spindle Symbols

Turnlng
V = 0 262 X

Milling
D,
X

Drllling
V = 0.262 X D, X

Broachmg
V

?*=02&'YDmX

rpm

rpm

rpm

f,=f,xrpm f ,
T,"
=

LY,

T,

uf,
MRR = wDadi4

MRR=12XdXfr

xK

MRR=wXdXf,, hp
=

x f",
hp = MRR X HP.

MRR=12XwXd XV

hp = MKRX HP, hp, = MRR X HPJE t, - 63,030 hphpm

MRR X HP,

hp, = MRR X MPJE fa - ( S , ~ O hplr~m

hp,, = iMRR X

HP, IE

D,= hameter of workp~ecein

turnmg, mches D, = Diameter of m~lling cutter, mches D, = D~amerer of dull,~nchas d = Depth of cut.tnches E = Effmency of spmdle dnve f6 = Peed mte, inches per mrnute f, = Feed,mwhes per rwolullon 6 = Fwd,m~hesper moth hp, = Horsepo\uer at motor MRR = Metal removal rate,in '/nun

hp =horsepower at sp~ndle L = Length of cut, Inches n = Number of teeth m cutter horsepower

N , =Revoluhon per m~nuteof work or

cutter
1, = Torque at sp~ndle, mch-pound

T, = C u n ~ a tune, g mrnutes V = Cutting speed, feet per minute w =W~dth dcut,tnches

Values forspedhc iwrsepewer (un~t power) are p e a mTable2W

Cutting e d g e d

Vertical spindle

Workpiece Slab milling - multiple tooth Slab milling is usually performed on a horizontal m ~ l l l n g mach~ne. Equations for,,T , and MRR derived in Chapter 25. The tool rotates at rpm N ., The workp~ece translates past the cutter at feed rate , , f the table feed. The length of cut, L, 1 s the length of workp~ece plus allowance, LA, Face m i l l ~ n g Mult~ple-tooth cutting Given a selected cutting speed Vand a feed per tooth f, the rpm of the cutter is N,= 1 2 V I ~ D f oa r cutting of diameter D. The table feed rate is = ,f ftnN,for a cutterwith n teeth. The cutting time, T , = 1L + LA + L,)lfrn W < Dl2 whereL, = L A = /-for or L, = LA = 012 for W* 012. d f , where d= depth of cut. The M R R = W

Tm = (L + LA)/frn The MRR = Wdf,,,where and d= depth of cut. W = wldth of the cut

FIGURE 20-6 Basics of milling processes (slab, face, and e n d milling) including equations for cutting time and metal removal rate (MRR).

a = rake angle

Select cutting speed V, fpm and feed, f , in./rev. Select drill. D = diameter of the drill which rotates 2 cutting edges at rpm N,V= velocity of outer edge of the lip ofthe drill. N,= 12Vl~D. = T , cutting time = (L+Allf,N, , is the feed rate in in, per rev. The where f allowance A = DI2. The MRR = (nD214)f,Ns in? l m i n which is approximately 3DVfi.

Drilling multiple-edge tool

FIGURE 20-7 Basics of the d r i l l ~ n g (hole-maklng) processes, including equations for cuttlng t i m e ano metal removal rate (MRR).

1
Process basics of ling. Equations for cutting id metal removal rate are developed in ?r 26.
E 20-8
t

Tool

The T, for broachin$hr~ T, L112V. The MRR (per toothi is 1 2 t W in3/min where V = cutting velociv in fpng, W is the width of cut, t = rise per tooth.

Depth

The tool cuts at velocity Vwlth a return of VR dictated by the rpm of the veloc~ty speed V = ( I + A)N,IlZR, crank, N,. The cun~ng where R,= stroke ratlo = 200/360" and the length of stroke is I = L + ALLOW. The tool feed IS f, Inches per stroke T , = WINsfc MRR = LdN,f, in3/min

>

,
//
/
t

i
4
/1

Shaper (quick return) mechanism for drlvlng tool pastwork.

'!

FIGURE 20-9 (a) Baslcs of the shaping process, lncludlng equations for cuttlng tlme (T,,) and metal removal rate (MRR). (b) The relat~onsh~p of the crank rpm N, to the cuttlng veloclty V

depending on the selection of the machine tool and the cutting tool. Milling, a multiple tooth process, has two feeds the amount of metal an ~ndividual tooth removes,called the feed per toothf,, and the rate at wh~ch the table translates past the rotating too1,called the table feed rate t;,, , in inches per minute. It a calculated from
fm

finNr

where n is the number of teeth in a cutter and N, is the rpm value of the cutter. Just as was shown for turning, standard tables of speeds and feeds for mill~ng provide values fo the recommended cutt~ng speeds and feeds per tooth,f,.

Operation Turnina

Block d~aaram

Most wmmonly used machines Lathe NC lathe machining center

Machines less frequently used Bor~ncl - mlll

Grinding

'

b=K7Y,
Work

Cyllndrlcal grinder

Sawlng (of plates and sheets)

C ~ n t o U rOr band saw

Laser Flame cuttlng Plasma arc

Drlllrng

Drill press Mdchlnlng center (nc) Vert m~llrng mach~ne Lathe Boring mill

Lathe Hor~zontal boring machlne

Lathe Dr~ll press

Boring mill

Reaming

Horizontal borlng rnachlne Machlnlng center

Grinding

Sawing

Contour or band saw

Broaching

Broaching machine

Arbor press (keyway broaching)

FIGURE 20-10

Operat~ons and machines used for machining cylmdrlcal surfaces.

Figure 20-10 prowdes an overvlew of t h e basic machlning processe 20 ~cal machlne tools that can generate cylindrical surfaces. F~gure overview of the bas~c processes that can generate flat surfaces. Table summary on typical sizes (min-max), the production rates (partihour), clslon or repeatability), a n d surface finish (roughness). M i l h g has prett shaprng and plan~ng, although gear shapmg is stdl a viable process. Millir

Most

Machines

Operation

Block diagram

commonly used machines

less
frequently
used

Machines seldom used

Broaching

Broaching
mach~ne
Surface grinder

Turret broach

Grinding

Lathe (with speciai attachment)

Contour saw

Sawing

Cutoff saw

Shaping
Work

Hor~rontal shaper Planer

Vert~cal

shaper

Planing

slab mill~na
Milling

Milling

mach'ne

Lathe with special m~lling tools

Lathe w ~ t h

face m~iling
Well

M~ll~ng
mach~ne Machlnlng

spec~al
mill~ng tools

center

Drill press il~ght cuts)

FlGURE 20.11

Operat~ons and machines used to generate flat surfaces

other rotational multiple-edge tool processes (drilhng or reamlng) is often performed in machimng centers rather than on milling machmes The turret lathe has been replaced mult~ple turrets in many factories. by CNC turmng centers' w ~ t h

Most of the cuttlng operations process descr~bed to this point are examples of oblique,or fotmation three-force, cutting. The cuttmg force system m a conventional, oblique-ch~p process is shown schematically In Figure 20-12. Obl~que cutting has three components:

1. F; Primary cuttlng force actmg in the direction of the cuttlngvelocityvector.% force

is generally the largest force and accounuj for 99% of the power requued by the process. 2. Ff. Feed force acting in the direction of the tool feed.Tms force is usually about 50% of F, but accounts for only a small percentage of thc power required because feed rates are usuaIly small compared to cuttmg speeds. 3. F,. radialar thrust force acting perpendicular to the machined surface. This force 1s typ~cally about 50% of F, and contributes very little to power requirements becausevelocity in the radial direction is negligible. Figure 20-12 shows the general relationship between these forces and changes in speed, feed, and depth of cut. Note that these figures cannot be used to determule forces for a specific process.
'Maehmtng centers and turnzng centers are NC or CNC machines, discussed m Chapter 27

Raw
MatexiaI

Typical

Form
Cylinders, prebrms, casti

Miinllm 1 . m.typical

Typical Tolerance All ferrous and nonfe~rons masetfa1wnsidered machmable

+8.002 in on dra. common, *U3WI in abttiunabk !kO.OOI m.on dia.

Surface

Rou$hnms a 5 2 5 0

iww
Em, rod. tube.
pmfoms 38 a . du X 03 in. bng Genetally 2 in di8. X h m. long Gallets adapt
3i s .dm.
eashng,
I

a m . dia.

Ang materisl wtth good maChmab~Iity rating

2n) marcr~al unth good

Bar, rod

$in. &a. and less

weight less than 1 ounce CollPts adapt to less than; in.
2m

where needed, iU.0005 m. passrhle iO,Ii005 in, posstble

t0.0lY to t O ' W 3 m. camon tO.ML% in to tQ001 la. common

nlachinshtlny r ~ t l o
Any m;llcrl:tl xith pood

Tbming ( S u k
automatle

63 and
betby 90-2@

mdrhmahilily tatlng All 3 i. m and aon6:rrou.s

wbmgb
Boring [veftkal)

98 in. x 7 2 m .

x 12%n

tO.NDS in.

preforms %plate, rod' tube

h k e d usually by aVSt~i to hold pmt

63-ZO
Any maBnal wiffi goad

Bhks
preforms, rods Pla& bar, preforms Sawing Bmaching Griadrng Wib.piate, sheet 'hhe,rod, bar, plate Plete,r6& bars

D . Z Wi n .dia
0m - m . drill dl&

63

mcbinahrhcy ratmg
Any unhardened inat&al: oarbides needed fqr Bome cO.W2-zkO 010in. cammos: tOAD1 m p~&bIe tO.OlS m po~slble

ease-hardened parts
0 . 0 1 0 in thi&
1 in.
0.020 io. dia.

A a y nonhardened maenat,

Shaping

Bar, plate,
Gastlng

Lmtted usually by ability to hvld pact

Any material wrrh good mashinabilltyraring Nearly all rnetallic matemaisplu~ many nometalkc Low-to medim-aarhon

OSnOl in. and less

fO.rn1-20 002 in.

Bar, plate,
casting
Gear shaping

~ftwwidexlsft
high X 76 ft long

Par@toolarge for Sh*F work

1m.d~a.

psras b w -t o m e d m n e b o o

(larpr pans) COXMO1-~LI OaQS a. @mall-medtum parts) -tO.OD1-t-&ObS in.

Blanks

t M C partshow

steels ar nonferrous materrals k t hymaterialwth gOod machmawity rathg

+O MIin. or better 4t 200D.P. to O,W65 in. at

3Q D,E
--

100 0

k
Radjal force 0 100 500 V, Cutt~ng speed -fi per rnln 1000

y,
1000 800 600 400 -

Depth of cut - in

e
0
m

1 ,
0 "z

200

100 60 40 20 0.001

1
0.002

I I I 0.005 0.01 0.02 , f Feed - in, per rev

1 0.05

0.1

3 Force
Fc
= Cuning force (vertical)

FR
FF

=Radial force (thrust)

= Feed force

h e power required for cutting is

F,V(ft-lhlmin)

The horsepower at the spindle of the machine is therefore

In metal cutting a very useful parameter is called the unit. or specific, horsepower HP,, which is defined as HP,

hp (hpiin?irnin) =-

MRR

In turning, for example, where MRR " 12Vf,,d, then

HP,

Fc 396,OOOf , d

Thus this term represents the approximate power needed at the spindle to remo cubic inch of metal per minute. Values for specific horsepower HF',, which is also called unit power, are gi Table 20-3. These values are obtained through orthogonal metalcutting expen described later in this chapter.

Material
Nonalloy carbon steel

Alloy steel

Stainless stt-el,annealed Steel castlngs

Stainlesssteel, annealed Heat-resistant alloy5

Hard steel Malleable iron Cast iron,low tenclle Cast Iron, hlgh tenslle Nodular SG uon Chilled cast Eon Alum~num alloys Alummum alloys (cast) Bronze-brass alloys

C 0.15% C 0.35% C.0.6(1% Annealed Hardened and tempered Hardened and tempered Hardened and tempered Annedlcd Hardened Martens~udfrrrit~c Nonalloy Low-alloy Hlgb-alloy Austenltlc Annealed Aged-Iron based Annealed-N~ckel or cobalt Aged Hardened sreel Manganese steel 12% Ferrlt~c Peariltic

Non-heat-treatable Heat-treatable Non-heat treatable Heat treatable

Z~nc alloy T~tanlum

Lead alloys,Pb>l% Brass, carlr~dge bras7 Bronze and lead-free copper Includes Electrolytrc copper Dieeast

25

1.8-2 0
0 31-0 83

0 90 025 034

Values assume normal feed ranses and sharp tools Mulllply values by 1 25 for a dull tool Caiculation of unll power (HP,)
HP
=

F,V11?000

HP, = HPIMRR Where

MRR = 1ZViw fortube tlllnrng

HP,
=

F VIlZVrw
=

33000 = )Jtw X 146000

Calcuiauon of specific energy (01

F.VWlw

rJrwfor tube tumxns

Specific power can be used in a number of ways Fust, it can be used to estimate the motor horsepower required to perform a machining operation for a given material HP,values from the table are multiplied by the approximate MRR for the process The motor horsepower HP,, is then HP, X MRR X C F HP, = E where E is the efficiency of the machine. The E factor accounts for the power needed to overcome friction and inertia in the machine and drive moving parts.Usually,80% is used. Usually the maximum MRR is used in this calculation. Correction factors (CFs) may also be used to account for variations in cutting speed, feed, and rake angle. There is usually a tool wear correction factor of 1.25,used to account for the fact that dull tools use more power than sharp tools. The primary cutting force F, can be roughly estimated according to HP, X MRR X 33,000 F,

This type of estimate of the major force F, is useful in analysis of deflection and vibration problems in machining and in the proper design of workholding devices,because these devices must be able to resist movement and deflection of the part during the process. In general, increasing the speed, the feed, or the depth of cut will increase the power requirement. Doubling the speed doubles the horsepower directly. Doubling the feed or the depth of cut doubles the cutting force F,. In genera1,increasing the speed does not increase the cutting force F,, a surprising experimental result. However, speed has a strong effect on tool life because most of the input energy is converted into heat, which raises the temperature of the chip, the work, and the tool, to the latter's detriment.Too1 life (or tool death) is discussed in Chapter 22. Equation 20-12 can be used to estimate the maximum depth of cut, &for a process as limited by the available power.
dm"
"=

HP, X E 12HP, VF, (CF)

(20.13)

Another handbook value useful in chatter or vibration calculations is cutting stiffness K,. In this text, the term specific energy U will be used interchangeably with cutting stiffness K,. It is interesting to compute the total specific energy in the process and determine how it is distributed between the primary shear and the secondary shear that occurs at the interface between the chip and the tool. It is safe to assume that the majority of the input energy is consumed by these two regions. Therefore, U
=

U,

+ U/
K, (turning)

where specific energy (also called cutting stiffness) is

u = - Fc The specific shear energy is

vf, d

- - Fc =

f,d

where V ,is the shear velocity and F, 1s the shear force. Specific friction energy is

where V, is the chip velocity and r, is the chip thickness ratio. See equation 20-18 for the calculation of r,.

Insulated jaws
Tubular!

/Center

-v
Dynamometer to measure forces
Milling

" 5 rm
Chuck
2.channel

Dynamometer
Oscillograph

machine table

amplif~er force stn

Readout
(a) OPM

' d
1
w&

-Cutting

V(Frontview1See Figure 21-14

(bl O T T (Top view) See Fic

~huck

r ~ o l t

FIGURE 20.1 3

Three ways to perform orthogonal machining. (a) Orthogonal plate mach~nlng on a hor~zontal m~ll~ng machine, good for low-speed cutting. (b) Orthogonal tube turning on a lathe, hlgh-speed cutting (see Figure 20-16). (c) Orthogonal dlsk mach~n~ng on a lathe; very h~gh-speed machining with tool feeding (~pr) In the facing d~rection.

-Dynamomet

(c) ODM (Topview)

Usually, 30 to 40% of the total energy goes into frlction and 60 to 70% shear process. Typical values for U are given in Table 20-3. This 1s expenmental data d by the orthogonal machining experiment described in the next section.

ir Orthogonal machining (OM) is carried out mostly in research labo~atories, better understand this complex process. In OM, the tool geometry 1s simpli the three-dimensional (oblique) geometry, as shown in Figure 20-1. Using this simplified tool geometry, metals can be cut to test machi chanics and theory. There are basically three orthogonal machining setups, ln F~gure 20-13.
cutt 2. Orthogonal Tube Turning end-cuttmg a tube wall in a turning s e t u p medium-speed ranges 3. Orthogonal Disk Machining end-cutting a plate feeding in a facing duecti~ high-speed cutting In oblrque machining, as in shaping, drilling, and single-point turning, the cut and the cutting motion are not perpendicular to each other. l n the orthogona cutting velocity vector and the cutting edge are perpendicular. The OPM 1

I. Orthogonal Plate Machining a plate in a milling machine-low-speed

SECTTON 20.4

Orthogonal Machining (Two Forces)

539

Workp~ece (plate) Dynamometer Traverse feed

for the

Cutt~ng tool

Overarm
Tool and QSO hammer holder

QSD hammer -Table Power traverse feed

Handle for the cross feed

VCutting speed

-Handle

-Tool

holder

Workpiece holdel Levers for select

Handle for the Infeed lvert~cal movement of the table)

(a)

Endview of OPM setup on horizontal mill. Tabte feed is used for cuttlng speed

Machlne bed

(b) Hor~zontal mllllng setup for OPM uslng

QSD Front vlew.

g machine. (c) Orthogonal machin~ng with fixed tool, plate. The feed ism of the mill is used to iuce low cutting speeds. The of the tool is t and the DOC the width of the plate.

Onset of s h e a r plane

Onset of shear angle

WOrkplece

- I
v
(plate)
J

Ic) End vbew (on left) and slde vlew of orthogonal plate machln~ng w~th f~xedtool a n d movlng plate

order to ied from ' ning meas shown

tting I case
OW-S

plate machining is shown in more detail in Figure 20-14, using a modified horizontal milling machine where the table traverse provides the cutting speed and the tool is mounted in a tool holder in the overarm. As shown in Figure 20.15, OTT can be done on solid cylinders that have had a groove machined on the end to form a tube wall w, or a tubular workpiece can be used. The tubular workpieces can be mounted in a lathe and normal cutting speeds developed for the machining experiment. This setup has the advantage of being very easy to modify so that cutting-temperature experiments can be performed, using the toollchip thermocouple method. The orthogonal case is more easily modeled for temperature experiments. Low-speed orthogonal plate machining nses a flat plate setup in a milling machine.The workpiece moves past the tool at velocity V.The feed of the work up into the tool is now called r, the uncut chip thickness.The DOC is the width of the plate w. The cutting edge of the tool is perpendicular to the direction of motion V.The angle that the tool makes with respect to a vertical from the workpiece is called the back rake angle a.A positive angleis shown in the schematic.The chip is formed by shearing.The onset of shear occurs at a low boundary deformed by angle 4 with respect to the horizontal.This model is sufficient to allow us to consider

Fundamentals o f Mach~n~ngIOrthogonal Machlnlng

w = wall thickness

F, FIGURE 20-15 Orthogonal

Cutt~ng force

@ In ton vtew

tube turning (OTT) produces a two-forcecutting operation at speeds equivalent to those used in most oblique machining operations. The slight difference in cutting speed between the inside and outside edge of the chip can be neglected.

the behavior of the work material during chip formation, the ~nfluence of the rn ical elements of the tool geometry (the edge radius of the cutting tool and the ba angle a),and the interactions that occur between the tool and the freshly genera faces of the chip against the mke face and the new surface as rubbed by the fl the tool. narrow zone. This large-strain, high-strain-rate, plastrc deformation evolves

in annealed metals. The applied stress level increases as the material apptoa

agrees with the material behavior observed in tensilelcompression testing-

SLLTION 20.4

Orthogonal Machining (Two Force)

541

Videograph ~rthogonal plate

ess.

Schematic
~fthe material

chip-forming defines the lower boundary. -ection of slip )n movement.

(increased di$location dens~ty)In metalcutting, we observe that the onqet of shear (to form the c h ~ p is ) delayed by increased hardness (so &increases d~rectly w~th hardness). As the mater~al bemg machined gets harder, dislocat~on motion becomes more difficult and plastic delormation (wah continuous chips) glves way to fracture (d~scon20-18 fot examples of chtps If tinuous ch~ps) just as it docs in tens~le testlng See F~gure the work material has hard second-phase particles dispcrscd ~nit, (hey can act as barriers to the shear front dislocations, whlch cannot penetlate the particle The dislocations createvoid$ around the particles. If there are enough part~cles oi the right size and shape, the c h ~ p will fracture through the shear zonc, forming segmented chrps. Free-muchmmg steels, which have small percentages of hard second-pha5e particles added to them, use this metallurgical phenomenon to bredk up the c h ~ p for s easiel chip handling.

FIGURE 20-18 Three c types of cnlpc c113rx~er1)~ D scolitln,ous, (lclt LO r~qhi) continuois, and continuous with built-up edge. Chip samples produced by quick-stop technique. (Courtesy of Eugene Merchant (deceased) at Cincinnati Milacron, Inc., Ohio.)

For the purpose of modelmg c h ~ p formation, assume that the shear process takes on a single narrow plane, shown in Figure 20-19 as A-B rather than o n the set of fronts that actually comprise a narrow shear zone. Further, assume that the tool' ting edge is perfectly sharp and no contact is being made between the flank of th( and the new surface. The workpiece passes the tool w t h velocity V, the cuttmg 4 The uncut chip thickness is r. lgnormg the compresston deformation, chips having1 ness f, are formed by the shear process.The c h ~ p has velocity Vc.The shear process has velocity V, and occurs at the onset of shear angle +The tool geometry is ~ N I the back rake angle a and the clearance angle y.The velocity triangle for V, V,, ar is also shown (see Figure 20-19).The chip makes contact with the rake face of tht over length I, The plate thickness is w. From orthogonal machin~ng experiments, the chip th~ckness l b measured an4 to compute the shear angle from the chip thrckness ratlo, r,,defined as Vt';

t, ABcos(d,-a) p the free surface. where AB is the length of the shear plane from the tool t ~ to Equation 20-18 may be solved for the shear angle 4 as a function of the able chip thickness ratio by expanding the coslne term and simplifying:

,.

t =-=

AB sin Q,

r, cos a tan $ = 1 - r, sin ru

ii

FIGURE 20-19 Velocity d~agram associated with Merchant's orthogonal mach~ning model.

There are numerous other ways to measure chip ratios and obtain shear angles both during (dynamically) and after (statically) the cutting process. For example, the ratio of the length of the chip L, to the length of the cut L can be used to determine 7,. Many researchers use the chip compression ratio, which is the reciprocal of r,, as a parameter. See Problem 2 at the end of the chapter for another method.The shear angle can be measured statically by instantaneously interrupting the cut through the use of quick-stop devices.These devices disengage the cutting tool from the u~orkpiece while cutting is in progress, leaving the chip attached to the workpiece. Optical and scanning electron microscopy is then used to observe the direction of shear. Figure 20-14 shows a QSD on an OPM setup, and Figure 20-18 was made using a quick-stop device. Highspeed motion pictures and high-speed videographic systems have also been used to observe the process at frame rates as high as 30,000 frames per second.Figure 20-16 is a high-speed videograph. Machining stages have been built that allow the process to he performed inside a scanning electron microscope and recorded on videotapes for highresolution, high-magnification examination of the deformation process. Using sophisticated electronics and slow-motion playback, this technique can be used to measure the shear velocity.The vector sum of V ,and Vcequals V. For consistency of volume, we observe that
y =-= <"

sin cos(4 - a )

v. v

~ndicating that the chip ratlo (and therefore the onset of shear angle) can be determined dynamically if a rel~able means to measure V , can be found. The ratio of V , to V is cos CY -V cos(4 - a )

v,

These velocities are important in power calculations, heat and temperature calculations, and vibration analysis associated with chatter in chip formation.

Orthogonal machining has been defined as a two-force system. Consider Figure 20-20, which shows a free-body diagram of a chip that has been separated at a shear plane. It is assumed that the resultant force R acting on the back of the chip is equal and opposite to the resultant force R' acting on the shear plane.The resultant R is composed of thefriction force Fand the normal force N acting on the tool-chip interface contact area.The resultant force R' is composed of a shear farce F, and normal force F, acting on the shear plane area A,. Since neither o f these two sets of forces can usually be measured, a third set is needed, which can be measured using a dynamometer (force transducer) mounted either in the workholder or the tool holder. Note that this set has resultant R, which is equal in magnitude to all the other resultant forces in the diagram. The resultant force R is composed of a cutting force F, and a tangential (normal) force F,. Now it is necessary to express the desired forces (F,, F,,, F, N) in terns of the measured dynamometer components, F, and F,,and appropriate angles.To do this, a circular force diagram is developed in which all six forces are collected in the same force circle (Figure 20-21).The only symbol in this figure as yet undefined is 0,which is the angle between the normal force Nand the resultanl R. It is called friction angle P and is used to describe the friction coefficient p on the tool-chip interface area, which is defined as FIN so that

The friction force F a n d its normal N can be shown to be

F , cos a - F, sin a

Work

FIGURE 20-20 Free-body dlagram of orthogonal chrp formation process, showing equ~l~brlum condttlon between resultant forces R and R'.

FIGURE 20-21 Merchant's clrcularforce dlagram used derlve equations for F,, F , F,, and N as functions of F,, F, and P

and the resultant R is

R = m

Notice that in the special situation where the back rake angle is zero, F = F, ap F,, so that in this orientation, the friction force and its normal can be d~rectly m s t by the dynamometer. a The forces parallel and perpendicular to the shear plane can be shown f circular force diagram (Figure 20-21) to be

F,

F, sin 4

+ F, cos 4

F, is of particular interest, because it is used to compute the &ear stress on the plane. This shear stress is defined as

where A, is the area of the shear plane, as

A,

=7

tW sin d,

SECTION 20.7 Shear Strain y and Shear Front Angle ip

545

~ecalling that t was the uncut c h ~ p thickness and w was the wldth of the workplece.The shear stress (flow stre~s) is, therefore,

For a given polycrystalline metal, this shear stress has been shown to be not sensitive to variations in cutting parameters, tool material, or the cutting environment. Figure 20-22 gives some typical values for the flow stress for a variety of metals, plotted against hardness. Specific horsepower is related to and correlates well with shear stress for a given metal, which will be derived later. Unit power is sensitive to material properties (e.g., hardness), rake angle, depth of cut, and Ceed, whereas r, is sensitive to material properties only.

Uslng Merchant's c h ~ p formation bubble model, which emulates the videographlc lm20-23, can be developed. From ages, a new "stack-of-cards" model, as shown in F~gure thix model,?train is expressed as
y = cos c./[sin(d,

+ p) cos(6 + p

a)]

where
9 = the angle of the onset of the shear plane

the shear front angle

Using the available machining data, $is observed to decrease, reach a mmlmum, and rise again for all rake angles for a given metal of some hardness. The minimum energy princ~plehas been reported to have use 111 varrous fields such as physics, metalforming processes. and mach~nlng proces3es. Applied to metal cuttmg, the spec~fic shear energy (shear energylvolume) equals shear stress X shear straln

150

16-2S6
n

9/

/F C C matrlx metals

5
A .

125 3 0 '4 , Inconel 600 a,p

/ '

x Q

-m
E

100 -

d Nickel 200
0 Cold itntshed

B C C matrix

(steels) Steels

$
U)

75 -

9445
0 8640

I
0

I 100

I 200

300

400

BHN,brine11 hardness number

Onset of shear plane angle 6

Workpiece Mechant's bubble model of chip forrnat~on

The shaded shear triangle

on the rlght is used to develop the basic equation for shear straln, y

FIGURE 20-23 The Black-Huang "stack-of-cards" model for calculating shear straln in metal cutting 1 s based on Merchant's bubble model for chip formation, shown on the left

directions) t o reduce shear energy t o a minimum.

That is
dUJd$ = 0
The shear front angle is obtarned by

Substituting

+i n equation 20-34into equation 20-31, the shear Strain can be expres

y = 2 cos oi/(1

45" - 6

+ oi/2

+ sin a)

1 to 2 in,/in.

o f t h ~ process. s

In order to verlfy equation 20-34, metalcutting experiments in copper with a hardness gradient ranging from dead soft to full hard were performed. The equation was experimentally venfied to 99% confidence!
The materlal begms to shear at the lower boundary of the shear zone, defined by the angle 4.As the hardness of a material increases, 4 increases while $decreases, so @ 4 = 45" t ct 1 2 is mamtamed lor all levels of hardness The material in the shear zone shears at an mclination angle $to Lhe plane of the onset 4 of shear plane for aluminum and steel. Shear strain and shear front angle can be determ~ned by
y = 2 cos u/(l t sin a)
$$

+ 4 = 45" + ru/2

where 4 and $vary with hardness

Machining is a dynamic process of large strain and high strain rates. All the process variables are dependent variables. The process is intrinsically a closed-loop interactive process as shown in Figure 20-24. Starting at the top,inputs to the processes (speed. fced,depth of cut) determine the chip load on the tool. The chip load determines the cutting forces (magnitude and direction) (usually elastic), which alters the chip load on the tool.The altered chip load produces new forces.The cycle repeats, producing chatter and vibration. Remember that plastic deformation is always preceded by elastic deformation. The elastic deflection behaves like a big spring. The mechanism by which a process

INPUTS = V, f,, DOC tool geometry material props., etc

nC i
Response Forces D~splacement and chip size
variation
mag, and

direction

design

-24 Mach~n~ng
)recess

a closed-loop that creates laczment response

Statlcs and dynamics

machining, it has long been obse dissipates energy is called chatter or vibrat~on-In in pracfice that rotational speed may greatly mfluence process stability and ch Experienced operators commonly listen to machining noise and interactively the speed when optimizing a specific apphcation. In addition, experience demons that the performance of a particular tool may vary significantly based on the mac tool employed and other characteristics such as the workpiece, fixture holderband like. Today more than ever, the manufacturing industry is more competitive and sponsive, charactemzed by both high-volume and smalI-batch production, see economies of scale. High produotivity is achieved by increased machme and tool pabilities along w~th the elunination ofall non-value-added acnvities.Few cornpame afford lengfhy tnal-and-error approaches to machimng-process optimization or tlohal processes to treat the effect of chatter. In metalcut&g,chatter is a self-excited vibration that 1 s caused by the closed-l force-displacement response of the machimng process. The process-induced varia in the cutting force may be caused by changes in the cuttlng velocity, chip cross se (area), tool-ch~p interface friction, built-up edge, workpiece variation, or, most c monly, protess modulation resulting in regeneratton of vibration. When more en input into the dynamic machining system than can be dissipated by mechanical damping, and friction, equilibrium (the state ofminimum potential energy) IS sou the machining system through the generation o f chatter vibration. The proper classification of the type of vibration is the first step in identifying solving thecause of unwanted Vibration (see Figure 20-25).

Free vzhmtion is the response to any initial condition or sudden change. The
tude of the vibration decreases with time and occurs at the natural frequency o system. Interrupted machining is an example that often appears as lines or sha followng a surfacediscontinuity. Forced vibration is the response to a periodic (repeating with time) input.Theresp and input occur at the same frequency.The amphtude of the vibration remains co for set input conditions and is linearly related to speed Unbalance, mis tooth impacts, and resonance of rotation systems are the most common ex
.Free Vibration The response to an lnitrai condit~on or sudden change. The amplitude of the vibration decreases with tlme and occurs at the natural frequency of the system often produced by interrupted mach~ning Often appears as l~nesor shadow foilowingasurfacediscontinuity.

Free v ~ b r a t ~ o n

Forced Vibration The response to a periodiC (repeating with time) Input. The response and Input ocwr at the same of the vibration frequency The ampl~tude remalns constant for a set Input condition and 1 s nonilnearly related to speed. Unbalance, misal~gnm&nt. tooth impacts, and resonance of rotaung systems are the most common examples.

d
Present

FICU

20-25 There are three vibration In rnach~ning

.Self-Excited Vibratton The perlodic response of thk system to a constant Input. The vibration may grow i n amplitude (unstable) and oscurs near the natural frequency of the system regardless of the Input Chatter due to the regeneration af surface WavInees is the most common metal cuttlng exampls

Self-excited vcbratcon is the period~c response of the system to a constant input. The vibrat~on may grow in amphtude (become unstable) and occurs near the natural frequcncy of the system regardless of the input. Chatter due to the regeneratlon of wavlness in the machmed surface 15 the most common metal cutting example.
How do we know chatter exists? Listen and look! Chatter 1s characterized by the following: 1. There is a sudden onset of vibration (a screech or buzz or whine) that rapidly increases in amplitude until a maximum threshold (saturation) is reached. 2. The frequency of chatter remains very close to a natural Gequency (critical frequency) of the machining system and changes little with variation of process parameters.The largest force-displacement response occurs at resonance and therefore the greatest energy dissipation. 3. Chatter often results in unacceptable surface finish, exhibited by a helical or angular pattern (pearled or fish scaled) superimposed over normal feed marks. 4. Visible surface undulations are found in the feed direction and corresponding wavy or serrated chips with variable thickness. l e the surface f~msh Figure 20-26 shows some typ~calexamples of chatter v ~ s ~ b in marks. There are several~mportant factors that mfluence the stabihty of a mach~mng process: Cutting stiffness of the workpiece material (related to the machinability), K, Cutting-process parameters (speed, feed, DOC, total width of chip) Cutter geometry (rake and clearance angles, edge prep, insert size and shape) Dynamic characteristics of the machining process (tooling, machine tool, fixture, and workpiece)

K,, cutting stiffness, is closely aligned with flow stress but simpler to calculate in that 4is not used. Like flow stress, cutting stiffness can be viewed as a material property of the workpiece, dependent on hardness.

CHIP FORMATION AND REGENERATIVE CHATTER In machining, the chip is formed due to the shearing of the workpiece material over the chip area ( A = thickness X width = t X w), which results in a cutting force F, The magnitude of the resulting cutting force is predominantly determined by the material cutting stiffness K, and the chip area such that F, = K, X f X w.The direction of the cuttingforce F, is influenced mainly by the geometries of the rake and clearance angles as well as the edge prep. Machining operations require an overlap of cutting paths that generate the machined surface (see Figure 20-27). In single-point operations, the overlap of cutting paths does not occur until one complete revolution. In milling or drilling, overlap occurs in a fraction of a revolution, depending on the number of cutting edges on the tool. The cutting force causes a relative displacement X between the tool and workpiece.which affects the uncut chip thickness t and,in turn, the cutting force.This coupled relationship between displacement in the Y-direction (modulation direction) and the resulting cutting force forms a closed-loop response system. The modulation direction is normal to the surface defining the chip thickness. A phase shift e between subsequent overlapping surfaces results in a variable chip thickness and modulation of the displacement, causing chatter vibration. The phase shift between overlapping cutting paths is responsible for producing chatter. However, there is a preferred speed that corresponds to a phase-locked condition ( E = 0) that results in a constant chip thickness t. A constant chip thickness results in a steady cutting force and the elimination of the feedback mechanism responsible for regenerative chatter. This what the operators are trying to achieve when they vary cutting speeds (see Figure 20-28).

Chatter in boring

Chatter in turning, plunge cutting, and chamfering during single point operations

Chatter in face milling visible in the surface finish

Chattar due to increased radi immersion leading to rnoree

due to force release and tool spring back

Chatter in end milling

FIGURE 20-26 Some examples of chatter that are visible o n the surfaces of the workpiece.

radial engagernen (full immersions

HOW DO THE IMPORTANT FACTORS INFLUENCE CHAlTER?

Cuning stqfnfness KK, and work hardening and is often described in a relative sense of the machin materials. Materials such as steel and titanium require much greater shear fo aluminum or cast i displacement in t Speed. The procegs parameters are the easiest factors to change chatter and its tude. The rotational speed of the tool affects the phase shift between overlapp

SEmoN 20.8 Mechanics of Machining (Dynamics)

Y -modulation d~rect~on F, - outting force

551

a -rake angle
y

-clearance angle Xo-prevlous surfacevib. X -present surface v~b. E - phasesh~ft

When the get out of other, a variable produced, ange in F,on the e.
i

Workp~ece W = Width of cut Into t h e piane

Number of cycles between cuttlng path overlap

Constant Uncut chip thickness

Variable uncut

chiu thickness

Regenerative gand milling fable uncut chip

faces and the regeneration of vibrati0n.A handheld speed analyzer2 that produces dynamically preferred speed recommendations is commercially available. When applied to processes exhibiting a relative rotational motion between the cutting tool and workpiece, it recommends a speed to eliminate chatter.
'Best speed by Des~gn Manufacturing Inc ,Tampa,Flonda.

The most successful applications are in

9

M~llimg, boring, and turning Multipoint tools Machining aluminum and cast iron Hlgh-speed machmmg Thin-chip, high-speed d ~ machinrng e

At slow speeds (relative to thevibrat~on frequency) process stability 1s mainly due to creased frictional losses occurring between the tool clearance angle (defined by y) the present surface vibration X.This interference and fiict~on dlss~pates energy m form of heat and is calledprocess dampmg. As machining speeds are increased wavelength of the surface vibration also Increases, whch reduces the slope of the s and eliminates process-mdnced damping. Additionally, chatter becomes more sig cant as speeds increase, because ex~sting forces approach the aatural fieqnencies o machining system. The analyzer measures and identifies the vibrational frequenci the chatter noise and delermines which speeds w~ll most closely result in E = 0 A z phase shift between overlapping surfaces eliminates the variation of the chip thic and elim~nates the modulation, resulting in chatter Feed.Thefeed per tooth defines the average uncut chip thickness t and influences magnitude of the cutting force. The feed does not greatly influence the stabilit the machiningprocess (is., whether chatter occurs) but does control the severity o v~bration. Because no cutting force ex~sts if the vibration in the Y-d~rection resu the loss of contact between the tool and workpiece, the maxlmum amplitude of ter vibration is limited by the feed. D O C T h e depth of cut is the prlmary cause and control of chatter.The DOC de the chip width and acts as the feedback gain in the closed-loop machining pro The stab~lity l i m ~ (or t borderline between stable machining and chatter) ma expenmentally determined by incrementally increasing the DOC until the ons chatter. It can also be analytically predicted based on a thorough understanding o machining system dynamics and mater~al cutting stiffness. Total wldth of chzp The total wrdfh of c/up is equal to the DOC times the numb cutting edges engaged in the cnt.The total width of cut directly Influences the st ty of the process At a fixed DOC that corresponds to the stability limit, increasln number of engaged cutting edges will result in chatter.The number of engaged in the cut may be increased by adding inserts (us~ng a fine-pitch cutter) or lncre the radial immersion of a milling cutter. Conversely, reducing the number of edg the cut will have a stabihung effect on the process. The cuttrng tool geometry Influences the magnitude and d~rection of the cu force, espec~ally the amount of the force component In the modulation direction greater projection of force in the Y-direction results in Increased displacemen vibration normal to the surface, leading to potential chatter.

As the back rake angle a increases (becomes more positive), the length of the of shear plane decreases, whchreduces the magnitude of the cutting force, F, A positive rake also directs the cutting force to be more tangential and reduces the component in the Y-direction. In general, a more positive cutting geometry incr process stability, especially at higher speeds. An insufficient feed compared to th radius results m less efficient machming, greater tool deflection, and poorer m ingstability. A reduced clearance angle y,which Increases the fnct~onal contact between the to workpiece, may produce process damping.The s t a b h m g effect is due to ene f heat, which potentially decreases tool life and may th sipation in the form o d~stort the workpiece or increase the heat-affected zone in the workpiece.The wear of a new cutting edge may have a stabihung effect on chatter.

MILLING

BORING

( = 1
Radial

0" Lead

60" Lead

15" Lead

Less stable

More stable

More stable

Less stable

nCURE 20-29 M~lllng and bor~ng operations can be made more stable by correct select~on of Insert geometry

The size (nose radius),shape (diamond, triangular, square, round), and lead angle of the insert all influence the chip area shape and the corresponding Y-direction (Figure 20-29). In milling, the feed direction is transverse to the tool axis (i.e., radial), and the DOC is defined by the axial immersion of the tool. In milling, as the lead angle of the cut increases or the shape of the insert becomes rounded (same effect as a large nose radius compared to a small DOC), the Y-orientation is directed away from the more flexible radial tool direction and toward the stiffer axial direction. The orientation of the modulation direction Y toward a dynamically more rigid direction results in decreased vibrational response and therefore greater process stability (less tendency for chatter). In boring, the feed direction is axial, and the DOC is determined by the radial direction. Therefore, in boring, a reduced lead angle or a less round (and smaller nose radius compared to the DOC) insert maintains a more axial (stiffer tool direction) orientation of Y, leading to greater stability. Because the stability of the machining process is a direct result of the dynamicforcedisplacement characteristics between the tool and workpiece, all components of the machining system (tool, spindle, workpiece, fixture, machine tool) may, to varying degrees, influence chatter. Maximizing the dynamics (the product of the static stiffness and damping) of the machining system leads to increased process stability. For example, the static stiffness of an overhung (cantilever) tool with circular cross section varies nonlinearly with the diameter D and the unsupported length L. Machining stability is increased by having the tool with the largest possible diameter with the minimum overhand.Thefrequency of chatter occurs near the most flexible vibrational mode of the machining system.

STABILITY LOBE DIAGRAM A stability lobe diagram (Figure 20-30) relates the total width of cut that can be machined to the rotational speed of the tool with a specified number of cutting edges.Tf the total width of cut is maintained below a minimum level (although this may be of limited practical value for some machining systems),then the process stability exhibits speed independence

lode region "Condnronal"stabil~ty Stable p r o c e damping region Bdrderline of stability

FfGURE 20-30 Dynam~c analysis of the cutting procezr produces a stability lobe d~agrarn, which defines speeds that produce stable and unstable cutting conditions.

speed frpm) Rotat~onai

o r "unconditional" stability. At slow speeds, increased stability may be achieved the process damping region.The "conditional" stability lobe resons allow increased width of cut @OC x number of edges engaged in the cut) at dynamcally preferred at whi& the phase shift a between overlapping or consecutive cuttingpaths appro zero.The stability lobe number Nindicates complete cycles of vibrabon that emst be overlapping surfaces.As can be seen by the diagram, the higher speeds correspo lower lobe numbers and provide the greatest potential increase in the. total width and material removal rate (due togreater lobe height and width). If the total width exce~ds the borderline of stability, even if the process is operating at a preferred chatter occurs.The greater the total w~dth of cut above the stabilrtyhmit,the more n and violent wdl be the chatter vibration. When a chatter condition occurs, such as at point a on the stability lobe dia the rotational speed is adjusted to the first recommended speed (N = I), which in stable machining at pomt b on the diagram.The DOC may be incrementally i until chat161 again occurs as the Stability border a crossed at point e. Usmg the er again, chattering under the new operating conditions wll result in a modified recommendation corresponding to point d. If desired, the DOC may agaln be ' mentally increased (conservative skps promote safety) to pomt e. In general, attempt to maintain the DOC (and total width of cut) right up to the borderline bility because workpiece variation affecting K,, speed errors, or small changes in namic characteristics of the m a c h i i g system may result in crossing the stabili into severe chatter. The amplitude of chatter vibratmn may be more safely S i i temporary reduction of the feed per tooth until a preferred speed and stable de cut have been established.
HEAT AND TEMPERATURE IN METALCUTTING In metalcutting, the power pat into the process (F,VJ is largely converted to he vating the temperatures of the chip, the workpiece, and the tool.These three of the process, along with the environment (which includes t h e cuttingfluid), heat sinks. Figure 20-31 shows the distribution of the heat to these three sinks as tion of cutting speed. As speed increases, a greater percentage of the heat ends chip to the paint where the chips can b k ? cherry red or even bum at high cutbn There are three main sources of heat. Listed in order of their heat-gene capacity, they are shown in I3gure 20-32.
1 . The shear front itseIf, where plasttc deformation results in the major heat Most of thii heat stays in the chip. 2. The tool-chip interface contact region, where additional plastic deformatio place in the chip and considerable heat is generated due to sliding friction. 3. The flank of the t o 4 where the freshly produced workpiece surface rubs the

1
Percent of total going to work Percent of total going to tool

Total heat generated

/
aSecondary sheal'zone

Percent of total going to chip

Low
Cutting

High

speed ( V ) A

Secondary shear zone

L
FIGURE 20-32 There are three main sources of heat in metal cutting. (1) Primary shear zone. (2) Secondary shear zone tool-chip U-C) interface. (3) Tool flank. The peak temperature occurs at the center of the interface, in the shaded region.

a Primary shear zone

There have been numerous experimental techniques developed to measure cuttlng temperatures and some excellent theoretical analyses of this "movmg" multiple-heat-source problem. Space does not permit us to explore this problem in depth. Figure 20-33 shows the effect of cutting speed on the tool-chip mterface temperature. The rate of wear of the tool at the interface can be shown to be directly related to temperature (see Figure 20-33b). Because cuttlng forces are concentrated in small areas near the cutting edge, these forces produce large pressures.The tool material must be hard (to resist wear) and tough (to resist cracklng and chipping).Tools used in

m
'O

>
C .-

,a 1000
m m a

E 1 2 0 0 ~

i104:

temperature

800 600 400 200

lu5-

lW6-

FIGURE 20-33 The typical relationship of temperature at -the tool-chip interface to cutting speed shows a rapid increase. ,Correspondingly,the tool wears ,at the interface rapidly with increased temperature, often created by increased speed.

lo-'
400 800 1200 1600 2000

i 400 800 1200 1600 2000

Temperature "F
(a)

Temperature "F

interrupted cutting, such as milling, must be able to resist impact loading as well. materials must sustain their hardness at elevated temperatures.The challenge to ufacturers of cutting tools has always been to find m conditions. Cutting tool materials that d o not lose hardn associated with high speeds are said to have "hot hardn usually requires a trade-off in toughness, as hardness and toughness are posing properties. In Chapter 21 cutting-tool materials will be addressed in more de

In this chapter, the basics of the machining processes hav through 29 provide additional information on the various Modern machine tools can often perform several basi Chapter 26, machining centers are now widely used.These chapters on basic proc must be carefully studied.Tlie relationship between the basic processes a tools that can be used to perform these processes has been presented. In Chapter anatomy lor automation will be presented, which will show that most of the ma described in this chapter will be of the A(2) or A(3) level of automat centers, which are numerical control machines, are A(4).Machining centers hav matic tool-change capability and are usually capable of tapping (hole threading), and other minor machining p you will need to become familiar with new terminolo processes will need inputs concerning rpm (given that you selected the cutting sp feeds, and depths of cut. Note also from these tables that the same process performed on two or more different machine tools.There are many ways to pr flat surfaces, internal and external cylindrical surfaces, and special geometries in Generally, the quantity to be made is the driving factor in the selection of processes, will be explained later. This chapter also introduced orthogonal machining. a laboratory machining pr used to experimentally determine values for specific horsepower, cutting stiffness, flow stress for machining. As noted previously, the properties of the work material are important in chip mation. High-strength materials produce larger cutting forces than materials of lo strength, causing greater tool and work deflection: increased friction, heat generati and operating temperatures; the structure and composition also influence metal ting. Hard or abrasive constituents, such as carbides in steel, accelerate tool wear.

Work hardness prior to machining is an important factor because it controls the onset of shear. Onset is delayed by increased hardness, so 4 mcreases, as does .r, Highly ductile materials not only permlt extensive plastic deformat~on of the chlp during cuttlng, which increases heat generation and temperature, but also result in longer,"cont~nuous"chips that remain in contact longer wlth the tool face, thus causing more frictional heat. Chlps of t h type are severely deformed and have a characterlstlc curl. On the other hand, materials that are already heav~ly work hardened or br~ttle,such as gray cast iron, lack the ductility necessary for appreciable plastic deformation Consequently, the compressed material ahead of the tool falls m brittle fracture. sometimes along the shear front, producing small Iragments. Such chips are termed d~scontinuou~ or segmented. A variation of the continuous chip, often encountered in mach~nlng ductlle materials, is associated wlth a buzlt-up edge (BUE) formatlon on the cutting tool. The local high temperature and extreme pressure in the cutting zone cause the work material to adhere or pressure weld to the cuttlng edge of the tool formlng the budt-up edge, rather like a dead metal zone in the extrusion procesa Although this materlal protects the cutting edge from wear, it modifies the geometry of the tool. BIJEs are not stable and w111 break off periodically, adhering to the c h ~ p or passlng under the tool and remalnmg on r minlm~zed by rethe machned surface. Bu~lt-up edge formation can he elunmated o ducing the depth of cut, altering the cuttlng speed, using pos~tive rake tools, applying a coolant, or changing cutt~ng-tool materials.

II Key Words
depth of cut drlllmg dynam~cs feed flow stress fnct~on force gnnding (abras~ve machmmg) machlne tool metalcutt~ng m~lhng oblique mach~n~ng onset of shear orthogonal machlnlng rcgcneratlve chatter sawlng self-excited vibration shaping shear angles shear force shear strain shear veloc~ty spec~fic honepower speed stab~l~ty lobe d~agram turnlng v~bratinn workhold~ng dev~ce workplece

II Review Questions
uttmgprooess resisted theoretical sblutron What var~ablesmust be considered m understanding a machmmg process? Whlch of the seven basic ch~p formatlon processes are smgle point, and which are multlple po~nt? See Figure 20-2. How 1s feed related to speed tn the machining operations
11. For orthogonal macbnmg, the cuttlng edge rahus k assumed

E o be small compared to tkeuneut chip thickness.Why? 12 Row do the magnitude ofthe strain and atrarn rate value5 o f metalruttrng compare to those of tensile testmg? 13. Why is htanium such a d~fficult metal to machme? (Note its hrgh value of HPJ, 14. Explain why youget Segmented or discont~nuous chips when you mach~ne cast iron Before you select speed and feed for a machlntng operatton, 15. Why is metal cutting shear stress such an important deterwhat have you had to dec~de? (Hmt See Figure 20-4.) mination? WImg has two feeds What are they,and whlch one IS an tnput 16. Which of tho three cutt~ngforces in obhque cutt~ng consumes parameter to the machine tool? most of the power? What is the fundamrntal mechanism of c h p formahon? 17. How is the energy rn a mdchmmgproceustyp~cally consumed? What are the impllcattons of Figure 20.17, glven that this 18. Wheredoatheenergyconsumedin metalcuttingultimately go? ndeograph was made at a very low cuttlng speed? 19. State two ways of estlmattng the primary cutting force F, What is the ditference hetween obhque mach~ning and or- 20 How 1s cutting speed related to tool wear? thogonal machlmng'? 21. What 1s the relatiunship between ha~dness and temperature m Note that the units far the approximate equation tor MRR metal cutting tool materials? for turmng are not CoTTect.When 1s the approximate equation 22 Why d6es the cutting force F, Increase w~th Increased feed or not very good (y~elds a large error m MRR values)? DOC?

558

C H A ~ 20 R Fundamentals ofMachin~ngiOrthogonal Machln~ng 25 You had a machlningoperation (bonng) running perfee you changed work matenals All of a sudden, you are lots of chatter Why? 26 Explarn F~gure 20-31. Why 1s the percentage of total he erated dnrlng machlmng changing as speed 1ncreases7

23 Why doesn't the cuttlng force F, Increase w ~ t h Increased speed V 7 24. How do the selection of the mach~nlng parameters (speed, teed, DOC) ~nfluence chatter?

III Problems
1. For a turning operation, you have selected an HSS tool and turning a hot rolled free machining steel, Bhn = 300. Your depth of cut will be 0.150 in.7he diameter of the workpiece is 1.00 inches. a. Whai speed and feed would you select for this job? b. Using a speed 01105 sfpm and afeed of 0.015,calculate the spindle rpm for this operation. c. Calculate the metal removal rate. d. Calculate the cutting time for the operation with a length of cut of 4 in. and 0.10-in. allowance. 2. For a slab milling operation using a 5-in.-diameter, 11-tooth cutter (see Figure 20-6), the feed per tooth ts 0.005 in.1tooth with acuttingspeed of 1M)sfpm [HSS steel). Calcnlate therpmof the cutter and the feed Fate (f,) of the table, then calculate the metal removal rate, MRR, where the width of the block being machined is 2 in. and the depth of cut is 0.25 in. Calculate the time to machine (T,,) a 6-in.-long block of metal with this setup. Suppose you switched to a coated-carbide tool, so you increase the cutting speed to 400 sfpm. Now recalculate the machining time (T,) with all the other parameters the same. 3. The power required to machine metal is related to the cutting force (F,) and the cutting speed. For Problem 1, estimate cutting force F, for this turning operation. (Hint:You have to estimate a value of HPs for this material.) 4. In order to drill a hole in the material described in Problem 1 using an HSS drill, you have to select a cutting speed and a feed rate. Using a speed of 105 sfpm for the HSS drill, calcnlate the rpm for a a-in.-diameter drill and the MRR if the feed rate is 0.008 inches per revolution. 5. Explain how the constant 33,000 in equation 20-8 is obtained. 6. Explain how the constant 396,000 in equation 20-10 is obtained. 7. Suppose you have thc following data obtained from a metalcutting experiment (orthogonal machining). Compute the shear angle, ihe shear stress, the specific energy, the shear strain, and the coefficient of friction at the tool-chip interface.

How do your HP, and ?,values compare with the value! in Chapter 20? 8. For the data in Problem 7, determine the specific shea gy and the specific friction energy. 9. Derive equations lor Fand N using the circular force di ( H i m Make a copy of the diagram. Extend a line Gon X intersecting force F perpendicularly. Extend a lin point Y intersecting the previous line perpendicu1arly.F angle ru made by these constructions.) 10. Derive equations for F, and F, using the circular force di (Hint Consuuct a line through X parallel to vector F,. vector F, to intersect this line. Construct a line from X I dicular to F , . Construct a line through point Y perpendi~ the line through X.) 11. For the data in Problem 7,calculate the shear strain an pare it to llr,. Comment on the comparison l i r , = l,/t assuming that W = W,. 12. A manufacturing engineer needs an estimate of the force F, to estimate the loss of accuracy of a machining: due to deflection.The material being machined is Inco with a BHN value of 100.The cutting speed was 250 fi feed was 0.020 in./rev, and the depth of cut was 0.250 chip from the process measured 0.080 in. thick. E! the cutting force F, assuming that F, = FJ2. 13. Using Figure 20-4 for input data, determine the maxim minimum MRR values for rough machining (turning) a 1 bon steel with a BHN value of 200.Repeat for finishma assuming a DOC value equal to 10% of the rough'lng Dl 14. Estimate the horsepower needed to removc metal at 550 with a feed of 0.005in.lrev at a DOC value of 0.675 in.?he force F, was measured at 10,000 lb. Comment on these. 15. For a turning process, the horsepower required was 24 metal removal rate was 550 in.'/min. Estimate the ! horsepower and compare to published values for. 10: at 200 BHN.

Data Run Number

1

F,
330 308 41 0 420 510 540

&
295 280 330 340 350 395

Feed rpr Xljl000 4.89 4 89 7.35 7 35 9 81 9 81

Chlp Ratlo

r,
0.331 0.381 0 426 0 426 0 458 0.453

m Usy

fi

! ! 4 5 6

Case Study

559

Ofthoganal Plate Madining Experimenf at Auburn University

eremy has just been to h ~ 1SNY3800 s metalcutting lab where he learned about orthogonal machin~ng. Thls lab pmvlded hlm with a hands-on-experience tn materlal cutting analysis. The experlment vaned hts cutting speed and uncut ch~p thickness (UCT), two of the most Important cunlng parameters in orthogonal metalcutt~ng. An orthogonal plate machining setup was used in the experlment. A horizontal mlli~ng machine equipped wlth QSD and a dynamometer was used to perform the experiment The material for the experlment was cartridge brass. E~ghteen runs were performeduslng 2 levels of

speed, 3 levels of UCT, and 3 levels of posltlve rake angles. A toolmakers microscope were used to measure the thickness of the chip (b), and the shear angle The cutting and thrust forces (Fc and Ft) were measured wlth a dynamometer The data was reaorded m Table CS-20a Your task Is to make a table wlth head~ngs for r , ph~ (+), F, N, beta, MU, F , , Tau (r5),HPs, and spectfa energy (U). Complete the table and then d~scuss [us~ng plots) the effect (on the forces and other calculated values) of the changlng level in the input parameters

s*d

Z$ ste

21.1 INTRODUCTION 21.2 CUTTING-TOOL MATERIALS Tool Steels High-Speed Steels TiN-Coated High-Speed Steels Cast Cobalt Alloys Carbide or Sintered Carbides Coated-Carbide Tools Ceramics Cennels

BI 21.1 INTROUUCTION
Success in metal cutting depends on the selection of the propcr cutting tool (material geometry) lor a given work material. A wide range of cutting-tool materials area

sinlcred polycrystallinc diamond, and singlc-crystal natural diamond. Figure 21-1 some of these common tool materials ranked by the cutting speeds used to mac

2000 --

k m , .5 1500 -,5 7s 5$
m c x ,z ..LO
'"
' ,

Tool material iortomorrow+

't:

1000--

FIGURE 21-1 Improvements

in cutting tool materials have led to significant increases in cutting speeds (and productivity) over the years.
1800 1850 1900 1950 2000 Year tool material introduced into practice

560

Secrio~ 21.1 Introduction

561

and tool geometry selected d~rectly Influence the productiv~ty of the machmhg operation. F~gure 21-2 outhnes the input variables that influence the tool mater~al selection decision The elements which mfluence the decis~on are: Work material characteristics, hardness, chemical and metallurgical state, Part characteristics (geometry, accuracy, finish, and surface-integrity requirements) Machine tool characteristics, including the workholders (adequate rigidity with high horsepower, and wide speed and feed ranges) Support systems (operator's ability, sensors, controls, method of lubrication, and chip removal) Tool material technology is advancing rapidly, enabling many difficult-to-machine materials to be machined at higher removal rates andlor cutting speeds with greater performance reliability. Higher speed andlor removal rates usually improve productivity. Predictable tool performance is essential when machine tools are computer controlled with minimal operator interaction. Long tool life is desirable especially when machines become automatic or are placed in cellular manufacturing systems. The cutting tool is subjected to severe operating conditions.Too1 temperatures of 1000C and high local stresses require that the tool have these characteristics.

1. High hardness (Figure 21-3)

2. H ~ g h hardness temperature, hot hurdness (refer to Figure 21-3)

3. 4. 5. 6. 7.

Resistance to abrasion, wear due to severe sl~dlng fr~ct~on Chippmg of the cutting edges H~gh toughness (impact strength) (refer to F~gure 21-4) Strength to resist bulk deformation Good chemical stability (inertness or negligible affinity with the work material)

Manufacturing process-cont~nuous vs tnterrupted CONSTRAINTS

Condition &capabilitiesof available machine tools (rigidity)

Geometry, finish, accuracy, & surface-integrity requirements Workholding devices (rigidity) Required processing time-production schedule

:
INPUTS OUTPUTS

Work mater~al-hardness,

state cornposltlon. & metallurg~cal Type of cutvs flnlshlng rough~ng Part geometry & me-r~g~dtty

Selected tools-spec~f~c tool mater~al. grade, shape, and tool geometry Speed irpm) Feed Depth of cut
Cutting fluids

CUTTINGDECISION

TOOL SELECTION

1 I

*
t

Lot size--small batch vs mass product~on Machinabillty data-tool l~feispec~fic HP Qualitylcapability needed
of decis~on maker Past exper~ence

Availabilitv (ava~lable mater~als, the~r compos~t~on, properties and appl~cat~ons, ava~lable slres, shapes, and geometry, ava~labil~ty, dellvery schedule, cost and performance data)

of the cutt~ng-tool rnater~al and geometry followed by the selection of cuttlng FIGURE 21-2 The select~on depends upon many variables. cond~t~ons for a given appl~cat~on

Diamond-naturallsynthetic

Sintered cubic boron nitrideCBN CVD-titanium carbide Sintered silicon carbide CVD-titanium nitride carbon nitride CVD-aluminum oxide CVD-chromium carbide Diffused layer-CVD-iron boride Sintered TIC-WC hard metals Nitrided case of an alloy steel Electrodeposited hard chrome plated Nitrided case of an unalloyed steel Hardened steel Hardened and tempered steel Iron
0 1

Knoop hardness scale-1000

Kp/mmz (at

200
I
I

600
I

1000 Temperature ("Fi

I
4

1400
I

1
5

100

6 Temperature ("Ci ibt

1 8

900

FIGURE 21-3 (a) Hardness of cutting materials and (b) decreasing hardness with incream temperature, called hot hardness. Some materials display a more rapid drop in hardness above some temperatures. (From Metal Cutting Principles, 2nd ed. Courtesy of lngersoll Cutting Tool Company)

8. Adequate thermal properties 9. H i g h elastic modulus (stiffness) 10. Correct geometry and surface finish
Figure 21-5 compares these propertiesfor various cutting-tool materials;$ p i n g characteristics exist i n m a n y cases.Exceptions t o the rule are very common.] 0bta.u classes of t o o l materials, a wide range of compositions and properties ar~e~

SECTION 21.1

Introduction

563

Toughness
Toughness (as considered for tool~ng materials) is the relative resistance of a materlal t o breakage, ch~pplng, or cracklng under Impact or stress Toughness may be thought of as the opposlte of brittleness Toughness testlng is not the same as standardized hardness testlng. It may be dlff~cult t o correlate the results of different test methods. Common toughness tests include Charpy impact tests and bend fracture tests

CHARPY V-NOTCH

CHARPY C-NOTCH

FRACTURE

Methods of toughness testlng

Wear Resistance
Alloy elements (Cr, V , W. Mol form hard carbrde particles In tool steel mlcrostructures Amount & type present influence wear reslstance.

Hardness of carbides:
H a r d e n e d steel Chrornlum carb~des M o l y , tungsten carbides Vanadium carbides Carbide I Conventlona( tool steel microstructure
60165 HRC 66/68 HRC 72i77 HRC 8 2 / 8 4 HRC

Microstructure of P/M tool steel versus conventional tool steels shows the fine carbide distribution, uniformly distributed.

P/M tool steels microstructure

FIGURE 21-4 The most important properties of t o o l steels are:

2.
3.

Hardness-resistance t o deforming a n d flattening Toughness-resistance t o breakage and chipping Wear reslstance-res~stancet o abrasion a n d erosion.

C a r h o n and Law-/MedmmAlloy Steels Toughness Hot hardness Impact strength Wear resistance Chtpplng residance Cuttingspeed Depth of cut

H~gh-Speed Steels

Stntered Cemented Carb~deq

Coated HSS

Coated Carbides

Polycrystalhne Ceramlcs

CBN

Dtamond

b
b

Decreasing Increas~ng Decreasing

Increasing

4
b 4
Decreasing

b b 4 b 4

b
Light to med~urn L~ght to heavy Rough Wrought cast, HIPsintering Machining and grinding

Light to heavy
Good Cold pressing and sintering, PM Grinding

L g h t to heavy Good PVDb after

Llght to heavy Good CVDE

Llght to heavy Very goad

L~ght to heavy Very good

Fmsh ohta~nable Rough Method of manufacture Fabrication

lhermdl shock res~stance

very light for s~ngle crystal diamond Excellent

Wrought

forming
Machinins and Grinding before grinding,coating coating

Cold pressing High-pressore High-pressureand sinlering or high-temperature high-temperature HIPsintering sintering sintering Grinding Grinding and polishing Incrrastng Increasing Grinding and polishing

TODLmaterial cost

, ,

Machining and grinding

"Overlappilig characlcristica exist in many cases. Exceptions to the rule are very common. In many claws of iml material%a wide range ofcotnpositions and properties are obtainable. hPhysicalvapor dcposilion. 'Chemical vapor deposition.

SEcrlo~ 21.2 cutting-~ool Materials

565

Flgure 21-3 compares various tool mdterlals on the basis of hardness, the most crit~cal characterist~c, and hot hardness (hardness decreases slowly with temperature) F~gure 21-4 compares hot hardness with toughness, or the abil~ty to take Impacts durlng interrupted cutting. Naturally,it would be wonderful if these materials were also easy to fabricate, read~ly available, and mexpenslve,slnce cutting tools are routinely replaced, but thls 1s not usually the case Obviously, many of the requirements confllct and therefore tool selection wllI always require trade-offs.

lnnearly all machining operations, cuttlng speed and feed are limited by the capablhty of the tool material. Speeds and feeds must be kept low enough to provide for an acceptable tool life. If not, the time lost changing tools may outweigh the prodnct~vity gains from increased cuttlng speed Coated hgh-speed steel (HSS) and uncoated and coated carbides are currently the most extensively used tool materials. Coated tools cost only about 15 to 20% more than uncoated tools, so a modest improvement in performance can justify the added cost About 15 to 20% of all tool steels are coated, mostly by thephy.~curl vapor depositzon (PVD) processesDiarnond and CBN are used for applications m wh~ch, despite hlgher cost, their use IS justlfied. Cast cobalt alloys are belng phased out because of the high raw-material cost and the increasing availabihty of alternate tool materials. New ceramic materials called cermers (ceramic mater~al in a metal binder) are having a significant impact on future manufacturing productivity. Tool requirements for other processes that use noncontacting tools, as in electrodischarge machining (EDM) and electrochemical machining (ECM), or no tools at are discussed in Chapter 20. Grindlng abrasives will be disall (as In laser mach~ning), cussed in Chapter 29.

TOOL STEELS Carbon steels and low-/medium-alloy steels, called tool steels, were once the most common cutting-tool materials Plain-carbon steels of 0.90 to 130% cdrbon when hardened and tempered have good hardness and strength and ddequate toughness and can be given a keen cuttlng edge. However, tool steels lose hardness at temperatures above 400F because of tempering and havelargely been replaced by other materials for metal cutting The most important properties for tool steels are hardness,hot hardness,and toughness. Low-/medium-alloysteels have alloying elements such as MOand Cr, which improve hardenability, and W and Mo, which Improve wear resistance. These tool materials also lose then hardness rapidly when heated to about them temper~ng temperature of 300' to 650QF, and they have limited abrasion resistance. Consequently, low-/medium-alloy steels are used in relatively inexpensive cuttlng tools (e.g., drills, taps, dles, reamers, broaches, and chasers) for certain low-speed cutting applications when the heat generated 1s not high enough to reduce thelr hardness s~gn~frcantly. mgh-speed steels, cemented carbides, and coated tools are also used extensively to make these kinds of cuttrng tools. Although more expenslve, they have longer tool life and Improved performance. These steels greatly benefit from PIM manufacturing due to uniformly dlstr~buted carbides.

HIGH-SPEED STEELS First Introduced III 1900by F W.Taylor and Whte,h~gh-alloy steel IS superlor to tool steel m that it letains its cuttlng ab~lity at temperatures up to llOOF,exhibitinggood"red hardness." Compared with tool steel, it can operate at about double or tr~ple cuttlng speeds to about 100sfpm with cqual life, resulting ~nits name: high-speedsteel, often abbreviated HSS. Today's hlgh-speed steels contain significanl amounts of W, Mo, Co,V, and Cr besides Fe and C. W, hlo, Cr, and Co in the fernte as a sol~d solution provlde strengthening of the matrix beyond the tempering temperature, thus increasing the hot hardness. Vanad~nm(V), along w t h W, Mo, and Cr, ~mproves hardness (R, 65-70) and wear resistance. Extenswe sohd solutioning of the matrlx also ensures good hardenability of these steels.

IHAPTER 21 Cutting 'voois'fbr Machining

Although many formulations are used, a typical composition is that of ihe 1 type (tungsten 18%,chromium 4%. vanadium I%), calledT1. Comparable perform can also be obtained by the substitution of approximately 8% molybdenum for the sten, referred to as a tungsten equivalent (W,). High-speed steel is still widely use drills and many types of general-purpose milling cutters and in single-point tools in general machining. For high-production machining, it has been replaced almost pletely by carbides, coated carbides,and coated HSS. HSS main strengths are: Great loughnesssuperior transverse rupture strength Easily fabricated Best for sever applications where complex tool geometry is needed (gear cutters,ta drills, reamers, dies) High-speed steel tools are fabricated by three methods: cast, wrought, and tered (using the powder metallurgy technique). Improper processing of cast and wro products can result in carbide segregation, formation of large carbide particles an nificant variation of carbide size,and nonuniform distribution of carbides in them The material will be difficult to grind to shape and will cause wide fluctuations ofpr erties, inconsistent tool performance, distortion, and cracking. To overcome some of these problems, a powder metallurgy technique has b developed that uses the hot-isostatic pressing (HIP) process on atomized, preallo tool steel mixtures. Because the various constituents of the P/M alloys are "lo in place by the compacting procedure, the end product is a more homogeneous Figure 21 -4. PIM high-speed steel cutting tools exhibit better grindability,greater to ness better wear resistance, and higher red (or hot) hardness; they also perform consistently. They are about double the cost of regular HSS.

TIN-COATED HIGH-SPEED STEELS Coated high-speed steel provides significant improvements in cutting speeds, wit creases of 10 to 20% being typical. First introduced in 1980 for gear cutters (hobs in 1981 for drills,TiN-coatedHSS tools have demonstsated their ability to more than for the extra cost of the coating process. In addition to hobs,gear-shaper cutters, and drills,HSS tooling coated by T i includes reamers, taps, chasers, spade-drill blades, broaches, bandsaw and circular blades, insert tooling, form tools, end mills, and an assortment of other milling cutt Physical vapor deposition has proved to be the most viable process for coa HSS, primarily because it is a relatively low-temperature process that does not exc the tempering point of HSS.Therefore,no subsequent heat treatment of the cutting is required. Films 0.0001 to 0.0002 in. in thickness adhere well and withstand minor tic, plastic, and thermal loads. Thicker coatings tend to fracture under the typical t momechanical stresses of machining. There are many variations to the PVD process, as outlined in Table 21-1. process usually depends on gas pressure and is performed in a vacuum chamber. processes are carried out with the workpieces heated to temperatures in the range o lo 900F Substrate heating enhances coating adhesion and film structure. Because surface pretreatment is critical in PVD processing, tools to be coat subject to a vigorous cleaning process. Precleaning methods typically involve de ing, ultrasonic cleaning, and Freon drying. Deburring, honing, and more active cle methods are also used. The advantages of TiN-coated HSS tooling include reduced tool wear. Less wear results in less stock removal during tool regrinding, thus allowing individual t to be reground more times. For example, a TiN hob can cut 300 gears per sharpe the uncoated tool would cut only 75 parts per sharpening. Therefore the cost per is reduced from 20 cents to 2 cents. Naturally, reduced tool wear means longer to Higher hardness,with typical values for the tbin coatings, "equivalent" to %8 as compared to R, 65-70 for hardened HSS,means reduced abrasion wear.Relati ertness (i.e.,TiN does not react significantly with most workpiece materials) result

S E C ~ O21.2 N

Cutting-Tool Matenalri

567

Steelsurface IS coated wrth n~tride layer by use of cyanide salt nt P000 to 1600Eor ammonia, ga% or N ~ t o n s 'tic )lamp

To 72 R,, Cascdepth U.DW1 to 0 LOO m

70-72 K:

H~gh productmnrates wtth bulk handling. Htgh suriace hardness Drffusesintnthe steel surfaces Slnnlates strmn hardening Low filctlon weffioient, anhgafhng, Gorrostonre%stance High hardne~s
I 3 r g c qunntttas prr b.l!ih.

ton
iJ

ten.

The part is the cathode in a chtomic acld solution, anode 1s lead Hard chrome plahngls thc u e a r most common process far m reatstance Depos~t~an of wattng matenal by chermcal rcactioas m the gasems phase. Reachve g a k s replace a protective atmosphere m a vacuum nhamber,At temperatures of 1800" to IZWRa thlnd~ffuswn zone ircreatad bstween the base metal and the coating

0 M)in to

0 100 m.

Sl,dn r:iictU,n times rudu;. stjtKtr_~t.: itrc\~? Cxcc!lcnl :adllsion. recommended for forming tools. MultlpIe coatings can he applied (liN,Ttc,&O,) Line-of stght not a problem

tion

mheam)

s generated tn a vacuum Plasma 1 chamberby Ion bomhardmcnt lo drslmlge parncles from a target made of the eoaringmatertal Mctal is evaporated and IS condensed or auraotcd to substrate surfaces A plasma 1s generatad m vacuum by evaporation from a molten pool that is hcated by an electionbeam guu

compounds, and refractones AppbcabIe for all toolmg. & I coat reasimable ouantiues ner bat& cycle katenals i r e metals, wmpaunds, alloys, and refraamles. Substrate metallurgy eprese~ed.Vq goad adhesion Elne particle depadhon Appl~oahlefvr all toohng Process at 90U'Voreserves deposloon of gram sire and pmwth. Bmension$surface fmn&,and sharp edges aEe preservsd Can coat all highsgeedsteels wrthout distortcon

Can anly he applied to steel. Process has cmbriltlingeffect becauseof greater hardness Post-heat freement needed for somealloys. Moderatcproducuon,p1eceb must hektured Part must n d o e n he vwy clean Coatmg d nQt dtffuse mto surface,which can affect unpact propties Hlgh temperatures can affect suhsnatemetallurgy, requlringpost-heat treatment,wh~chcancausen caw dtmensroual dtstoruon (except when mating smtered carb~des)Necessary to reduce ~ffeots of hydrogen chlortdP.on material pmpertlessuch as Impact strength Usually nat drffussd Tolerances of +0.001 requued for HSS tools Not a hsgh-production method Requlrcs care m cleamng Usually not diffused.

PartS~qulrr fixtwing and orienration ih Ime-of-czght process Ultra-cleanliness required

Titanium is evaporated in a vacuum and reacted with nitrogen Gas. Resulting titanium nilride plasma is ionized and electrically attracted to the substrate surface. A high-energy process with multiple plasma guns.

To85 R.:

by products formed m reachon Usually pnly mrnor d~ffus~on

greater tooItire'ihrough a reduction in adhesion.TiN coatings have a low coefficient of friction.This can produce an increase in the shear angle, which in turn reduces the cutting forces,spindle power, and heat generated by the deformation processes. PVD coatings generally fail in high-stress applications such as cold extrusion, piercing, roughing, and high-speed machining.

CAST COBALT ALLOYS Cast cobalt alloys, popularly known as stellite tools, are cobalt-rich, chromiumtungsten-carbon cast alloys having properties and applications in the intermediate range between high-speed steel and cemented carbides. Although comparable in roomtemperature hardness to high-speed steel tools, cast cobalt alloy tools retain their hardness to a much higher temperature. Consequently, they can be used at higher cutting

CHAFER21

Cutting Tools for Machining speeds (25% higher) than HSS tools. Cast cobalt alloys are hard as cast and can softened or heat treated. Cast cobalt alloys contain a primary phase of Co-rich solid sol by Cr and W and dispersion hardened by complex hard, refractoly ca Other elements added include V, B, Ni, and Ta. The casting provide elongated grains normal to the snrface.The structure is not,howe Tools of cast cobalt alloys are generally cast to shape and fin ing.They are avaflahle only in simple shapes, such as single because of limitations in the casting process and the expense inv ing (grinding).The high cost of fabrication is primarily due to the high ha material in the as-cast condition. Materials machinable with this tool mat plain-carbon steels, alloy steels, nonferrous alloys, and cast iron. Cast cobalt alloys are currently being phased out for cutti cause of increasing costs, shortages of strategic raw materials (Co, W, and Cr), development of other, superior tool materials at lower cost.

CARBIDE OR SINTERED CARBIDES Carbide cutting-tool inserts are traditionally divided into two primary groups:
1 .Straight tungsten grades,which are used for machining cast irons, austeniticst steel, and nonferrous and nonmetallic materials. 2. Grades containing major amounts of titanium, tantalum, and or columbium c which are used for machiningfemtie workpieces.There are also the titanium c grades, which are used for finishihg and semifinishingferrous alloys.
The classification of carbide insert grades employs a United States and I S 0 P and M classification system in Eur sifications are based on application, rather than composition or properties Each tool vendor can provide proprietary grades and recommended applications. Carbides, which are nonferrous alloys, are also called sintered (or cement bides because they are manufactured by powder metallurgy techniques.Thz PiFYI is outlined in Figure 21-6. See Chapter 16 for details on powder metallurgy pr 1 , as they afforded a fourThese materials became popular during World War 1 fold increase in cutting speeds The early versions had tungsten carbide as th constituent, with a cobalt binder in amounts of 3 to 13%.Mo

in vacuum furn

press. The compacted powder is sintered in a high-temperature vacuum furnace.The so

21.2 SECTION

Cutting-Tool Materials

569

are either straight WC or multicarbides of W-Ti or W-Ti-Ta, depending on the work material to be machined. Cobalt is the binder. These tool materials are much harder, are chemically more stable, have better hot hardness, have high stiffness, have lower friction, and operate at higher cutting speeds than HSS.They are more brittle and more expensive and use strategic metals (W,Ta, Co) more extensively. Cemented carbide tool materials based on TiC have been developed primarily for auto industry applications using predominantly Ni and Mo as a binder. These are used for higher-speed (>I000 ftlmin) finish machining of steels and some malleable cast irons. Cemented carbide tools are available in insert form in many different shapes: squares, triangles, diamonds, and rounds. They can be either brazed or mechanically clamped onto the tool shank. Mechanical clamping (Figure.21-7) is more popular because when one edge or corner becomes dull, the insert is rotated or turned over to expose a new cutting edge. Mechanical inserts can be purchased in the as-pressed state or the insert can be ground to closer tolerances. Naturally.precision-ground inserts cost more. Any part tolerance less than 10.003 normally cannot be manufactured without radial adjustment of the cutting tool, even with ground inserts. If no radial adjustment is performed, precision-ground inserts should be used only when the part tolerance is between 20.006 and e0.003. Pressed inserts have an application advantage because the cutting edge is unground and thus does not leave grinding marks on the part after machining. Ground inserts can break under heavy cutting loads because the grinding marks on the insert produce stress concentrations that result in brittle fracture. Diamond grinding is used to finish carbide tools. Abusive grinding can lead to thermal cracks and premature (early) failure of the tool. Brazed tools have the carbide insert brazed to the steel tool shank.These tools will have a more accurate geometry than the mechanical insert tools, but they are more expensive. Since cemented carbide tools are relatively brittle, a 90' corner angle at the cutting edge is desired. To strengthen the edge and prevent edge chipping, it is rounded off by honing, or an appropriate chamfer or a negative land (aTland) on the rake face is provided. The preparation of the cutting edge can affect tool life. The sharper the edge (smaller edge radius), the more likely the edge is to chip or break. Increasing the edge radius will increase the cutting forces, so a trade-off is required.Typica1 edge radius values are 0.001 to 0.003 in.

1-

Insert
[holder

size

Insert type a n d size

Adapter

j _
\
\ \

Groove width Groove depth

Boring head lsert cutting tools. lave a chip groove the chips to curl ?ak into small, I lengths.

A chlp groove (see F~gure 21-7) \nth a posltlve rake angle at the tool tip ma be used to reduce cuttmg forces w~thout reduc~ng the overall strength of the inse nificantly.The groove also breaks up the chips for easler disposal by causmg th curl tightly. For very low-speed cutting operations, the chips tend to weld to the tool fa cause subsequent microchipping of the cutting edge. Cutting speeds are general1 range of 150 to 600 ftimm. Higher speeds (>I000 ftlmin) are recommended for less-difficult-to-machine materials (such as alum~num alloys) and much lower (100 ftimin) for more difficult-to-machine materials (such as titanium alloys). rupted cuttmg applications, it is important to prevent edge chlpping by choosing propriate cutter geometry and cutter position with respect to the workpie interrupted cutkng, finer grain size and higher cobalt content improve toughn straight WC-Co grades. After use, carbide inserts (called disposable or throwaway inserts) are genera cycled in order to reclaim the Ta, WC, and Co. T h ~ recycling s not only conserves gicmaterials but also reduces c0sts.A new trend is to regrind these tools for futu where the actual size of the insert is not of critical concern.

COATED-CARBIDE TOOLS Beginning in 1969 with Tic-coated WC, coated tools became the norm in the working industry because coating can consistently Improve tool hte 2200 or 30 more. In cutting tools, material requirements at the surface of the tool need to sion resistant, hard, and chemically inert to prevent the tool and the work mate lnteractlng chemically w ~ t h each other dur~ng cutting. A thin, chermcally stable, fractory coating of TiC,TlN, or ALO, accomplishes this objective.The bulk of 1s a tough, shock-resistant carbide that can withstand hgh-temperature plasti mation and resist breakage.The result is a composite tool as shown m Figure 2 To be effectwe, the coatmgs should be hard, refractory, chem~cally stabl chemically Inert to shield the constituents of the tool and the workp~ece from in ing chem~cally under cutting condit~ons. The coatings must be fine gra~ned, binders and porosity. Naturally, the coatings must be metallurg~cally bonded substrate. Interface coatmgs are graded to match the properties of the coatinga substrate. The coatings must be thick enough to prolong tool life but thin eno prevent br~ttleness. Coatings should have a low coefficient of friction so the ch~ps do not adh the rake face. Coating materials include slngle coatings of TiC, TIN, A124,H HfC. Multiple coatings are used, with each layer Imparting its own characterisbc tool. Successful coating combinations include T~NITICIT~CNITIN and TiCIAl$I Chemical vapor deposition is used to obtain coared carbides.The coatings are f by chemrcalreactions that take place only on ornear the substrate. Like elect CVD is a process in which the deposlt is built up atom by atom. It is therefor of producing deposits of maximum denslty and of closely reproducing fine detail substrate surface Control of critical variables such as temperature, gas concentration, an pattern is required to assure adhesion of the coatlng to the substrate. The coa substrate adhesion must be better for cutting-tool inserts than for most other c applications to survive the cutting pressure and temperature con&tlons without off. Gram size and shape are controlled by varylng temperature and/or pressur The purpose of multiple coatings is to tailor the coating thickness for pro tool life.Multiple coatings allow a stronger metallurgical bond between the coati the substrate and provide a variety of protection processes for machining differ materials, thus offermg a more generaI-purpose tool mater~al grade. A very thin of 'IiN coating (ym) can effectively reduce crater formation on the tool face two orders of magnitude relative to uncoated tools. Coated inserts of carbldes are findmg w ~ d e acceptance in many meta apphcations. Coated tools have two or three tnnes the wear resistance of the best u tools w ~ t h the same breakage resistanee.This results in a 50 to 100% increase ins the same tool life, Because most coated~nserts cover a broader application ran

S E C I O N

21.2

Cutting-Tool Materials

571

T~tanium carb~de remalns as the bas~c materlal coverlng the substrate for strength and wear resistance The second layer is alum~ntum ox~de, whlch has proven chemical stabil~ty at high temperatures and resists abras~ve wear. The third laver IS a thin coating of titanium nitride to $ive the insert a lower coefficient of friction and to reduce edge build up. Titanium nitride coating-low

Titanium carbide (TiCN) as first layer-strength

R Relat~ve thickness of coatlngs

Titanium mtride coating Alum~num oxide-2nd layer Tltan~um carb~de-1st layer substrate

and wear resistance A1203 Aluminum oxide 2nd layer--chem~cal stabil~tyat h~gh temperature-r~sists abrasive wear

FIGURE 21-8 Trtple-coated carb~de tools provide reststance to wear and plastic deformabon in machlnlng of steeL abrasive wear in cast iron, and built-up edge formation.

grades are needed, and therefore inventory costs are lower. Aluminum oxide coatings have demonstrated excellent crater wear resistance b y p r o v i d ~ n g a chemical diffusion reaction barner at the tool-chip interface, p e r m ~ t t i n g a 90% increase m cutting speeds in m a c h i n ~ n some g steels Coated-carbide tools have progressed t o t h e place where in the U n l t e d States about 80 t o 90% o f the carbide tools used in melalworkmg are coated.

CERAMICS

ordinary ceramics, sintenng occurs w~thout a vitreous phase.

tungsten carbide do= at lower speeds, As shown m Table 21-2, ceramics are us

have greater transverse rupturestrength have been developed.The whiskers are Ir~m silicon carbide.

CERMETS
Cermets are a new dass of tool materials best suited for finishing. Cemets are cer TIC,nickel, cobalt, and tantalum nitrides.TiN and other carbides a s used for binders mets have superior wear resistance, longer tool life, and can operate a t higher cn speeds with superior wear resistance. Cermets have higher hot hardness andoxid resistance than cemented carbidesThe better flnish imparted by a cermet is due low level of chemical reaction w ~ t h iron [less cratering and built-up edge (BUE)].

mtlling by using high nitrogen content in the titanium carbonitride phase finer grain size) and adding WC and l s C t o improve shock resistance PVDand they perform well vith a coolant. Flgure 23-9 shows a comparison of speed feed coverage of typical cermets clearly cover a wide range of important metalcutting applications

DIAMONDS
Diamond is the hardest material known. Industrial diamonds are now ava~lable form of polycrystalline compacts, which are finding Industrial application in th

H~gh-speed steel

86 R ,
91-94

600

RA

l0a-125

600-650 400-650

2000

PCD or CBN

Coated cermets Coated carbldes Tungsten carbide Hlghspeed steel

Tool Material Group PCD (polycrystal diamond) CBN (cubic boron nitride]

General Applications High-speed machining of aluminum alloys, nonferrous metals, and nonmetals. Hard workpieces and high-speed machining o n cast irons.

Versus Cermet Cermets can machine same materials, but at lower speeds and significantly less cost per corner. Cermets cannot machine the harder workpieces that CBN can. Cermets cannot machine cast iron at the speeds CBN can. The cost per corner of cermets is significantly less. Cermets are more versatile and less expensive than cold press ceramics but cannot run at the higher speeds. Cerrnets cannot machine the harder workpieces or run at the same speeds on steels and irons but are more versatile and less expensive. Cermets cannot machine cast iron at the high speeds of silicon nitride ceramics, b u t in moderate-speed applications cermets may be more cost effective.

Ceramics (cold press)

High-speed turning and grooving of steels and cast iron.

Ceramics (hot press1

Turning and grooving of hard workpieces; high-speed finish machining of steels and irons. Rough and semirough machining of cast irons in turning and milling applications at high speeds and under unfavorable conditions. General-purpose machining of steels, stainless steels, cast iron, etc. Tough material for lower-speed applications on various materials.

Ceramics (silicon nitride)

Coated carbide

Cermets can run at higher cutting speeds and provide bettertool life at less cost for semiroughing to finishing applications. Cermets can run at higher speeds, provide better surface finishes and longer tool life for semiroughing to finishing applications.

Carbides

FIGURE 21-9

Comparison of cerrnets wlth vartous cuttlng-tool mater~als.

chining of aluminum, bronze, and plastics, greatly reducing the cutting forces as compared to carbides. Diamond machining is done at high speeds, with Line feeds for finishing,and produces excellent finishes. Recently,single-crystaldiamonds, with a cutting-edge radius of 100 or less, have been used for precision machining of large mirrors. However, single-crystal diamonds have been used for years to machine brass watch faces,thus eliminatingpolishing.They have also been used to slice biological materials into thin films for

CHAPTER 21

Cutting Tools for Machining viewing in transmission electron microscopes. (This process, known as ultramicrotom one of the few industrial versions of orthogonal machining in common practice.) The salient features of diamond tools include high hardness; good thermal ductivity; the ability to form a sharp edge of cleavage (single-crystal, natural diam very low friction; nonadherence to most materials; the ability to maintain a sharp for a long period of time, especially in machining soft materials such as copper aluminum; and good wear resistance. To be weighed against these advantages are some shortcomings, which incl a tendency to interact chemically with elements of Group IVB to Group VIII of periodic table. In addition, diamond wears rapidly when machining or grinding steel. It wears less rapidly with high-carbon alloy steels than with low-carbon stee has occasionally machined gray cast iron (which has high carbon content) with lon Diamond has a tendency to revert at high temperatures (700C) to graphite an oxidize in air. Diamond is very brittle and is difficult and costly to shape into tools, the process for doing the latter being a tightly held industry practice. The limited supply of, increasing demand for, and high cost of natural dia led to the ultra-high-pressure (50 Kbar), high-temperature (1500C) synthesis of di from graphite at the General Electric Company in the mid-1950s and the subseq development of polycrystalline sintered diamond tools in the late 1960s. Polycrystalline diamond (PCD) tools consist of a thin layer (0.5 to 1.5 mm) o grain size diamond particles sintered together and metallurgically bonded to a ceme carbide substrate. A high-temperaturelhigh-pressure process, using conditions clo those used for the initial synthesis of diamond, is needed. Fine diamond powder ( 30 pm) is first packed on a support base of cemented carbide in the press. At the propriate sintering conditions of pressure and temperature in the diamond gion, complete consolidation and extensive diamond-to-diamond bonding ta Sintered diamond tools are then finished to shape, size, and accuracy by laser cutti grinding. See Figure 21-10. The cemented carbide provides the necessary elast' port for the hard and brittle diamond layer above it. The main advantages of si polycrystalline tools over natural single-crystal tools are better quality, greater toug and improved wear resistance, resulting from the random orientation of the diam grains and the lack of large cleavage planes. Diamond tools offer dramatic performance improvements over carbides.Tool is often greatly improved, as is control over part size, finish, and surface integrity. Positive rake tooling is recommended for the vast majority of diamond too applications. If BUE is a problem, increasing cutting speed and using more positive1 angles may eliminate it. Tf edge breakage and chipping are problems, one can red feed rate. Coolants are not generally used in diamond machining unless, as in chining of plastics, it is necessary to reduce airborne dust particles. Diamond tools be reground. There is much commercial interest in being able to coat HSS and carbides dii with diamond, but getting the diamond coating to adhere reliably has been di Diamond-coated inserts would deliver roughly the same performance as PCD t when cutting nonferrous materials but could be given more complex geometries chip breakers while reducing the cost per cutting edge.

POLYCRYSTALLINE CUBIC BORON NITRIDES Polycrystalline cubic boron nitride (PCBN) is a man-made tool material widely us the automotive industry for machining hardened steels and superalloys. It is made compact form for tools by a process quite similar to that used for sintered poly talline diamonds. It retains its hardness at elevated temperatures (Knoop 4700 at 2 4000 at 1000C) and has low chemical reactivity at the toolkhip interface.This mat can be used to machine hard aerospace materials like Inconel 718 and Ren6 95 as as chilled cast iron. Although not as hard as diamond, PCBN is less reactive with such materi hardened steels, hard-chill cast iron, and nickel- and cobalt-based superalloys. P can be used efficiently and economically to machine these difficult-to-machine m

Raw material from sintaring and compacting

Laser dicing dlsk into

Carbide insens

Standard tungsten carbide insen \

21-10

Polycrystalline

d tools are carbides with

Compax blank (0020 in. [O 51 mml thick

jiarnond inserts. They are .estricted to simple geometries.

diamond layer with carbide substratel

R. ~a re I.. !.~ -.

--

als at higher speeds (fivefold) and w~th a h~gher removal rate (fivefold) than cemented carbide, and with superior accuracy, finish, and surface mtegnty. PCBN tools are available in basically the same sizes and shapes as smtered diamond and are made by the same process. The cost of an insert is somewhat higher than either cemented carbide or ceramc tools, but the tool life may be five to seven times that of a ceramic tool.Therefore, to see the economy of using PCBN tools, it is necessary to consider all the factors. Here is an industrial example of analysis of toollng economics, where a comparison 1s being made between two tool materials (insert tools). A manufacturer of diesel engines is producing an m-line six-cylmder engme block that is machined on a transfer h e . Each cyl~ndel hole must be bored to accept a sleeve liner. This operation has a depth of cut of 0 062 in. per side, lor a total of 0.125 in stock removal. The toleranceon this bore is +0.001 in. and the spindle is operating at 2000 sfpm. Ceramic inserts are used on this operabon, but with these mserts, wear was severe enough to require lndexrng after only 35 pieces. The ceramic insert was replaced with PCBN inserts made of a highcontent BCN. Both inserts had a 0.001- to 0.002-in. radlus hone for edge preparation

576

CHA~E 21 R Cutting Tools for Machining

Table 21-3 is a cost comparison between the ceramic and PCBN insert.ThePCBN sert is a layer of PCBN material. At first glance the PCBN tool appears to be extr expensive.Each insert costs $208.00 and provides only three usable edges, whereas t ramic insert costs $14.90 and provides sixusable edges.However, the ceramic tool mus indexed every 35 pieces.The PCBN tool is indexed every 500 pieces.The cost per bor

application, the ceramicinsert required 10.75hours of downtime for indexing,whereas PCBN tool required only 0.75 hour of downtime for indexing. Use of the PCBN cutting f significantly reduces the total cost per piece by eliminating 10 hours of downtime o machine. Later in this chapter the economics of machining will be addressed again. The two predominant wear modes of PCBN tools are notching at the depthcut line (DCL) and microchipping. In some cases, the tool will exhibit flank wear of

carbides and ceramics.

Cost pet edge Time per iildea (6 tools) Cost per index at $45 per hour Indexes per 1500 b1,odks. Tndexing cost (indexes ~ ' $ 1 1 1 5 ) Insert rast for 6spindles tabor and tool Gast Cost per bore Total number of tool changes Downtime for 1300 blocks

$2.48 0.25 hr $tl.25 43

gaiS3.13

$6933 0.25 hr $11.25

3 $33.75
$1248.00 $1281.00 8.142

$638.34
$1122.09

$.I25
43

carbon steels Alloy-steels,

alloy cast iron

Aluminum, brass Hlgh-silieon dluminum. Nickel-based Titanium Plastic composites

X
X
X

X
X

X
X

Since diamond and PCBN are extremely hard but brittle matenals, new demands are being placed on the machine tools and on machining practice in order to take full advantage of the potential of these tool matenals. These demands include: Use of more rigid machine took and machining practices mvolving gentle entry and exit of the cut in order to prevent microchipping Use of high-precision machine tools, because these tools are capable of producing high finish and accuracy Use of machrne tools with hlgher power, because these tools are capable of higher metal removal rates and faster spindle speeds

21.3 TOOL GEOMETRY

Figure 21-11 shows the cutting-tool geometry for a single-point tool (HSS) used in turning.The back rake angle affects theability of the tool to shear the work material and fonn the chip. It can be positive or negative. Positive rake angles reduce the cutting force&resulting in smaller deflections of the workpiece, tool holder, and machme. In machining hard work matertals, the back rake angle must be small, even neganve f ~ carbide r and diamond tools, Generally speaking, the higher the hardness of the workpiece, the smaller the back rake angle. For high-speed steels, back rake angle 1s normally chosen in the positive range, depending on the type of tool (turning,planing, end milling, face milling, drilling, etc.) and the work material.

Locatlon of resultant

cation of back rake

Ic)

Clearance or end reltef angle

C H A ~ 21 R Cutting 1 -ools for Machln~ng

For carbide tools, inserts for different work materials and tool holders can be plied with several standard values of hack rake angle: 6 " to +6". The side rake and the back rake angle combine to form the effective rake angle.Th1s 1s also calle true rake angle or resultant rake angle of the tool. m True rake lnclinatlon of a cutting tool has a major effect ~ndetemming the a of chip compression and the shear angle. A small rake angle causes high compres tool forces, and friction, resulting in a thick, highly deformed, hot chip.Increased angle reduces the compression, the forces, and the friction, yielding a thinne deformed, and cooler chip. Unfortunately, it e difficult to take much advantage des~rable effects of larger positive rake angles, since they are offset by the re strength of the cutting tool, due to the reduced tool section, and by ~ t great1 s duced capacity to conduct heat away from the cutting edge. To provide greater strength a t the cutting cdge and better heat conductivity, or negative rake angles are commonly employed on carbide,ceramic, polydiam PCBN cutting tools Thesematerrals tend to be brittle, but their ahillty to hold perlor hardness at high temperahrres results in their selection for high-speed an tinuous machining operat~ons. While the negative rake angle increases tool fo keeps the tool in compression and provides added support to the cutting edge. particularly important in makmg Intermittent cuts and in absorbing the impact the initial engagement of the tool and work In general, the power consumpkon is reduced by approximately 1% for each alpha ( a ) . The end relief angle is called gamma (y) The wedge angle O determ strength of the tool and its capacity to conduct heat and depends on the values The relief angles ma~nly affect the tool life and the surface quality of the wo reduce the deflections of the tool and the workplece and to provlde good surface q larger relief values are required. For high-speed steel, relief angles in the range of 5" are normal, with smaller values be~ng for the harder work materials For carhldes,the angles are lower to gve added strength to the tool. The side and end cutting-edge angles define the nose angle and characte tool design.The nose radlus has a major influence on surface firush. Increasing radtus usually decreases tool wear and lmproves surface finlsh. Tool nomenclature varies with different cutting tools, manufacturers, and Many terms are still not standard hecause of all this vanety. The most commo terms wiU be used in later chapters to describe specific cutting tools. The introduction of coated tools has spurred the development of improve geometries Specifically, low-force groove (LFG) geometries have been develope reduce the total energy consumed and break up the chips into shorter segments. grooves effectively increase the rake angle, whtch increases the shear angle and the cutting force and power.Th1s means that h~gher cutting speeds or lower cuthn peratures (and better tool lives) are possible. As a chip breaker, the groove deflects the chip at a sharp angle and cau break into short pieces that are easier to remove and are not as likely to beco gled in the machine and possibly cause iqury to personnel.This is particularly tant on h~gh-$peed, mass-production machines The shapes of cutting tools used for various operatlona and n~atenals are promises, resulting ram experience and research so ac to provide good overall mance. For coated tools, edge strength is an important consideranon A thin coat the edge to retain high strength, but a thicker coat exh~bits better wear reslstan mally, tools for turning have a coating thickness of 6 to 12 pm Edge strength is for multilayer coated tools The radius of the edge should be 0.0005 to O 005 in.

PROCESSES

The two most effective coating processes for improving the life and perform tools are the chemlcal vapor deposition and physical vapor deposlhon of titani (TB) and titanlum carbide (TiC).The selection of the cuttmg vnaterlals for cutu depends on what property you are seeking. If you want

Oxidation and corrosion resistance; high-temperature stability Crater resistance, Hardness and edge retention, Abrasion resistance and flank wear, Low coefficient of friction and high lubricity, Fine grain size,

select select select select select select

A1,03, TiN,T i c A1203, TiN,TiC Tic, TiN, A1,03 Alz03, TiN, Tic TiN,A1,03, Tic TiN, TiC,A1,03

The CVD process, used to deposlt a protective coating onto carbide inserts, has been benefiting the metal removal industry for many years and is now belng apphed w~th equal success to steel.The PVD processes have quickly become the preferredT2N coatlng processes for hlgh-speed steel and carbide-tipped cutting tools.

CVD
Chemical vapor deposition is an atmosphere-controlled process carried out at ternperatures in the range of 950" to 1050UC (1740 to 1920UF).Figure 21-12 shows a schematic of the CVD process. Cleaned tools ready to be coated are staged on precoated graphite work trays (shelves) and loaded onto a central gas distribution column (tree). The tree loaded with parts to be coated is placed inside the retort of the CVD reactor. The tools are heated under an inert atmosphere until the coating temperature is reached.The coating cycle is initiated by the introduction of titanium tetrachloride (TiCL,), hydrogen, and methane (CH,) into the reactor.TiCL, is a vapor and is transported into the reactor via a hydrogen carrier gas; CH, is introduced directly. The chemical reaction for the formation of T i c is: To form titanium nitride, a n~trogen-hydrogen gas mixture is substituted for methane.The chemical reaction for TiN is:

PVD
The simplest form of PVD is evaporation, where the substrate is coated by condensation of a metal vapor. The vapor is formed from a source material called the charge, whichis heated to a temperature less than 1000C. PVD methods currently being used include reactive sputtering, reactive ion plating, low-voltage electron-beam evaporation, triode high-voltage electron-beam evaporation, cathodic evaporation, and arc evaporation. In each of the methods, the TiN coating is formed by reacting free titaniumions with nitrogen away from the surface of the tool and relying on a physical means to transport the coating onto the tool surface.

Electric coating furnace

RE 21-12 Chern~cal .depositton is used to apply ,~ I C TIN, , <:tc.) to carb~de lg tools.

1 4 To scrubber

Cutting Tools for Mach~n~ng All of these PVD processes share the following common features.
1 . The coating takes place inside a vacuum chamber under a hard vacuum with workpiece heated to 200' to 405'C (400 to 900F) 2. Before coating, all parts are given a final cleanmg inside the chamber to remove ides and improve coating adhes~on. 3. The coating temperature 1s relatively low (for cutting and forming tools), typ~c about 842" F (450" C) 4. The metal source is vaporized in an inert gas atmosphere (usually argon), and metal atoms react with gas to form the coating. Nitrogen is the reactwe gas for mtri and methane or acetylene (along with nitrogen) is used for carbides. 5. A11 four are ion-assisted deposition processes.The ion bombardment compresses atoms on the growing film, yielding a dense, well-adhered coatmg.

A typical cycle trme for the coatmg of functional tools, includmg heat-up and cool do is about six hours. Of the three, PVD arc evaporation, shown m Figure 21-13, is the most recent velopment The plasma sources are from several arc evaporators located on the sides top of the vacuum chamber Each evaporator generates plasma from multiple arcsp In this way a highly localized electrical arc discharge causes minute evaporation of material of the cathode, and a self-susta~ning arc is produced that generates a hi energy and concentrated plasma. The kinetlc energy of deposition is much greater than that found in any ot PVD method Duringcoating, this energy is of the order of 150 electron volts and m Therefore, the plasma is h~ghly reactive and the greater percentage of the vapor is ato and ionwed

arc evaporation process will preserve substrate metallurgy, surface finish, edge sharpn geometrical straightness, and d~mensions.

CVD A N D PVD-COMPLEMENTARY PROCESSES CVD and PVD are complementary coating processes.The differences between th processes and resultant coatings dictate whch coating process to use on d~fferent
be superior to a PVI-CVD-deposited coating. CVD coatlngs are normally depos' thicker than PVD coatings (6 to 9 pm for CVD, I to 3 pm for PVD). See Figure 21-

Evaporated niaterlal
FIGURE 21-13 Schematic of PVC arc evaporation process.

SECTION 21.4 ~mpar~son of PVD Process Character~stlcs Processing Temoerature 'C Throwlug Power L~ne of slght Coatlne Materials Chrefly metal, espec~ally A 1 (a few slmple alloys/ a few ssmple compounds) Usually N (B,C)

Tool Coat~ng Processes

581

Coatlug Appllcatrons and S ~ e c ~Features al Eledronic, optlcal, decorative, srmple maskmg.

cuum evaporation

RT-700,

usually < Z O O

nimplantatron

2 0 M 0 , best 4 5 0

L~ne o f srght

for N

Wear resistance for tools, dies, etc. Effect much deeper than original implantation depth. Precise area treatment, excellent process control. Electronic,optica1,decorative. Corrosion and wear resistance. Dry 1ubricants.Thicker engineering coatings Electronic, optical, wear resistance.Architectural (decorative).Generally thln coatings.Excellent process control. Thin,wear-resistant films on metal and carbide dies, tool&etc. Free-standing bodies or refractory metals and ppolytic C or BN.

R T 4 7 T, of coating Best at elevated temperatures

Moderate to good

Ion platrug Al, other metals (few alloys) ARE TiN and other compounds Metals, alloys,glasses, oxides TiN, and other compounds

R T 4 . 7 T," of metal cuatlngs Best >ZOO for nonmetals.

300-2000, usually 60&1200

Line-of-sight

Very good

Metals, espec~ally refractorvliN and other compounds; pyrolytic B N

&room temperuture;ARE = activated reactive evapuration; 7, =absolute melting temperature. (a) Compounds:oxidss,nitrides.carbides,silicides,and boridea ofAl, Cr,HfMo, Nb,Ni, Re. Si,Ta,Ti,V,W Zr. urie Advanced Materials and RocerscS December 2001

CURE 21-14

Comparison of PVD methods for depos~t~ng thin films on m~croelectron~c devices as well

cutting tools.

With CVD multiple coat~ngs, layers may be readily depos~ted but the tooling materials are restricted CVD coated tools must b e heat treated after coating.This lunits the apphcation to loosely toleranced tools. However, the CVD process, being a gaseous process, results in a tool that is coated uniformly all over; this includes blind slots and blind holes. Since PVD is mainly a line-of-slght process, all surfaces of the part to be coated may be masked. PVD also requires fixturing o f each part in order to effect the substrate has.

APPLICATIONS Applications for the two different processes are as follows.

CVD
Loosely toleranced tooling punches g P~ercing and blanking punches, trim dies, phillips p ~ n c h e s ~ u p s e t t i n AISI A, D, H, M, and air hardening Sohd carbide tooling

and tool steel parts

PVD
All HSS, solid carbide, and carbide-tipped cutting tools

Fine blanking punches, dies (0.001 in. tolerance or less)

steels and

Non-composition-dependent process; virtually all tooling materials, including mold bronze

W 21.5 TOOL FAILURE AND TOOL LIFE

In metalcutting, the fallure of the cutting tool can be classified into two broad categ according to the failure mechanisms that caused the tool to die (or fil):

wear) or both.
2. Chemical jafallures,which Include wear on the rake face of the tool (called crate

are rapid, usually unpredictable, and often catastrophic failures resulting from premature death of a tool.

ria is also widely varied. Figure 21-15 also shows a sketch of a "worn" tool, sho crater wear and flank weur, along \nth wear of the tool nose radlus and an outer-d eter groove (the DCL groove). Tools also fail by edge ch~ppmg and edge fracture As the tool wears,its geometry changeaThis geometry change wfflinfluence th ting forces, the power being consumed. the surface finish obtained, the dimensional

any or all o f these mechan~sms may be operatlve at a given tlme in a gwen process. Tool failure by plast~c deformation, brittle fracture, fatigue fracture, or edge c ping can be unpred~ctable. Moreover, it is difficult to predict whlch mechanism will inate and result m a toolfailure in a particular situation. What can be said is that t like people, die (orfail) from a great vanety of causes under andely varying condit

Groove at DCL or outer edge of cut Cross sectlon AA

Due to wear at the DCL or outer edge of the cut

A portlon of the workp~ece materlal adheres to the lnsertcun~

484

CHAPTER 21 Cutting Tools for Machining

FIGURE 21-18 Construction of the Taylor tool life curve uslng data from determ~nist~c tool wear plots like those of 21 -1 7. Curves lhke thls F~gure can be developed for both flank and crater wear.

wi (see the dashed-hne construction), Flgure 21-18 was developed. When V and T plotted on log-log scales, a linear relationship appears described by the equatlon

VT"

constant

lished his now-famous paper "On the Art of Cutting Metalsmin ASMETransactions, in tool Me (T) was related to cutting speed (V) and feed (fl.This equation had the constant fxVy which over the years took the more widely puhhshed form VT" = C*

F~gure 21-19 shows typical tool life curves for one tool material and three work terials. Notice that all three plots have about the same slope, n. Typical values for n 0.14 to 0.16 for HSS, 0.21 to 0.25 for uncoated carbides, 0.30 ForTiCinserts, 0.33 for po diamonds, 0.35 for TiN inserts, and 0.40 for ceramic-coated Inserts. equation, as each combination of tool and work material will have different const Note that for a tool life of 1minute, C = V,or the cutting speed that yields about 1min of tool life for this tool. theTaylor equation, wherein constants for other input parameters (typically feed,d of cut, and work material hardness) are experimentally determined, for example,
VTnFmdp= K'

*Carl Earth, who wils Taylor's mathematicalgenius, is generally thought to be the author of these forrnulah mks. along m t h early verslons of sl~de

VT"

=G

HQh-carbon steel

8.14,6 6,6 15.3164 8.14.6.6.6 l5,3$64

BrOnze ( 9 Cn, 1.5n) Cast Iron 160 Bhn Cast tron, Nlckel, 164 Bhn Cast ,ran, NI-Cr, 207 Bhn Stell,SAE ~ 1 1 1 c 3D SLeU,SAEBlllZ C D Stell.SAEB1120 C D Stell, SAE 81120 + Pb C.D. Stell, SAE B1035 C D Stell,SAE B1035 Ph C D Stel1,SAE 1045 C D Stell, SAE2340 185 Bhn Stell, SAE2345 198 Bhn Stell,SAE3140 190Bhn SteU,SAE 4350 363 Bhn Stell, SAE 4350 363 Bhn Stell, SAE 4350 363 Bhn Stell,SAE 4350 363 Bhn Stell, SAE4350 363 Bhn Stell,SAE 4140230 Bhn Stell, SAE 4140271 Bhn Stell. SAE 6140 240 Bhn Mnnel meld 215 Bhn

050
.1W)

HSS 18-4-1

8 14,6 6,6.15,~64

8 14.6 6,6.15,3164
8 14.6 6,6 13,3166 8.14,6 6,6 15,3164 814,6.6,615,3164 8.14,6.6.6 15,3164

,030 .(EO ,050 MO 050 -050 .050 ,050 .050 .I50

,025-5 ,0127 0255 .0255 ,0255

,0127 -@I27
0127 0127 ,5127 .0127 01'27 0125 .0255 0125 0127

Dry Dry Dry DL-y Dry ~ r y Dry Dry Dry

086 .I11 101

190 232 172

111
088

186
102
SO

am
105 -100 860

22s 270 290

D l y
Dry

,110
,110
,110 ,147 105 .I60

130 147
192 143 126 178 181 146 95

m y
Dty

100
.M0 -100 ,0125 ,0125 100 100 ,050 ,050 050 ,140 .150 ,100 ,106

Dry
Dry Dry Dry Dry Dry Dry Dry Dry Dry Dry Dry
Em

080
125 ,125 110 110 180

W .m

.a255
0255 .QI27 0259 0127 0127 ,0127 0127 .0252

78
46 190 159 197 170 127 185 189 7.15 240

HSS-18-el

8.14,6 6,6 15.3164

180
150 080

HSS-18-4-1

8.22,6 6.6 15,3164

@I27
01'27 031 031 031 . I 3 3 1 031

Stelhfe 2400

0.0,6 6,6 0,3132

Steel,SAE 3240 annealed

.I87
,125

StelliteNo 3 Carbide (TM)

0.0,6.6,6.0,3/32 6 12,s 5,10.45

Cast uon 290 Bhn Steal, SAE 1040annealed Steel, SAE 1060annealed Steel,SAE 1060 annealed Steel, SAE 1060 annealed Steel, SAE ID60 annealed Steel, SAE 1060 annealed Steel, SAE 1060 annealed

062 ,031 .Q62

SMO Dry Dry Dry Dry Dry

.om
.1M

.M4

190
190 1W 190 ,150

' 2 7 0
310 205 800 660 615 560 8 8 8 0 510 4W

,156
,167 167 .I67 167

.I64
,162

where n, n, and p are exponents and K' is a constant. Equations of this form are also deterministtc and determined empirically. The problem has been approached probabiitically in the follomng way. Sinm T depends on speed,feed, materials, and so on, one writes
K "- T=-=
vt/n

K V"

(21-6)

where K is now a random variable that represents the effects of all unmeasured factors and is an input variable. The sources of tool life variability include factors such as:

1. Vanation in work material hardness (from part to part and within a part) 2. VariabiiQ in cutting-tool materials, geometry, and preparation
3. Vibrations in machine tool, including rigidity of work and tool-holding devices

4. Changing surface characteristics of workpieces

FIGURE 21-19

Log-log to01 life plots for three steel work materials cut with HSS tool rnater~al.

20

I 1

10 Tool life, T, minutes

100

FIGURE 21-20 To01 life

viewed as a random variable has a log normal distribution with a large coefficient of variation.

Tool life or time to failure

The examnation of the data from a large number of tool hfe studies in which a nety of steels weremachined shows that regardless of the tool material or proces$t life dlstrlbutions are usually log normal and typically have a large standard devlati As shown in Flgure 21-20, tool life distributions have a large coefficient of variation! tool llfe IS not very predictable Other criterla that can be used to define tool death in addition to wear limits .When surface finjsh deteriorates unacceptably When workpiece dimension is out of tolerance When power consumption or cuttlug forces ~nctease to a limit Sparking or chip discoloration and disfigurement Cuttmg tune or component quantity In automated processes, tt is very beneficial to be able to monitor the tool wear on so that the tool can be replaced prior to fallure, whereln defectlve products m result. The feed force has been shown to be a good, indirect measure of tool That is, as the tool wears and dulls, the feed force increases more than the cutting increases.

SECTION 21.6

Flank Wear

587

Once criteria for failure have been established, tool life is that time elapsed between start and finish of the cut, in minutes. Other ways to express tool life, other than time, include: 1. Volume of metal removed between regr~nds or replacement of tool 2. Number of pieces mach~ned per tool
3. Number of holes dnlled with a given tool

Drilling tool falluse m discussed more in chapter 23 and 1s very complex because of the varied and complex geometry of the tools and as shown here ~nFigure 21-21, the tool material.

RECONDITIONING CUTTING TOOLS In the reconditioning of tools by sharpening and recoating, care must be taken in grinding the tool's surfaces,The following guidelines should be observed:
1. Resharpen to original tool geometry specificatioiis.Restoring the original tool geometry will help the tool achieve consistent results on subsequent uses. Computer numerical control (CNC) grinding machines for tool resharpening have made it easier to restore a tool's original geometry 2. Grind cutting edges and surfaces to a fine finish. Rough finishes left by poor and abusive regrinding hinder the performance of resharpened tools. For coated tools, tops of ridges left by rough grinding will break away in early tool use,Ieaving uncoated and unprotected surfaces that will cause premature tool failure. 3. Remove all burrs on resharpened cutting edges. If a tool with a burr is coated, premature failure can occur because the burr will break away in the first cut,leaving an uncoated surface exposed to wear. 4. Avoid reshmpeningpractices that overheat and burn or melt (called glazed over) the tool surfaces,as this will cause problems in coating adhesion. Polishing or wire brushi n g of tools causes similar problems.

99

1000 rpm 65.5 sfpm

.TiN coating by company D o Bright finished twist drills-uncoated O T i N coating by company B CTD black-oxide drill x CTD drills with TIN coat by company A ATiN coating by company E

E 21-21 Tool l~fe test

r varrous coated drills. coated HSS drills outperform ted dr~lls. L~fe based on the er o f holes drilled before

No, of holes drilled

Drill performancebased on the number of holes dr~lled wlth 114 in.-dlameter drills i n T-1 structural steel.

Tools for Machining The cost of eash recoating is about one-fifth the cost of purchasing a new tool, recoating, the tooling cost per workpiece can be cut by between 20 and 30%, de ing on the number of parts being machined.

F MACHINING
The cutting speed has such a great inUuence on the tool life compared to the fee depth of cut that it greatly infIuences the overall economics of the machining For a given combination of work material and taol material, a 50% increase in spe sults in a 90% decrease in toollife, while a SO% increase in feedreaults in a 60% desr in tool life. A 50% increase in depth of cut produces only a 15% decrease in too Therefore, in limited-horsepower situations, depth of cut and then feed shoul maximized while speed is held constant and horsepower consumed is maintain in limits. As cutting speed is increased, the machining time decreases but the to out faster and must be changed more often. In terms of costs, the situation is as in Figure 21-22, which shows the effect of cutting speed an the cost per piece. The total cost per operation is comprised of four individual costs: machiming c taol costs, tool-changing costs, and handling costs.The machining cost is observed to crease with increasing cuttingspeed because the cutting time decreases Cuttrng timeis p o r t i d t o the machining cost@. Both the toal costs and the tool-changmg costs incr with Increases in cutting specds. The hadling costs are independent of cuttings Addlng up each of the individual costs results in a total unit cost curve that 1s obs go through a minimum point,For a turning operation, the total wst per piece C

:IG
Rr

Utt

C"

c = cl + c2+ c3+ c d
= Machining cost

eq'

+ tooling cost + tool-changing cost + handling cost per piece

Expressing each of these cost terms as a function of cutting velocity unll pernnt summation of all the costs. Ci = Tm x Co where C , = operating cost ($Emin) T , = cutting time (minipiece) C, = ($)C, w h n r T = tool life (minitool)
C, = initial cost of tool($)
C.3 =

t, X

Go -

(3

where t, = time to change tool (min) T labor, overhead, and machine tool costs consumed while p being loaded or unloaded, tools are being advanced, machi has broken down, and so on. Since T, and T
= =

Tm -- number of tool changes per piece

c 4

L / N f r for turrmg rDL/12Vfr ( K / v ) ' I n,by rewriting equation 21-3,

and uslng "K"for the constant "C',the cost per unit, C, can be expressed m terms of V
c = - LwDC, 12Vf,

c,V'/" +-+-

K'/.

I~C~V'~" + Cd ~'1"

To find the minimum, take dc/dV

= 0 and

solve for V :

SECTION 21.7

Economics of Machining

589

cost

c*

Cost per unit process ve rsus dote that ttie 2 and relatti.d the same Taylor tool life

cost per piece C3

VM Cutting speed (sfpm) for minimum costlpiece
VI"C.~,S,",

So V, represents a cutting speed that will minimize the cost per unit, as depicted in Figure 21-22. However, a word of caution here is appropriate. Note that this derivation was totally dependent upon the Taylor tool life equation. Such data may not be available because they are expensive and time consuming to obtain. Even when the tool life data are available, this procedure assumes that the tool fails only by whichever wear mechanism (flank or crater) was described by this equation and by no other failure mechanism. Recall that tool life has a very large coefficient of variation and is probabilistic in nature.This derivation assumes that for a given V, there is one T-and this simply is not the case, as was shown in Figure 21-16. The model also assumes that the workpiece material is homogeneous, the tool geometry is preselected, the depth of cut and feed rate are known and remain unchanged during the entire process, sufficient horsepower is available for the cut at the economic cutting conditions, and the cost of operating time is the same whether the machine is cutting or not cutting. Another example of tooling economics is summarized inTable 21-6, where a comparison is made between four different tools, all used for turning hot-rolled 8620 steel with triangular inserts. Operating costs for the machine tool are $60 per hour.The lowforce groove insert has only three cutting edges available instead of six. It takes 3 minutes to change inserts and half a minute to unload a finished part and load in a new 6-in.-diameter bar stock.The length of cut is about 24 in.The student should study and analyze this table carefully so that each line is understood. Note that the cutting-tool cost per piece is three times higher for the low-force groove tool over the carbide but really of no consequence, since the major cost per piece comes from two sources: the machining cost per piece and the nonproductive cost per piece.

MACHINABILITY Machinability is a much maligned term that has many different meanings but generally refers to the ease with which a metal can be machined to an acceptable surface finish. The principal definitions of the term are entirely different, the first based on material properties, the second based on tool life, and the third based on cutting speed.
1. Machinability is defined by the ease or difficulty with which the metal can be machined. In this light, specific energy, specific horsepower, and shear stress are used as

590

C H ~21 R Cutting Tools for Machining

Cutting speed (surfac&fl/mtn) Feed (in Itev) Cunrng edges ava~lable per Insert Cost of an Insert ($/~nsert) Tool l ~ f (p~ecesisutnn~ e edge) Tool-change tune per p ~ e m (mm) Nonprixluct~vecostper piece ($lpcJ Machlnlng flme per piece (mmlpc) Machrnlng cast per plece ($/unit) Tool-change cost per piece ($/PC) cut tin^-too1 cost Ber piece j$/pc) Totd cost perplece (%PC) Production rate (pceceslhr) Impravoment ln productivity

0 015

0.075

0 075

48
48 0.08

2.7 27 0 08

1.50
15

0 08

measures, and, in general, the larger the shear stress or specific power values, more difficult the material is to maeh~ne, requiring greater folce~ and lower sp In this definition, the material is the key. some material, compared to a standard material cut with the same tool mat&

plies that steel X has a cutting speed of 70% of El112 for equal tool life. Not this defiition assumes that the too1 fads when machining X by whatevermech
30 min of ton1 life with flank wear of 0.33 mm.

automatic xrew bar machine.'' 4 Other definitions of machiibility are based on the ease of removal of the chips disposal), the quality of the surface finish of the part itseIf, the dimensional sta of the process, or the cost to remove a given volume of metal.

tool life curves.

FIGURE 21-23 Machinab~l~ty ratings defined by determin~stic

Tool life, T, minutee

60

Further defmitions are be~ng developed based on the probabilistic nature of the tool failure, in which mach~nability is defined by a tool reliabihty index. Using such Indexes, various tool replacement strategies can be examined and optimum cutting rates obtained.These approaches account for the tool life varlab~htyby devetoping coefficxents of variation for common combinat~ons of cuttlng tools and work materials The results to date are very prormsirlg One thing 1s clear, however, from t h ~ sort s of research: although many manufacturers of tools have worked at developing materials that have greater tool hfe at hlgher speeds,few have worked to develop tools that have less variability in tool hfe at all speeds The reduction In variab~lity e fundamental to achieving smaller coefficients of variat~on, which typically are of the order of 0.3 to 0.4. %s means that a tool with a loll-min average tool life has a standard deviation of 30 to 40 min, so there is a good probability that the tool will fall early. In automated equipment, where early, unpredicted tool fallures are extremely costly, reduction of the tool life vanabllity will pay great benefits in improved producti~ity and reduced costs.

'ING
From the day that Frederick W.Taylor demonstrated that a heavy stream of water flowing directly on the cutting process allowed the cutting speeds to be doubled or tnpled, cutting flutds have flourished m use and variety and have been employed m virtually every inachining process.The cutting fluid acts pr~marily as a coolant and secondly as a lubricant, reducing the friction effects at the tool-chip interface and the work flank regions. The cutting fluids also carry away the chips and provide fiiction (and force) reductions In regions where the bodies of the tools rub against the workpiece. Thus in processes such as dnlling, sawlng, tapping, and reaming, portions of the tool apart from the cutting edges come in contact with the work, and these (sliding friction) contacts greatly increase the power needed to perform the process, unless properly lubricated. The reduction in temperature greatly a ~ d in s retaining the hardness of the tool, thereby extending the tool life or permitt~ng increased cutting speed with equal tool life In addition, the removal of heat from the cutting zonc reduces thermal distortion of the work and permits better d~mensional control. Coolant effectiveness is closely related to the thermal capacity and conductivity of the flu~d used. Water is very effective in this respect but presents a rust hazard to both the work and tools and also is meffective as a lubricant. Oils offer less effective coolant capacity but do not cause rust and have some lubricant value. In practice, straight cutting oils or emulsion comb~nations of oil and water or wax and water are frequently used.Var~ous chemicals can also be added to serve as wetting agents or detergents, rust inhibitors, or polaruing agents to promote formation of a protectwe 011 film on the work The extent to which the flow of a cutting fluld washes the very hot chips away from the cutting area is an important factor in heat removal.Thus the application of a coolant should be copious and of some veloctty. The possibility of a cuttingflu~d providing lubncahon between the chip and the tool f a ~ is e an attractive one, An effective lubrrcant can modify the process, perhaps producing a cooler chip, discouraging the formation of a bu~lt-up edge on the tool, and promoting improved surface finish. However, the extreme pressure at the toolxhip interface and the rapid movement of the chip away from the cutting edge make it virtually impossible to maintain a conventional hydrodynamic lubr~eat~ng film at the tool-chip interface Consequently, any lubr~cation action is associated primarily with the formation of solid chem~cal compounds of low shear strength on the freshly cut chip face, thereby redncir~g chip-tool shear forces or frict~on.For example, carbon tetrachloride is very effective in reduc~ng friction in machining several different metals and yet would hardly be clasvfied as a good lubricant in the usual sense. Chemically actlve compounds,snch as chlorinated or sulfurized olls, can be added to cutting fluids to achieve such a lubr~cation effect. Extreme-pressure lubricant5 are especially valuable in severe operatlons, such as internal threading (tapping), where the extensive tool-work contact results in high friction with limited access for a fluid. In addition to functional effecnveness as coolant and lubncant,cutting fluids should be stable in use and storage,noncorrosive to

I
Category Solids ?amp fluids Contaminants Metallic fines, chips Grease and sludge Debris and wash Hydraulic oils (coolant) Water (oils)
Bactetia

Effect4
Scrarcb product's surface

Biologicals (coolants)

Fungi Mold

PlugcoOlant lines Produce wear on tools a n d m a c ~ e s Deaeese coaling efficiency Cause smoking Clog paper filters Grow bacteria faster Acidity coolant Break dawn emulsions Cause rancidity, dermatitis Require toxic biocides

I
I

DIRTY FLUIDS 011sor coolants contammated from usage

HYDROCYCLONE (coolant) or FILTER loll) removes sol~ds to 15-30 microns

AUTOMATIC DISCHARGE HIGH-SPEED CENTRIFUGE to 1 mtcron and removes sol~ds tramp fluids to 114 of 1%

MESH FILTER removes all coarse solids

PASTEURIZER reduces biologicals to levels of fresh coolants

CLEAN FLUIDS

available for immediate reuse

A well-designed recycling system for coolants will return more than 99% of the fluid for reuse.
FIGURE 21-24

work and machines, and nontoxic to operating personnel.The cutting fluid should ; b e restorable by uslng a closed recycl~ng system that w l l punfy the used coolant cutting 011s. Cutting fluids become contammated in three ways (Table 21-7.) All th contaminants can be eliminated by filtering, hydrocycloning, pasteurizing, and c trifugmg. Coolant restoration e h m ~ n a t e s 99% of the cost of d~sposal and 80% or m of new fluid purchases. See Figure 21-24 for a schematic of a coolant recycling systt

Key Words
alum~num ox~de BUE (built-up edge) carb~des cast cobalt alloy ceramcs cermet8 chem~cal vapor depos~t~on coated tools crater wear cubic boron nitride (CBW) cutting fluids cutting tool materials DCL (depth-of-cut line) diamonds flank wear hardness hot hardness HSS (high-speed steel) machinability metal cutting physical vapor deposition powder metallurgy sintered carbides stellite tool life titanium nitride wear land

III Review Questions

Firimetalcutting tools, what is the most important material property (i.e., the most critical characteristic)? Why? What is hot hardness compared to hardness? What is impact strength and how is it measured? Why is impact strength an important property in cutting tools? , What is HIP, and how is it used for tool fabrication? , What are the primary considerations in tool selection? , What is the general strategy behind coated tools? What is a cermet? : How is a CBN tool manufactured? , E W. Taylor was one of the discoverers of high-speed steel. What else is he well known for? What casting process do you think was used to fabricate cast cobalt alloys? Discus the constraints in the selection of a cutting tool. , What does cemented mean in thc manufacture of carbides? Failure Excesslve flank wear

14. What advantage do ground carbide inserts have over pressed carbide inserts? 15. What is a chip groove? 16. What is the DCL? 17. Suppose you made four beams out of carbide, HSS, ceramic, and cobalt.'Ihe beams are identical in size and shape, differing only in material. Which beam would do each of the following? a. Deflect the most, assuming the same load b. Resist penetration the most c. Bend the farthest without breaking d. Support the greatest compressive load 18. Multiple coats or layers are put on the carbide base for what different purposes? 19. What are the various ways a cutting tool can fail and what can he done to remedy this? See Figures 21-A and 21-15,

w
/

Tool material Cuttcng condltlons

Basic Remedy

Use a more wear-resjstant grade

carb~de

Decrease speed

.Enlarge the rake angle .Select the correct chip breaker

. speed .Decrease Reduce the depth of cut and feed .aIf less built-up edge occurs, change to susceptible grade (cermet)
the rake angle .Reduce Increase speed (if caused by edge

- Increase the cutting speed and

FIGURE 21-A

594

C H A ~21 R

Cutting Tools for Mach~ning

20. What makes the process that makes xC coatings for tools a problem? See equation 21.1. 21. Why does a 'EN-coated tool consume less power than an uncoated HSS under exactly the same cutting conbtions? 22. For what work material are CBN toolsmore wmmonly used and why? 23. Why is CBN better for machining steel than diamond? 24. Whal is the typical coefficient of variation for tool life data, and why is this a problem? 25. Whatismeant by the statement "Tool life is a random variable"? 26. The typical value of a coefficient of variation in metalcuttiig tool life distributions is 0.3. How could it be reduced: 27. Machinability is defined in many ways. Explain how a rating is obtained.

28. 29. 30. 31. 32. 33.

What are the chief functions of cutting fluids? How are CVD tools manufactured? Why is the PVD process used to coat HSS tools? Why is there no universal cutting-tool material? What is an 18-44 HSS composed of? Over the years, tool materials have been developed allowed stgnificant increases in MRR. Nevertheles still widely used. Under what conditions might HSS b terial of choice? 34. Why is the rigidity of the machine tool an importar eration in the selection of the cutting-tool material? 35. Explain how it can he that the tool wears when it ma times as hard as the work material. 36. What is a honed edge on a cutting tool and why is it

Problems
1. Figure 21-B gives data for cutting speed and tool l ~ f e Determ~ne the constants for theTaylor tool llEe equatlon forthese data

you thrnk the tool materlal might have been?

Tool life in minutes (TI

FIGURE 21-B

2 Suppose you have a turomgoperahon usrng a tool w ~ t h *zero back rake and 5" end rehef. The insert flank has a wear land on it of 0.020 in How much has the dlameter of the workpiece grown (mcreased) due to thls flank wear, assuming the tool has not been reset to compensate for the flank wear?

3. In Frgure 21-C, a srngle polnt tool is shown. Identrfy through G usmg tool nomencl3ture

Single-pant tool

A= 8=

C= D=

*14c

view

YF I-

FIGURE 21-C

.Thefollowtng data have been ohta~nedfor maeh~n~ng an Si-A1 alloy.

Workpiece Matenal
Sand casting Permanent mold castrng PMC wlth flood coollog

Tool Mater~al
Dlamond poly~rystal Dlamond polycrystal Diamond polycrystal WC K 20

Cutting Speed (mimin) for Tool Llfe (mi&) 20 mm

731 591

of

30 min
642 517 554 161

60 min
514 41 1

6118
175

472
139

Sand castrng

Compute the C and n values for the Taylor tool Life equation. How do these n values compare to the typical values? , InFigure 2 1 - 4 the insert at the top is set with a 0' side cutting-edge angle.The insert at the bottom is set so that the edge contact length is increased from a 0.250-in. depth of cut to 0.289 %.The feed was 0.010 ipr. a. Determine the side cutting-edge angle for the offset tool. b. What is the uncut chip thickness in the offset position? c. What effect will this have on the forces and the process?

3
-------- ---

Tube turning with insert tool

tu- 0'

- ----- -- -

0.250

= a 0 .-:

a*&X
Feed = 0.010 ipr

--------- --- -- - --

Tube turning
with
a=OO

inserttool

---

Feed = 0.010 ipr

FIGlJRE21-D

Tool cost is often used as themajor criterion for justifying tool selection. Either silicon nitride or PCBN insert tips can be used to machine (bore) a cylinder block on a transfer line at a rate of 312,000 partslyr (material: gray cast iron).The operation requires 12 inserts (2 per tool), as six bores are machined simultaneously.The machine was run at 2600sfpm with a feed of 0.014 in. at 0.005-in. DOC for finishing. Here are some additional data. SiN
TIPS in use per part Tool life (parts per tool) Cost per lip
12
200 $1.25

PCBN
12
4700

$28.50

a. Whieh tool material would you recommend? b. On what basis?

7. A 2-in.-diameter bar of steel was turned at 284 rpm, and tool failure occurred in 10 min.The speed was changed to 132 rpm, and the tool [ailed in30 min of cutting.Assume that astraightline relationship exists. What cutting speed should be used to obtain a 60-min tool life of V60? 8. Table 21-6 shows a cost comparison for four tool materials. Show how the data in this table were generated. 9. Refer to Problem 1. Show the relationship between cutting speed and tool temperature. What does this mean with regard to tool failure? 10. 'The outside diameter of a roll for a steel (AISI 1015) rolling mill is to be turned. In the final pass, the starting diameter = 26.25 in. and the length = 48.0 in.The cutting conditions will he feed = 0.0100 in.irev and depth of cut = 0.125 in. A cemented carbide cutting tool is to be used, and the parameters of theTaylor tool life equation for this setup are n = 0.25 and C = 1300.It is desirable to operate at a cuttingspeedsuch that

596

C H ~ L 21 K Cutting Toals for Machining

the tool will not need to he changed during the cut. Deterrmne the cutting speed that will make the tool hie equal to the tlme required to complete this turning operatloll (Problem suggested by Groover, Fundamenral~ of Modern Manufacrurzrrg: &Sons, Mntenals, Procrrser, and Systems,2nd ed ,John W~ley
2002 ) I I Using dam from Problems 8 and 10, determine the necesyary horsepower for the mathme tool to make th15 cut.

12 F~gure 21-E shows d sketch of a smgle-po~nt tool and if8 c ~ a t e dtool signature Put the slgnaturc from the to Figure 21-C in the same oideras shown in F1gure21-E tool would produce the larger F, glven that both are cuttld the same V, f,, and DOC In the same matend?

"
l Y U l l i l d l >,Ut:

"-Norm.

side rake angle

rellef angle

End cun~ng-edge angie

Tool Signature Normal back rake angle Normal s ~ d e rake angle Normal end rellef angle Normal side reiief angle End cumng-edge angle S~de cutting edge angle Nose rad~us

l2

FIGURE 21-E

Case Study

597

Comparing Tool Materials Based on Tool Life

The Llnus Dr~lllng Company is trying to decide on what kind of inserts to use in the~r indexable Insert dr11ls.These drills do not cut at dead center but rather form a thin center slug that a pushed out of the way by the dr~ll body.These ~ndexabte dr~lls often prov~de a marked productivity improvement comHDD drills, parbcularly when pared to convent~onal they are combined wlth caated Insert tools The company IS trylng to determine whrch type of Insert to use in the drill for rnachlnlng some hot rolled 8620 steel shafts uslng triangular lnserts.The operating cost of the machine tool IB $60.00 per hour. It takes about three mlnutes to change the Inserts and about part and load a new 30 seconds to unload a f~n~shed part In the machine The company IS currently uslng uncoated inserts at the following operating condit l o n s 4 0 0 sfpm and 0.020 ipr These speeds and feeds resulted in each cuttlng edge producing about 40 pieces before the tool'scuttlng edge became dull. The tool was then Indexed. Slnce it was a trranguiar tool, each tool had 6 cuttlng edges available before it had to be replaced At these speeds and feeds, the drilllng hme was 4.8 mtnutes ahd the produdon rate was 11 parts per hour, whlle the machln~ng cost per plece was 54.80 (51.00imln x 4.8 m~n/pc.).The manufacturingengineer on the job, Br~an Paul, has found three new tool mater~als being used in triangular insert tools They are llsted In Table CS-21 along wlth the data for the uncoated tool. The new materials are TIC-coated carbide, A l , O , c o a t e d carL~de and a ceramlc-coated Insert wlth a single srde, low force groove geometry.fhe expected cuttlng conditions, speeds and feeds, are glven in the table along wlth Brian's estimates of the production rates in p~eces per hour for each of these new tool materiak The low force groove geometry tools can only be used three tim-they cannot be fl~pped over-so only three cuttlng edges are ava~lable par insert before it has to be replaced Bnan has argued that even though the cerarn~c-coated lnserts cost more, they result in a lower cost per piece, cons~derlng all the costs. Determine the machln~ng cost per piece, the tool changing cost per plece, and the tool cost per p~ece which make up the total cost perplece, and verify Brian's belief that these coated tools will provtde some cost savings

Cutrmg edges avmlable per Insert Ci>sfufan Insort ($/insert) (p>cucdcutt~ng sdgc) Tool l~fe Toal change time per pltctcc (mm) Nonproductive cost per plece (Wpc) Machining t ~ m e p cpiece r (mmipc) MdGllnmgcdst pw piece ($ipc) 4.w) Toal change cost perpteca ($/PC) 0 08 Cutt~ng tool cost per paece ($ipc) 0 . U Total cast prrplecc ($) 540 Pmduct~on rate (p%eeasihr) Improvemsnt m productlvlty hilsed on p~ecesihr

TURNING AND BORING PROCESSES

Special Attchments 22.1 INTRODUC~ON 22.2 FUNDAMENTAL^ OF TURNING, DimenSionalAccuracy ' BORING, AND FACING TURNING 22.3 L A TDEIGN ~ AND TERMINOLOGY Boring Facing Lathe Desfgn Parting Size Designation uf Lathes Deflection Types .elf Lathes TOOLS FOR LATHES 22.4 (SJ'ITING Precision Boring Drilling Lathe Cutting Tools Reaming Form Tools Knurling Turret-Lathe Tools 22.5 WORKHOWING IN Lathe Chucks

Ei 22.1 INTRODUCTION
Turning is the process of machining external cylindrical and conical sulfates. It is ally performed on a machine tool called alathe, as shown in Figure 22-1, using a cu tool.The workpiece is held in a workholder. More details on lathes are shown lat relatively simple work and tool mov this chapter. As indicated in Figure 22-2, are involved in turning a cylindrical surface.The workpiece is rotated and the single
axis of rotation, an external conical surface results. This is called ruper turning. I f tool is fed to the axis of rotation, using a tool that is wider than the depth of the cut, operation is calledfacing, and a flat surface is produced on the end of the cylinder. By using a tool having a specicific form or shape and feeding it radially or inw against the work, external cylindrical, conical, and irregular surfaces of limitedlength also be turned. The shape of the resulting surface is determined by the shape and

FIGURE 22-1 Schematic of a standard engine lathe performing a turning operation, with thecutting tool shown in inset.

598

Taper turn~ng

Facing

Contour turnlng

Form turning

End

facing

Pan~ng or

cutoff

Taper baring

Internal threading

Internal forming

Knurling

'eed the too. ong'tua nally lor rdlming and can what o rection the tool 1s fed and on what portion d~fferent names The dasnea arn~ w s ~nd~cate the

of the cutting tool. Such machining is called form turning. If the tool is fed all the way to the axis of the workpiece, it will be cut in two. This is calledparting or cutoff, and a simple, thin tool is used. A similar tool is used for necking or partial cutoff. Boring is a variation of turning. Essentially, boring is internal turning. Boring can use single-point cutting tools to produce internal cylindrical or conical surfaces. It does not create the hole but, rather, machines or opens the hole up to a specific size. Boring can be done on most machine tools that can do turning. However, boring also can be done using a rotating tool with the workpiece remaining stationary. Also, specialized machine tools have been developed that will do boring, drilling, and reaming but will not do turning. Other operations,like threading and knurling, can be done on machines used

600

CHAPTER 22 Turnlng and Boring Processes

for turnmg. In addition, driUing,reaming, and tapplng can be done on the rotation of the work. In recent years, turnlng centers have been developed that use turrets to h multiple-edge rotary tools 1n powered heads. Some new machine tools feature opposrng spindles with automatic transfer from one to the other and two turrets oft

22.2 FUNDAMENTALS OF TURNING, BORING, AND FACING TURNING

Turnirig constitutes the majority of lathe work. The cutting forces, resulting from ing the tool from r~ght to left, should be directed toward the headstock to force workpiece against the workholder and thus provide better work support. If good finish and accurate size are desrred, one or more roughing cuts are us followed by one or more finislung cuts. Roughing cuts may be as heavy as proper thickness, cutting dynamics,tool life, lathe horsepower, and the workp~ece permit. L depths of cut and smaller feeds are preferred to the reverse procedure,becausefewer are required and less time is lost in reversing the carriage and resettrng the tool for following cut. On workpieces that have a hard surface, such as castings or hot-rolled mate1 containing mill scale, theinitial roughing cut should be deep enough to penetrate materials. Otherw~se, the entire cutting edge operates m hard, abras~ve material t out the cut, and the tool w~ll dull rapidly. If the surface is unusually hard, the cut speed on the first roughing cut should be reduced accordrngly. Finishing cuts are light, usually being less than 0.015 in.in depth, with the fee fine as necessary to grve the deslred fimsh. Sometimes a special finishng tool is used often the same tool is used for both roughing and finishing cuts. In most cases,one ishing cut is all that is required. However, where exceptional accuracy is requued, fm~shing cuts may be made. If the d~ameter 1s controlled manually, a short fmishlng ('I4rn long) is made and the dlameter checked before the cut is completed Bec previous rn~crometer measurements were made on a rougher surface, it may h sary to reset the tool in order to have the ftnal measurement,made on a smoo face, check exactly. In turning, the primary cutting motion is rotatronal, with the tool feeding par to the axls of rotation (Figure 22-3).The desrred cutting speed establrshed the news rpm (N,) of the rotating workpiece. The feedf,is given m inches per revolution (ip The depth of cut is d, where

d = DOC=

4 - Dz mches .
2

The length of cut is the distance traveled parallel to the mrs L plus some allowance overrunA to aUow the tool to enter and/or exit thecut. Here is how the mputs to the turning process are detemnlned. First it is necess that the englneer select the cutting speed, V,1n feet per minute, the feed CfJ, and depth of cut, based on what matenal is beingmachined, what tool material is bemg u

FIGURE 22-3 Baslcs of the turn~ng process normally done on a lathe. The dashed arrows ~nd~cate the feed mollon of the tool relat~ve to the work

slngle pomt tool process

and other factors l ~ k what e process 1s being performed. The rpm value for the machlne tool can be determined by hold two ools. (using the larger diameter), where the factor of 12 is used to convert feet to inches.The cutting time is

eed:the

where A is overrun allowance, and f, is the selected feed in inches per revolution.

P~J~Y chip arge cuts r the

Example of Turning
The 1.78-inch diameter steel bar shown in Figure 1-10 is to be turned down to a 1.10-inch diameter on a standard engine lathe.The overall length of the bar is 18.750 inches, and the region to be turned is 16.50 inches. After turning, the bar looks like stage 2.The part ismade from cold-drawn free-machining steel (this means the chips breakupnicely) witha Bhn of250. Since you want to take the bar from 1.78 to 1.10,you havea total depth of cut,d,of 0.34 inch (0.68/2).Yuudecide you want to make two cut$ a roughing pass and a finishing pess. Rough at d = 0.300 and finish with d = 0.04 inches. Looking at the table for selecting speed and reed (Figure 20.4) you select V = 100 fpm and feed = 0.020 ipr because you have selected high-speed steel cutting tools. The bar is held in a chuck with a feed through the holein thespindle and is supported on the tight end wilh a live center. The ends of the bar have been center drilled.Allowance should be 0.50 inches far approach (no avertravel). Allow 1.0 minutes to reset the tool after the first cut.To determine the inputs ta thc lathe, we calculate the spindle rpm:

rials hard

ughtting

:d as .but ,filltwo 5 cut :the :cesSur-

N, = 12V/rD,

= 12 X 100/3.14 X 1.78 = 214

But your lathedoes not have thisparticuiairpm,so you select theclosest rpm,which is200.Youdon't need any further calculations for lathe inputs as you input the feed in ipr directly. The time t o make the cut is L ALL _ 1650 050 T ,= = 425mm 0 020 X 200 fr X N.

You could reduce this time by changing to a coated carbide tool that would allow you tu increase the cutting speed to 925 sfpm. (See table). The time for the second cut will be different if you change the feed and/or the speed to improve the surface finish.Again from the table in Figure 21-4, speed = 925, feed = 0.007, so

N,

12 X 925/3.14 X 1.10 = 3213 with 3200 rpm the closest value:

T,

+ ALL

0 007 X 3200

O75mm

The metal removal rate is MRR

f the nse4
=

volume removed (710: - .rrD?)L time 4LliN

(omttmg the allowance term). By rearranging and substituting N,, an exact expression for MRR is obtained: MRR Rewriting the last term
=

12Vf,

(0: - 0 : ) 401

Therefore, since

4 - 4 and 4 - 4
2

20,

-- 1

for small d

then,

MRR

Note that Equation (22-5) 1 s an approximate equation that assumes that the depth(
d is small compared to the uncut diameter Dl.

12 V f r d in?/min

BORING
Boring always involves the enlarging of an existing hole. which may have been mat
a dnll or may be the result of a core in a casting An equally important and concu purpose of bonng may be to make the hole concentric with the axls of rotation o workpiece and thus correct any eccentrictty that may have resulted from the drill: C boredb ing or drlfting off the center line. Concentricity is an important attribute o When bor~ng is done in a lathe, the work usually is held in a chuck or on a plate. Holes may be bored straight, tapered, with thready, or to irregular cont, Figure 22-4 shows the relationship of the tool and the workpiece for borlng 7 t h boring as internal turning while feeding the tool parallel to the rotation axls o workpiece, with two important differences. First, the relief and clearance angles on the tool should he larger and the tool ( hand (length to diameter) must be considered with regard to stability and deflei problems Given Vand f,fos a cut of length L, the cutting time 1s

where N, = 12VIvD, for D,,the diameter of bore, and A, the overrun allowance. metal removal rate is

MRR

L(rrD:

rrD;)/4

LIJN

where D,is the original hole diameter

MRR

-- 12 V j d

(omitting allowance term), where d is the depth o C cut. In most respects, the same princrples are used for boring as for turning. Agair tool should be set exactly at the same height as the axis of rotation. Larger end c ance angles help to prevent the heel of the tool from rubbing on thc inner surfacec hole. Because the tool overhang will be greater,feeds and depths of cut may be red to reduce forces that cause tool mbration and chatter. In some cases, the bonng bar be made of tungTten carbide because of this mateiial's greater st~ffness.

FACl NG Fuczngis the producing of a flat surface as the result of the tool be~ng fed across thc of the rotating workp~ece, as shown in Figure 22-4b Unless the work is held on ai drel, d both ends of the work are to be faced, it must be turned end for end after the end is completed and the facing operation repeated. Machined surface surface Mach~ned

Tool
FIGURE 22-4 Bas~c movement of borlng, fac~ng, and cutoff (or partlng) process.

( a ) Borlng a dnlled hole

ib) Faclng from the cross s~de

(c) Cutoff or

partlng

SECTION 22.2

Fundamentals of Turning, Bonng, and Facing Turning

The cuttlng speed should be determined from the largest d~ameter of the surface to be faced. Facing may be done e~ther from the outs~de ~nward or liom the center outward In either case, the point of the tool must be set exactly at the he~ght of the center of rotation. Because the cutting torce tends to push the tool away Cram the work, it is usually des~rable to clamp the carnage to the lathe bed durmg each fac~ng cut lo prevent it from moving slightly and thus producing a surface that 1s not flat. In the facing of castings or other materials that have a hardsurface, the depth of the first cut should be sufficient to penetrate the hard material to avo~d excesslve tool wear In facing, the tool feeds perpendicular to the axis of the rotating workp~eccBecause the rpm is constant, the cutting speed 1s cont~nually decreas~ng as the axls 1s approached. The length of cut L is D,!2 or (Dl - D,)!2 for a tube

T,, = cutting time

=~lllnutes

L + A

frj"

VOL - .irD?df,N = 6Vf,d in. '/min 4L T,,, where d 1s the depth of cut and L = D,!2 is the length of cut

MRR

=-

PARTING
Parting is the operation by which one section of a workpiece is severed from the ramamderby means of a cutoff tool,as shown m Figure 22-4c Bccause parting tools are qulte thm
and must have cons~derable overhand, this process 1s more ddficult to perform accurately. The tool should be set exactly at the height of the axls of rotation, be kept harp, have ptoper clearance angles,and be fed into the workpzece at a proper and uniform feed rate In partmg or cutoff work, the tool 15 fed (plunged) perpendlculal to the rotat~onal axis as it was in facing.The length of cut for solid bars 1s D,12. For tubes.

In culoff operations. the width of the tool IS d i n inches, the w~dlh of the cutolf operatlon.7he equations for T,, and MRR are then basically the same a$ for tacing.

DEFLECTION
In bonng, facing, and cutoff operations, the speeds, feeds,and depth of cut selected are generally less than those recommended for stralght turn~ng because oi the large overhang of the tool often needed to complete the cuts Recall the baslc equation for deflection of a cantilever beam, modifying for machining,

In equation 22-9,l represents the overhang of the 1001, which greatly affects the deflect ~ o nso , it should be minimized whenever poss~ble. In equat~on 22-9,

E I
where

= =

modulus of elasticity moment of lnertta of cross Yeet~on of tool

=

P = F,

apphed load or cutllng force

I = rrD:/64 solid round bar

I
D,

= = =

v(D; - (D;))/64 barw~th hole diameter of tube or bar Inside d~ameter of the tube

D,

Turnlng and Bor~ngProcesses Deflection 1s proportional to the fourth power of the boring bar dlameter and the1 power of the bar overhand. Select the largest-diameter bar drameter and minimm shank boring bars (E 80,000,000 psi). and select tool gec overhand Use carb~de tnes that direct cuttlng forces into the feed d~rectlon to m~mmize chatter. The re tlon of the feed or depth of cut reduces the forces operatrng on the tools The cu speed usually controls the occurrence of chatter and vibration. See the dynamii machining discu~slon in Chapter 20. Any ~mbalance in the cuttlng forces w~ll detlect the tool to the slde, resulbngm of accuracy in cutoff 1engths.At the outset, the forces will be balanced if there 1 s no rake on the tool As the cutoff tool redches the axis of the lotatlngpart, the tool wi deflected away from Lhe spindle, resulting ~na change in the length of the part.

--

PRECISION BORING Sometimes bored holes are shghtly bellmouthed because the tool detlects out of the as it progresses into the hole.This often occurs in castmgs and forgings where the holes draft angles so that the depth of cut increases as the tool proglesses down the bore. problem can usually be corrected by repeatlug the cut with the same tool se however, the total cutting time for the part is increased.Alternately, a more robust can be used. Large holes may be prec~sion bored using the setup ~ h o w n m figure where a pilot bushing is placed in the sprndle to mate with the hardened ground pll the boring bar.This setup eluninates the cantilever problem? common to honng. Because the rotational relationship between the work and the loo1 1s a simple and i s employed on several types oI machine tools,mch as lathes,drilllng machine milllng machines, boring is very frequently done on such machines. However, se machine tools have been deveIoped prlmarlly for bor~ng, espec~ally in cases lnvo large workpleces or for large-volume boring of smaller parts. Such machines as thes also capable of perform~ng other operatcons,such as milling and turnlng Because rng frequently follows drilhng, many borlng machmes also can do dnlllng, pemi both operattons to be done w ~ t h a smgle setup of the work DRILLING
Dr~Nlng, discussed in detail in Chapter 23, can be done on lathes with the drill mo

In the tallstock qu~ll of engine lathes or the turret on turret lathes and fed aga

Cunlng tool (HSStool b ~ t i

P~lot boring bar mounted in tallstock of lathe for precision boring large hole in castlng The size of the hole is controlled by the rotation diameter of the cutting tool
FIGURE 22-5

SLC'II~N 22.2

Fundamentals of Turn~ng, Bor~ng, and Faclng Tum~ng

605

rotatmg workpiece. Straight-shank drills can be held in Jacobs chucks. or dr~lls with taper shanks mounted directly in the quill hole can drrll holes online (center of rotat~on). Drllls can also be mounted in the turrets of modern turret centers and fed automat~cally on the rotat~onal axis ot the workpiece or off axls with power heads. It also is possible to dr~ll on a lathe wlth the drill blt mounted and rotated in the spmdle while the work remains stationary, supported on the tallstock or the carnage of the lathe The usual speeds used for drihng should be selected for lathe work.Because the feed may be manually controlled, care must be exercised, particularly m d n h g small holes. Coolants should be used where required. ln drilling deep holes. the drill should be withdrawn occasionally to clear chips from the hole and to a ~ in d getting coolant to the cuttmg edges.This is called peck drillmg. See Chapte~ 23 for further d~scussion on drilling.

REAMING Reaming on a lathe involves no spec~alprecautions. Reamets are held in the tallstock quill, taper-shank types being mounted directly and stra~ght-shank types by means of a drlll chuck. Rose-chuckmg reamers are usually used (see Chapter 23) Fluted-chuckmg reamers may also be used, but these should be held in some type of holder that will permlt the reamer to float (~.e., have some compliance) in the hole and conform to the geometry created by the bonng process. KNURLINC Knurlrng produces a regularly shaped, roughened suriace on a workp~ece. Although knurlmg also can be done on other machine tools, even on flat surfaces. in most cases it rs done on external cylindrical surfaces using lathes Knurling is a chipless,cold-formmg process See Figures 22-6 and 22-18 for examples. The two hardened rolls are pressed against the rotating workpiece with sufficient force to cause a shght outward and lateral displacement of the metal to form the knurl in a ra~sed, diamond pattern. Another type ot knurlmg tool produces the knurled pattern by cuttmg chips Because it mvolves less pressure and thus does not tend to bend the workpiece, thls method 1 s often preferred for workpieces of small d~ameter and for use on automatic or semlautomatlc machmes. SPECIAL ATTACHMENTS For englne lathes, taper turning and m~lling can be done on a lathe but require special attachments.The rnrlhng attachment 1s a special vise that attaches to the cross slide to hold work.The rnilhng cutter 1 s mounted and rotated by the spmdle.The work is fed by means Tool-posIgrindersare often used toperm~t grmd~ng to be done of the crosy-slide sc~elv on a lathe. Taper turning will be discussed later. Duplzcaling attachments are available that, gulded by a tcmplate, will automatlcafly control the tool movements for turnmg irregularly shaped parts. In some cases, the first piece, produced in the normal mamer,may serve as the template for duplicate parts.To alarge extent duplicat~ng lathes using templates have been replaced by numerically controlled lathes and m~lling is done with power tools in NC turret lathes,

(a) Knurling in a ie, using a forming-typetool, jshowing the resulting tern on the workpiece; knurling tool with forming s. (Courtesy of Armstrong thers Tool Company.)
;URE 22-6

rolls ib)

606

CHA~ER 22

Turning and Boring Processes

DIMENSIONAL ACCURACY Dzmenszonal accuracy m turning operations is controlled by many factors. Pmn influenced by deflection due to the cutting forces and surface roughness.Too1 wear the workplece dirnenslon to change from the ~nltxal diameter when the tool is sha the d~ameter obtained after the tool has worn The cutt~ng forces Increase as th wears, resulting in increased deflection between the workpiece and the cutting built-up edge (BUE) may form at the tip of the cutting tool. A BUE has the te to change the actual diameter of the workpiece.Thus, to hold close tolerances, t of the wear land, the magnitude of the radial (thrust) force, and the elimination B U E should be taken into account. See F~gure 22-7 Dimensional accuracy w~ll also be d u e n c e d by the workpiece shape,the mat the rigidity of all elements, the surface finish, and vibrations. For example, hold dimens~onal accuracy of a boring operation on a deep hole is a problem, due deflection (rigidity) of the b o ~ ~ bar. ng Turned surfaces d~splay character~stic turnlng grooves that are produce feed and the tool tip corner radius, as shown in Figure 22-7. The roughness from feed marks from a round-nosed tool can be approximated by the formula

The feed and the comer radrus (CR) of the cutting tool influence the surface roughnes Diameter after

force tool)

worn tool, no BUE Change in diameter due to tool wear The d~ameter of a workp~ece becomes larger, as do the thrust forces as the cuttlng tool wears on the flank durlng turn~ng

T
d

FIGURE 22-7 Accuracy and preclslon in turning is a funct~on of many factors, lnclud~ng tool wear and BUE

Regardless of whether the tool 1 s dull or sharp, a bullt-up edge (BUE) causes the diameter of the workpiece to be smaller than deslred

SECTION 22.3 Lathe Des~gn and Terminology

607

where y is the roughness height, CR the corner rad~us of insert, and f, the feed rate (tn./rev).To improve the surface finish, reduce the feed and increase the corner radus. Other factors like BUE formations, cutt~ng-edge sharpness, and tool wear grooves in the flank wear area also affect the surface finish in turning Flank wear and BUE can combme to atfect both surface finish and accuracy, as shown m Figures 22-1 and 22-7. Wear on the comer radius may cause grooves and nicks, which produce additional surface roughness on the finish-turned surfaces. Thus, to hold the surface roughness wlthin specified limits, minimize tool wear and use small feeds and large-corner-radius tools. To minimize BUE formation, employ cutting speeds higher than those used m rough turning operations.

Knowing the terminology of a machine tool is fundamental to understandng how it performs the basic processes, how the workholding devices are interchanged, and how the cutting tools are mounted and interfaced to the work. Lathes are macbme tools designed primarily to do turning, facing, and boring. Very little turning is done on other types of machine tools, and none can do it w t h equal fac~lity. Lathes also can do facing, drilling, and reaming, and recent designs permit milling and drilling operations using live (also called powered) spindles in mulhple-tool turrets, so their versatihty p e m t s multiple operations to be done with a slngle setup of the workpiece. Consequently, the lathe i y probably the most common machine tool, along with milling machines Lathes in various forms have existed for more than 2000 vears. but modern lathes

a changegear system that could connect the motions of the spindle and leadscrew and thus enable threads to be cut. LATHE DESIGN - The essential components of an engine lathe (Figure 22-8) are the bed, headstock assembly (which includes the spindle), tallstock assembly,carriage assembly,quick-change gearbbx, and the Leadscrew and feed rod. The bed is the base and backbone of a lathe.

Quick-change gearbo for feed

Compound rest and slide (swivels)

- - - - - power feed

FIGURE 22-8

Schernat~c diagram of an engine lathe, showing bas~c components.

Turnlng and Boring Processes The bed is usually made of well-normalized or aged gray or nodular cast Iron components are nloun provides a heavy, rigid trame on whlch all the other bas~c Two sets of parallel, longitudinal ways, inner and outer, are contained on the bed. modern lathes, the ways are surface hardened and preasion machined, and care sh be taken to assure that the ways are not damaged Any inaccuracy m them usually m that the accuracy of the entire lathe 1s destroyed. The headstock, mounted in a fixed position on the inner ways, provides pow means to rotate the work at various rpm values. Essenhally, it conssts of a hollowsp mounted in accurate bearings, and a set of transmrssion gears-s~milar to a truck t mission-through whrch the spindle can be rotated at a number of speeds. Most la provide from 8 to 18cholces ofrpm. On modemlathes all the rpm rates can be obt merely by rnovlng from two to four levers ot the lathe has a continuouslyvariables rpm using electrical or mechanical drives. The accuracy of a lathe 1s greatly dependent on the sprndle It carries the holders and is mounted in heavy bearings,usually preloaded tapered roller or ball The spindle has a hole extendrng through its length, through which long bar stoc be fed.The size oi this hole is an important dimension of a lathe because it deter the maximum size of bar stock that can be machined when the materials must b through the spindle, The spindle protrudes from the gearbox and contains means for mounting v types of workholding devices (chucks,face and dog plates, collets). Power rs suppli the spindle from an electric motor through a V-belt or hilent-chain drive. Most mo lathes have motors of from 5 to 25 hp to provide adequate power for carb~de and cer tools cutting hard materials at high cutting speeds. For the classic englne lathe, the ta~lstockassembly consists, essent~ally, of t parts A lower castlng fits on the rnner ways of the bed, can shde longitud~nally be clamped in any desired location. An upper castlng fits on thc lower one an moved transversely upon it, on some type of keyed ways, to permit ahgning the and headstock spindles (for turnlng tapers).The third major component of the bly is the tallstock quill, This is a hollow steel cylinder, usually about 2 to 3 in in ter, that can be moved longitudinally in and out of the upper casting by means handwheel and screw. The open end of the quill hole has a Morse taper. Cutti or a lethe center are held in the quill. A graduated scale is usually engraved on f the quill to aid in controlling its motlon ~nand out of the upper casting. side o ing device permits clamping the quillin any desired posit~on. 1 1 1 recent years, dualNC turning centers have emerged, where a subspindle replaces the tailstock as Parts can be automatically transferred from the spindle to the subsplndle for tu the back end of the part. See Chapters 31 and 36 for more on NC turnlng centers The carnage assemhLy, together wth the apron, provrdes the means for mountin moving cutting tools.The carnage, a relatively flat H-shaped casting,rides on theo of ways on the bed.The cross slde is mounted on the carriage and can be moved b of a feed screw that is controlled by a small handwheel and a graduated dial.The c thus provides a means for movlng the lathe tool in the facing or cutoff direction. On most lathes, the tool post is mounted on a compound r a t . See Figure 2 compound rest can rotate and translate with respect to the cross slide, permit tber pos~tloning of the tool with respect to the work.The apron, attached to the the carriage, has the controls for providing manual and powered motion for riage and powered motion for the cross shde.The carrrage is moved parallel tot by turning a handwheel on the front of the apron, which isgeared to a pin~on on t side.This pinion engages a rack that is attached beneath the upper front edge of th In an inverted position. Powered movement of the carnage and cross slide 1 s provided by a rotat1 rod.The feed rod, which contarns a keyway, passes through the two reversing be ions (Figure 22-10) and is keyed to them. Either pin~on can be actrvated by means feed reverse lever, thus providing "forward" or "reverse" power to the carnage. SUI clutches connect e~ther the rack pinion or the cross-slide screw to provide longitu motlon of the carriage or transversemotion of the cross slide.

SECI~ON 22.3

Lathe Design and Terminology

For cutting threads, a leadscrew is used When a Friction clutch is used to drive carriage, motion through the leadscrew 1s by a dlrect, mechanical connection betw the apron and the leadscrew. A split nut is closed around the leadscrew by means o lever on the front of the apron directly driving the carriage without any slippage. Modem lathes have qurck-change gearboxes, driven by the spmdle, that conn the feed rod and 1eadscrew.Thus when the spmdle turns a revolution, the tool (mount on the carnage) translates (longitudmally or transversely) a spec~fie d~stance in inch that is, inches per revolut~on (lpr). This revolutions per mmute, rpm or N,, time feed,f,, gives the feed rate.5 m rnches per minute that the tool is movmg. In this w the calculations for turning rpm and feed m Inches per revolution are "mechanica related."Typical lathes may prov~de as many as 48 feeds, ranging from 0.002 to 0.118 (0 05 to 3 mm) per revolut~on of the spindle, and, through the leadscrew, leads from 92 threads per inch.

SIZE DESIGNATION OF LATHES

The slze of a lathe is designated by two dimensions.The first is known as the swmg. IS the maximum diameter of work that can be rotated on a lathe. Swing is approxlma twice the distance between the line connecting the lathe centers and the nearest p on the ways.The maxlmum diameter of a workp~ece that can be mounted between ters 1s somewhat less than the swing d~ameter because the workpiece must clear nagc assembly as well as the ways.The second size dimension is the mmzmum between centers.The swing thus indicates the maximum workp~ece diameter that c turned ~nthe lathe, whlle the distance between centers Indicates the maximum len of workp~ece that can be mounted between centers.

TYPES OF LATHES
Lathes used m manufacturing can be classified as speed, engine, toolroom, turret matics, tracer, and numerical control turning centers Speed lathes usually have headstock, a tallstock, and a simple tool post mounted on a light bed. They ord~n have only three or four speeds and are used primarily for wood turning, polishi metal spinning. Sp~ndle speeds up to about 4000 rpm are common. Engine lathes are the type most frequently used in manufacturmg. Figure 22-1 F~gure 22-8 are examples of t h ~ type.They s are heavy-duty mach~ne tools wlth all the ponents descr~bedprev~ously and have power drive for all tool movements except o compound rest.They commonly range in size from 12- to 24-in. swing and from up to 12 ft are no 48-111. center distances, but swlngs up to 50 in. and center d~stances common. Very large englne lathes (36- to 60-ft-long beds) are therefore capabl performingroughing cuts in iron and steel at depths of cut of 'I2 to 2 m. and at cu WC tools run at 0.010 to 0.100 in /rev. To perfom speeds at 50 to 200 sfpm w ~ t h heavy cuts requires rigidity in the machlne tool, the cutting tools, the workholder, the workp~ece (uslng steady rests and other supports) and large horsepower (5 I00 hp). Most engine lathes are equipped with chip pans and a built-~n coolant clrcu system. Smaller englne lathes, with swings usually not over 13 m , also are ava~la bench type, des~gned fot the bed L o be mounted on a bench or table. Toolroom lathes have somewhat greater accuracy and, usually, a wider r speeds and feeds than ord~nary englne lathes. Designed to have greater versa meet the requirements of tool and d ~ work, e they often have a continuously spindle speed range and shorter beds than ordinary engine lathes of comparab relatively small parts. They may bee smce they are generally used for machin~ng bench or pedestal type. Several types of special-purpose lathes are made to accommodate specific work. On a gap-bed Lathe, for example, a sechon of the bed, adjacent to the hea can be removed to permit work of unusually large dlameter to be swung. example is the wheel lathe, which 1s designed to permlt the turnlng of rail wheel-and-axle assemblies. Figure 22-11 shows a CNC vert~cal turning lathe. Vertical lathes are an ex alternative to large horrzontal CNC lathes. Gravity-aided seating of largeheavy

22.3 Lathe Design and Termln, SECTION

i 22-11 This CNC

turning center 1s used for large c~rcular parts under vert~cally mounted

pieces allows a high degree of process repeatability. A smaller footprint, lower initial cost, and increased productivity are all advantages when compared to traditional horizontal lathes. Although engine lathes are versatile and very useful, the time required for changing and setting tools and for making measurements on the workpiece is often a large percentage of the cycle time. Often, the actual chip-production time is less than 3056 of the total cycle time. Methods to reduce setup and toolchange time are now well known, reducing setups to minutes and unloadnoad steps to seconds. The placement of singlecycle machinetools into interim or lean manufacturing cells will increase the productivity of the workers because they can run more than one machine.Turret lathes, screw machines, and other types of semiautomatic and automatic lathes have been highly developed and are widely used in manufacturing as another means to improve cuttingproductivity.
Turret Lathes The basic components of a turret lathe are depicted in Figure 22-12. Basicalk a longitudinally feedable, hexagon turret replaces the tailstock. The turret, on which six tools can be mounted, can be rotated about a vertical axis to bring each tool into operating position, and the entire unit,can be translated parallel to the ways, either manually or by power, to provide feed for the tools. When the turret assembly is backed away from the spindle by means of a capstan wheel, the turret indexes automatically at the end of its movement, thus bringing each of the six tools into operating position in sequence. The square turret on the cross slide can be rotated manually about a vertical axis to bring each of the four tools into operating position. On most machimes, the turret can be moved transversely,either manually or by power, by means of the cross slide, and longitudinally through power or manual operation of the carriage. In most cases, a fixed tool holder also is added to the back end of the cross slide; this often carries a parting tool. Through these basic features of a turret lathe, a number of tools can be set up on the machine and then quickly be brought successively into working position so that a complete part can be machined without the necessity for further adjusting, changing tools, or making measurements. The two basic types of turret lathes are the ram-type turret lathe and the saddletype turret lathe. In the ram-type turret lathe, the ram and turret are moved up to the cutting position by means of the capstan wheel, and the power feed is then engaged. As the ram is moved toward the headstock, the turret is automatically locked into position so that rigid tool support is obtained. Rotary stopscrews control the forward travel of

---

Adiustment Feed ---*Spindle speed

Square turret (indexes1

Ram-turret iathe (light duty) Rear tool (1 tool station)

Top view
I

HeadStock

FIGURE 22-12

Block of ram- and d~agrams saddle-turret lathe.

Cross-slide turret (4 tool stations1 square turret

ram turret

thc ram, one stop being provided for each face on the turrct.The proper into operating position automatically when the turret is indexed. A similar set of st is usually provided to limit movement of the cross slide provides a more rugged mounting for the hexagon turret than can be ram-type mounting. In saddle-type laches, the main turret is mounted saddle, and the entire saddle and turret assembly reciprocates. Larger turret lath usually have this type of mounting. However. because the saddle-turret assembly rather heavy, this type of mounting provides less rapid turret reciprocation. When suc lathes are used with heavy tooling for making heavy or multiple cuts, apilor arm a tached to the headstock engages a pilot hole attached to one or more rct to give additional rigidity.Turret-lathe headstocks can shift rapidly between spindl speeds and brake rapidly to stop the spindle vcry quickly. They also stock-feeding for feeding bai- stock through the spindle hole. If the work is to be held a chuck, some type of air-operated chuck or a special clamping fixture is often employ to reduce the time required for part loading and unloading. Single-SpindleAutomatic Screw Machines. There ar spindle screw rncrchines. One, an American development and common ret type (Brown & Sharpe), is shown in Figure 22-13.The other is of Swiss origin and referred to as the Swiss typc.The Bvown & Sharpe screw machine is e automatic turret lathe, designed lor bar stock, with the main turret mounted in a ve cal plane on a ram. Front and rear tool holders can be mounted on the cross slide. motions of the turret, cross slide, spindle, chuck, and stock-feed mechan

Upper slides

Spindle

Tool turret

reverse and turret index

J
I

, , .

IGURE 22-13 On the tuhet-type single-spindle automatic, the tools must take turns to make cuts.

by cams,The turret cam is essentially a program that defines the movement of the turret during a cycle.These machines are usually equipped with an automatic rod-feeding magazine that feeds a new length of bar stock into the collet (the workholding device) as soon as one rod IS completely used. with a transfer Often, screw machlnes of the Brown & Sharpe type are equ~pped or "picking" attachment.This device picks up the workpiece from the spindle as it is cut off and carries it to a position where a secondary operation is performed by a small, auxiliary power head. In this manner screwdriver slots are put in screw heads, small flats are milled parallel with the axis of the workpiece,or holes are dnlled normal to the axis. O n the Swiss-type automatic screw machine, the cutting tools are held and moved in radial slides (Figure 22-14). Disk cams move the tools into cutting position and provide fecd into the work in a radial direction only; they provide any required longitudinal feed by reciprocating the headstock.

614

CHAPTER 22

Turning and Boring Processes Most machining on Swiss-type screw macl~ines is done with single-point tools.Because they are located close to the spindle collet, the workpiece is not su to much deflection. Consequently, these machines are particularly well sui machining very small parts. Both types of single-spindle screw machines can produce work to close tole the Swiss-type probably being somewhat superior for very small work. Tolera 0.0002 to 0.0005 in. are not uncommon.The time required for setting up the ma

piece is the sum of the individual machining and tool-positioning times. On m spindle screw machines, sufficient spindles usually four, six,or eight, are provided

a cutting tool (or tools) for each spindle.Tools are fed by longitudinal reciprocati Most machines have a cross slide at each spindle position so that an additional

the end tool slide for each spindle (except for a stock-feed stop at position 6), w slide moves forward, these tools cut essentially simultaneously.At the same time,t in the cross slides move inward and make their cuts. When the forward cutting of the end tool slide is completed,it moves away from the work, accompanied by of the spindle carrier, to position each part for the next operation to be perfor spindle position 5,finished pieces are cut off.Bar stock 1in. in diameter is fed to length for the beginning of the next operation.Thus a piece is completed each t' siderable range of sizes, determined by the diameter of the stock that can be

are typical.

H 22.4 CUTTING Toow FOR LATHES

LATHE CUTTING TOOLS
Most lathe operations are done using single-point cutting tools, such as the cla designs shown in Figure 22-16. On right-hand turning (and left-hand turning) ing tools, the cutting usually takes place on the side of the tool; therefore, the SI

SECTION 22.4

Cutting Tools for Lathes

615

Cutoff

End tool

n lm6
Headstock

slides

? ! slides !c

All sprndles on mult~ple-sp~ndie autornatlc have the same tool path

The

SIX

spindle autornatlc

Spindle arrangement for6spindle automatic. The barstock is usually fed to a stop at position 6. The cutoff position is the one preceding the bar feed position.

s"ss<a"ENr

OPC".

51r*=o

. -

Tooling sheet for making a part on a six-spindle.

FIGURE 22-15

The multiple-spindle, automatic screw machine makes all cuts simultaneously and then performs the noncutting functions (tool withdrawal, index, bar feed) a t high speed.

616

CHAmen 22

Turning and Boring Processes

Right-hand

turning tool

R~ght-hand facing tool

(b)

Grooving Cutoff

Boring

FIGURE 22-16 Common types of forged tool holders: (a) right-hand turning, (b) facing, (c) grooving cutoff, (d) boring, (e) thread~ng. (Courtesy of Armstrong Brothers
Tool Company.)

Threading

angle is of primary importance, particularly when deep cuts are being made. On t round-nose turning tools, cutoff tools, finishing tools, and some threading tools, cut takes place on or near the tip of the tool, and the back rake is therefore of importan Such tools are used with relatively light depths of cut. Because tool materials are expensive, it is desirable to use as little as possible. the same time, it is essential that the cutting tool be supported in a strong,rigid ma to minimize deflection and possible vibration. Consequently, lathe tools are suppo in various types of heavy, forged steel tool holders, as shown in Figure 22-16. The hi speed steel (HSS) tool bit should be clamped in the tool holder with minimum overh otherwise tool chatter and a poor surface finish may result. In the use of carbide, ceramic, or coated carbides for higher speed cutting, thr away inserts are used that can be purchased in a great variety of shapes, geometries (n radius, tool angles, and groove geometry), and sizes (see Figure 22-17 for some example

Triangular insert

35" d~amond

FIGURE 22-17 Typical insert shapes, available cutting edges per insert, and insert holders for throwaway insert cutting tools. (Adapted from Turning Handbook of High Efficiency Metal Cutting, courtesy of General Electric Company.)

TURNING. FACING. AND BORING TOOL HOLDER

V-Slot holds round bor~ng bars as well

as

Quick-change ornpanyit'9 rtesy of '8 Tool

TURNING AND FACING TOOL HOLDER

KNURLING TOOL HOLDER

Takes turning and

facing tool blts

Revolving head, selfcentering 3 palrs of knurls

When several different operations on a lathe are performed repeatedly in sequence, the time required for changing and setting tools may constitute as much as 50% of the total cycle time. Quick-change tool holders (Figure 22-18) are used to reduce manual tool-changing time. The individual tools, preset in their holders, can be interchanged in the special tool post in a few seconds. With some systems, a second tool may be set in the tool post while a cut is being made with the first tool and can then be brought into proper position by rotating the post. In lathe work, the nose of the tool should be set exactly at the same height as the axis of rotation of the work. However, because any setting below the axis causes the work to tend to "c1imb"up on the tool, most machinists set their tools afew thousandths of an inch above the axis, except for cutoff, threading, and some facing operations.

and Bor~ngProcesses

Form turnlng

FIGURE 22-19

Clrcular and block types of form tools. (Courtesy of Speed, Tool Cornpan): Incorporated)

FORM TOOLS
In Figure 22-15, the use of form tools was shown in automatic lathe work. Form tools made by grinding the mverse of the deslred work contour Into a block of HSS or t stee1.A threading tool is often a form tool.Although form tools are relatively expe to manufacture, it is possible to machine a falrly complex surface with a angle in feeding of one tool. For mass-product~on work, adjustable form tools of either flat o tary types, such as are shown in Figure Z-19, are used. These are expensive to m initially but can be resharpened by merely gnnding a small amount off the face and t raising or rotatlng the cutting edge to the correct pos~tion The use of form tools is limited by the difficulty of grinding adequate r for all points along the cuthng edge. A ngld setup 1s needed to resist the la forces that develop with these tools Light feeds with sharp,coated HSS tools are us on multiple-spindle automatics, turret lathes, and transfer line machines.

TURRET-LATHE TOOLS
In turret lathes, the work is generally held in collets and the 0 1rect amount of bar st is fed into the machlne to make one part.The tools are arranged in sequence at the stations with depths of cut all preset. The following factors should be considered w setting up a turret lathe.

1. Setup time: time required to install and set the toohng and set the stops. Stan
tool holders and tools should be used as much as possible to minlmlze setup Setup tune can be greatly reduced by eliminating adjustment in the setup. Workholding time: time to load and unload parts and/or stock. Machme-controllzng trme: time required to manipulate the turrets. Can be reduce comb~ning operations where poss~bleDependent on the sequence of operat established by the design of the setup. Cutting time time during which chips are being produced Should be as short a economically practical and represent the greatest percentage of the total cy time possible. Cost. cost of the tool, setup labor cost, lathe operator labor cost, and the number pieces to be made.

2.

3.

4.

5.

There are essentially eleven tool~ng stations,as shown in F~gure 22-20, w ~ t h six in the ret, four in the indexabIe tool post, and one in the rear tool post. The tooling is m rugged in turret lathes because heavy, simultaneous cuts are often made.Tools moun in the hex turret that are used for turning are often equipped w~th pressure rollers se the opposite side of the rotating workpiece from the tool to counter the cutting for Turret lathes are most econom~cal in producing lots too large for engine la but too small for automatic screw machines or automatic lathes. In recent years m of this work has been assumed by numerical control lathes or turning centers. For ample, the component (threaded shaft) shown in Figure 22-20 could have also b made on an NC turret lathe wlth some savings in cycle bme

Threaded shaft

Stock

dThread

see INSET

./

Is sre
I ceol

naive pard 19r 10-

IGURE 22-20 Turret-lathe )oling setup for producing part /lawn. Numbers in circles fdicatethe sequence of the perationsfrom 1 to 9. The pers A through F refer to the rrfaces being machined. lperation 3 is a combined peration. The roll turner is jrning surface F while tool 3 on jesquare post is turning lrface 8. The first operation lops the stock at the right pgth. The last operation cuts ~efinished bar off and puts a hamfer on the bar, which will gxt be advanced to the lock stop.
i

Roller turner has rolls to support the work against the cutting forces

:d;rrd
I.

: 1 m .
,

WORKHOLDING DEVICES FOR LATHES Five methods are commonly used for supporting workpieces in lathes:

1 . Held between centers

2. Held in a chuck 3. Held in a collet 4. Mounted on a faceplate 5. Mounted on the carriage

'~d by
ians :!a8is $cle
'I

p5 f

In the first four of these methods, the workpiece is rotated during machining. In the fifth method, which is not used extensively, the tool rotates while the workpiece is fed into the tool. A general discussion of workholding devices is found in Chapter 25, and the student involved in designing working devices should study the reference materials under Tool Design. For lathes, workholding is a matter of selecting from standard tooling.
LATHE CENTERS Workpieces that are relatively long wrth respect to thelr diameters are usually machined between centers. See Figure 22-21. Two lathe centers are used, one in the spindle hole and the other in the hole in the tailstock quill Two types are used, called dead and h e . Dead centers are solid, that is, made of hardened steel with a Morse taper on one end so that it will fit into the spindle hole.The other end is ground to a taper. Sometimes the tip of t h ~ taper s is made of tungsten carbide to provlde better wear resistance. Before a center

Solid or "dead" lathe center.

plate (Courtesy of South Bend Lathe.)

Live centers are shown in F~gure 22-22. A mechanical connection must be

Plaln solid mandrel

done in a dnll press or on the lathe with a tool dr~lled in each end.This IS typ~cally in the rear turret A comb~nation center drill and countersink ordinaslly is used,wth taken that the center hole is deep enough so that it will not he machined away m facing operation and yet is not drllled to the full depth of the tapered portion of center drill (see Chapter 23) Because the work and the center of the headstock end rotate together, no lu cant 1s needed in the center hole at this end. The center m the tailstock qu~ll does rotate; adequate lubr~catioti inust be prov1ded.A mixture of white lead and oil is o used.Failure to provide proper lubrication at all times w~ll result m scorlng of the w plece center hole and the center,and Inaccuracy and senous damage may occur. Live ters are often used m the tailstock to overcome these problems

of the workpiece w~ll reduce ihe clearances m the setup.

MANDRELS
Workpieces that must be machrned on both ends or are disk-shaped are often mo on rnnndreL~ for t u r m g between cenlers.Three common types of mandrels ares

between centers for lathe work

is pressed on the mandrel.The mandrel should be mounted between centers so th

SECTION 22.5 Workholding in Lathes cutting force tends to tighten the work on the mandrel taper. Solid mandrels permit the work to be machined on both ends as well as on the cylindrical surface. They are available in stock sizes but can be made to any desired size. Gnng (or disk) mnndrels are used for production work because the workpieces do not have to be pressed on and thus can be put in position and removed more rapidly. However, only the cylindrical surface of the workpiece can be machined when this type of mandrel is used. Cone mandrels have the advantage that they can be used to center workpieces having a range of hole sizes.

LATHE CHUCKS
Lathe chucks are used to support a wider variety of workpiece shapes and to permit more operations to be performed than can be accomplished when the work is held between centers.Two basic types of chucks are used (Figure 22-24). Three-jaw self-centering chuck.7 are used for work that has a round or hexagonal cross section.The three jaws are moved inward or outward simultaneously by the rotation of a spiral cam, which is operated by means of a special wrench through a bevel gear. If they are not abused, these chucks will,provide automatic centering to within about 0.001 in. However, they can be damaged through use and will then be considerably less accurate. Each jaw in a four-jaw independent chuck can be moved inward and outward independent of the others by means of a chuck wrench.Thus they can be used to support a wide variety of work shapes. A series of concentric circles engraved on the chuck face aid in adjusting the jaws to fit a given workpiece. Four-jaw chucks are heavier and more rugged than the three-jaw type, and because undue pressure on one jaw does not destroy the accuracy of the chuck, they should be used for all heavy work. The jaws on both three- and four-jaw chucks can be reversed to facilitate gripping either the inside or the outside of workpieces. Combination four-jaw chucks are available in which each jaw can be moved independently or can be moved simultaneously by means of a spiral cam.Two-jaw chucks are also available. For mass-production work, special chucks are often used in which the jaws are actuated by air or hydraulic pressure, permitting very rapid clamping of

\ , ~ e r n o v a b l e

jaws

4-law with sach jaw

independent

one of three

FIGURE 22-25 Hydraulically actuated through-hole three-jaw power chuck shown in sectlon view to left and in the spindle of the lathe above connected to the actuator.

chuckbody

Hous~ng

\
Top jaw

the work. See Figure 22-25 for a schematic.The rapid exchange of tooling is a ke] manufacturing strategy in manufacturing cells. Chuck jaw sets are dedicated and cus tomized for specific parts.7he lirst time a chuck jaw set is used, each jaw is marked w i t the number of the jaw slot where it was installed and an index mark that corresponds with the alignment of the iaw serrations and the first tooth on the chuck master iaw The jaws can now be reinstalled on the chuck exactly where they were bor adjustability of the chuck body lets the operator dial in part concentrically resetting the jaw.

COLLETS
Collets are used to hold smooth cold-rolled bar stock or machined workpieces more ac' curately than with regular chucks. As shown in Figure 22-26, collets are relatively thit tubular steel bushings that are split into three longitudinal segments over about two thirds of their length. At the split end, the smooth internal surface is shaped to fit tht piece of stock that is to be held.The external surlace of the collet is a taper that mate! with an internal taper of a collet sleeve that fits into the lathe spindle. When the collel is pulled inward into the spindle (by means of the draw bar), the action of the twomat. ing tapers squeezes the collet segments together, causing them to grip the worl Collets are made to fit a variety of symmetrical shapes. If the stock su smooth and accurate, good collets willprovide very accurate centering, with runout lesr than 0.0005 in. However, the work should be no more than 0.002 in. larger or 0.005 in, smaller than the nominal size of the collet. Conseuuentlv. collets are used onlv on drill-rad, cold-rolled, extruded, or previously machined stock. Collets that can open automatically and feed bar stock forward to a stop nism are commonly used on automatic lathes and turret lathes.An example of a collei chuck is shown in Figure 22-27. Another type of collet similar to a Jacobs drill chucj has a greater size range than ordinary collets; therelore, fewer are required. I
9.

SEC~O 22.5 N

Workholding in Lathes

623

FIGURE 22-27 (a) Method of using a draw-in collet in lathe spindle. (Courtesy of South Bend Lathe.) (b) Schematic of a collet chuck in which the clamping force can be adjusted.

624

CHAmEn 22

Turn~ng and Bonny P:ocesses

FIGURE 22-28 Cuttlng a thread on a long, slender workpiece, uslng a follow rest (left) and a steady rest (r~yht) on an enylne lathe Note the use of a dog and face plate to dr~ve the workp~ece.[Courtesy of South Bend Lathe )

FACEPLATES
Faceplates are used to support irregularly shaped work that cannot be gripped easi chucks or collets.The work can he bolted or clamped directly on the faceplate ore2 supported on an auxiliary fixture that is attached to the faceplate.Thelatter proce~ is time saving when identical pieces are to he machined.

MOUNTING WORK O N THE CARRIAGE When no other means is available, boring occasionally is done on a latlie 6jiBiothe work on the carriage, with the boring bar mounted between centers and drivf means of a dog.

1
I
:',

STEADY AND FOLLOW RESTS If one attempts to turn a long, slender piece between centers, the radial force ed@ by the cutting tool or the weight of the workpiece itself may cause it to be de out of line. Steady rests and follow rests (Figure 22-28) provide a means for such work between the headstock and the tailstock. The steady rest is clam lathe ways and has three movable fingers that are adjusted to contact the wo align it. A light cut should be taken belore adjusting the fingers to provide a s contact-surface area. A steady rest also can be used in place of the tailstock as a means of sup the end of long pieces, pieces having too large an internal hole to permit usin lar dead center, or work where the end must be open for boring. In such cases slock end of the work must be held in a chuck to prevent longitudinal moveme feed should be toward the headslock. The follow rest is bolted to the lathe carriage. It has two contact fingers adjustcd to bear against the workpiece, opposite the cutting tool, in order to the work from being deflected away from the cutting tool by the cuttingf~ces.

III Key Words

apron dutomatlc lathe bed hor~ng carrlage chuck6 collct5 cutoff cuttmng tools depth of cut dog drilling engine lathc faceplates facing feed follow rest headstock knurlii~g lathe centers mandrels metal removal rate milling patting quill reaming screw machine steady rest tailstock taper turning turning turret lathe workholdine;

I I Review Questions
How is the tool-work relationship in turning different from that in facing? What different kinds of surfaces can be produced by tuming versus facing? How does form turning differ from ordinary tuming? What is the basic difference between facing and a cutoff operation? Whichmachining operations shown in Figure 22-2 do not form a chip? Why isit difficult to make heavy cuts if a form turning tool is complex in shape? ' Show how equation 22-5 is an approximate equation. Why is the spindle of the lathe hollow? What function does a lathe carriage have? Why is feed specified for a boring operation typically less than that specified for turning if the MRR equations are thesame? What function is provided by the leadscrew on a lathe that is not provided by the feed rod? I How can work be held and supported in a lathe? How is a workpiece that ismounted between centers on a lathe driven (rotated)? What will happen to the workpiece when turned, if held between centers, and the centers are not exactly in line? Why is it not advisable to hold hot-roUed steelstock in a collet? How does a steady rest differ from a follow rest? What are the advantages and disadvantages of a four-jaw independent chuck versus a three-jaw chuck? Why should the distance a lathe tool projects from the tool holder be minimized? What is the difference between a ram- and a saddle-turret lathe? How can a tapered part be turned on a lathe? Why might it be desirable to use a heavy depth 01 cut and a light feed at a given speed in turning rather than the opposite? 22. If the rpm for a facing cut (assuming given work and tool materials) is being held constant, what is happening during the cut to the speed? To the feed? 23. Why is it usually necessary to take relatively light feeds and depths of cut when boring on a lathe? 24. How does the corner radius of the tool influence the surface roughness? 25. What effect does a BUE have on the diameter of the workpiece in turning? 26. How does the multiple-spindle screw machine differ from the single-spindle machine? 2 7 . Why does boring ensure concentricity between the hole axis and the axis of rotation of the workpiece (for boring tool), whereas drilling does not? 28. Why arc vertical spindle machines better suited for machining large workpieces than horizontal lathes? 29. What is the principal advantage of a horizontal boring machine over a vertical boring machine for large workpieces? 30. In which figures in this chapter is a workpiecc bung held in a three-jaw chuck? 31. How is the workpiece in Figure 22-14 being held? 32. In which figures in this chapter is a dead center shown? 33. In which figures in this chapler is a live center shown? 34. In which figures in this chapter showing setups do you find the following being used as a workholdmg device? a. Three-jaw chuck b. Collet c. Faceplate d. Four-jaw chuck 35. How many form tools are being utilized in the process shown i n Figure 22-15 to machine the part? 36. From the information given in Rgure22-20, start with a piece of round bar stock and show how it progresses, operation by operation, into a finished part-a threaded shaft.

I I Pro
A cutting speed of 100 sfpm has been selected for a turning cut. At what rpm should a 3-in.-diameter bar be rotated? Assume that the workpiece in Problem I is 8 in. (203.2 mm) long and a feed of 0.020 in. (0.51 mm) per revolution is used. What is the machining time for a cut across its entire length? Don't forget to add an allowance. If the depth of cut in Problem 2 is 0.25 in., what is the metal removal rate (MRR) exactly? What is the MRR approximately? Using the same recommended cutting speeds and feeds, calculate the machining time to cut off the bar in Problem 2. The following data apply for machining a part on a turretlathe and on an engine lathe: Engine Lathe
Times,in minutes to machine part Cost of special tooling T u n e to set up the machine Labor rates

' h r e t Lathe
5 min

30 min
0
30 min

$300
3hr $8hr

Machine rates (overhead)

$S/hr $10/hr

$12/hr

a. How many pieces would have to be made for the cost of the engine lathe to just equal the cost of the turret lathe? This is the BEQ. b. What is the w s t per unit at the BEQ? 6 . A finish cut for a length of 10 in. on a diameter of 3 in, is to be taken in 1020 steel with a speed of 100 fpm and a feed of 0.005 ipr.What is the machining time? 7. A workpicce 10 in. in diameter is to be faced down to a diameter of 2 in. on the right end. The CNC lathe (see Chapter 26) controls the spindle speed and maintains the cutting speed at 100 fpm throughout the cut by changing the rpm. What should be the time for the cut? Now suppose the spindle rpm for the workpiece is set to give a speed of 100 fpm for the 10-in. diameter and is not changed during the cut. What is the machining time for the cut now? The feed rate is 0.005 ipr. 8. A hole 89 mm in diameter is to he drilled and bored through a piece of 1340 steel that is 200 mm long, using a horizontal boring, drilling, and milling machine. High-speed tools will be used.The sequence of operations will be center drilling: drilling with an 18-mm drill followed by a 76-mm d~ill; then boring to size in one cut, using a feed of 0.50 mdrev. Drilling feeds will be 0.25 mmlrev for the smaller drill and 0.64 m d r e v for the

set or change any given tool and set the proper machine speed and feed requires 1 min. Select the initial cutting speeds, and compute the total time required for doing the job. (Neglect setup time for the fixture.) This is often referred to as the run time or the cycle time. (Hini: Check in Chapter 21 for recommended speeds for turning.) 9. Figure 22-A shows the fixed and variable costs for a part being produced on an engine lathe. Figure 22-B has three plots of unit production cost ($/unit) versus production volume (Q= huild quantity). (Note that this plot is made on log-log paper.) Cost per unit for a particular process decreases withincreased volume as fixed costs are spread out over more units. For a particular process there is no minimum cost but rarher production volumes within which particular processes are most economical.

divided by quantity, which means each is the sum I fixed cost per unit (mainly setup and tooling costs) ar variable costs per unit (direct labor, direct material). the data on the plots, estimate the fixed costs for tt gine lathe, the NC lathe, and the singlc-spindle antor b. For what build quantities is the NC lathe most econc (approximately)? c. What cost per unit does the NC lathe approach as the quantity becomes very large? d. Explain how to go from a cost vs, quantity plot to a co! vs. quantity plot? e. What happens to these plots if you plot them on re Cartesian coordinates? Try it and comment on you find. f. Many Japanese manufacturers have found innovative to eliminate setup time in many of their processes. l + the impacl of this on these kinds of plots, on cost pe economics, and on job shop inventories? 10. The derivation of the approximate equation 22-5 for the for turning process requires an assumption rcgardin diameters of the parts being turned. Determine the er the equation for Problem 3. What is the assumption?

Production quantity

FIGURE 22-A

Dark llne shows lowest unlt cost over all quantltles

c
2

a
4 Vi -

0 U

10

I
100

I
1,000
Q

automat1

10

10,000

Small

Production quantity

Degree of automatton
FIGURE 22-8

Large

s the plant manufacturing engineer at ERG, inc , Jay Strom has been called Into the productiondepartment at the tlme is a I-inch micrameter. Katrin Zachary, the production manager, would like to

= =

60 sfpm where7 = 100 mlnutes and another at 85 sfpm, where T = 10 minutes

'

drameter of 6 in.The onginal dtarneter of 10 in. to a f~nal forging has a BHN of 300 to 400.The turning is to be performad on a heavy-duty lathe, which is equipped wtth a50 HP motor and a conttnuousiy varrable speed dnve on the spindle The work will be held between centers, and the

3. Jay dec~des to make two test cuts at the standarcl feed of 0.020 tpr. He assumes that the chip thickness ratio varies almostlinearly between the speeds of 20 and 80 fpm, the values betng 0.4 at the speed of 20 fpm and 0.6 at 80 fpm. The chip thickness values were determ~ned by micrometer measurement in orderto determlne the value d rc. 4. The mach~ned forging (log) will be used as a roller in a newspaperpress and must be prec~sely machlned If the log deflects dur~ng the cuttlng more than 0.005 in., the roll wll end up barrel-shaped after f~nal grind~ng and

manufacturehave provlded the following mformatton:

and that F , causes thedeflection.

DRILLING AND KELATED

23.1 INTRODUCTION OFTHE 23.2 FUNDAMENTALS DRILLING PROCESS 23.3 TYPES Or DRILLS Depth-to-Diameter Ratio M~crodrilhng 23.4 TOOL HOLDERS FOR DRILLS 23.5 W O R K H ~ ~ DRILLING ~ G FOR 23.6 MACHINE TOOLS
FOR DRILLINO

23.8 COUNTERBORING, COUNTERSINKING, SPOT FACING Reaming Practice Case Study: BOLT-DOWN LEG ON A CASTING

23.7 C ~ NFLWI G DRILLING

I In manufacturing it is probable that more holes are produced than any other shape, a large proportion of these are made by drall~ngOf all the n~ach~ning processes perfom drilling makes up about 25% Consequently, drilling is a very important process Altho drilling appears to be arelatively simple process,lt a really a complex process Most dril is done wlth a tool having two cuttlng edges, or lips, as shown in Figure 23-1.Thls is a t drill, the most common drill geometry.The cutt~ng edges are at the end of a relat~vely f ible tool. Cutting action takes place Inside the workpiece.The only exit for the chipsis hole that is mostly f~lled by the dr~llFriction between the margln and the hole wall] duces heat that 1s addllional to that due to c h p formation.Tl~e counterflow of thecl ~nthe flutes makes lubrication and cooling difficult. There are four major actlons tal placc at the point of a drill.
1. A small hole is formed by the web--chips are notcut here in the normal sense.. 2. Chips are formed by the rotatlug lips.

clearance

I /\

th~ckness

FIGURE 23-1 Nomenclature and

geometry of conventtonal tw~st dr~ll. Shank style depends upon the method used to hold the dtlll. Tangs or notches prevent slippage: shank w~th tang, fa) stra~ght (b) tapered shank wlth tang, (c) stra~ght shank wlth whistle notch, (d) straight shank wlth flat notch

Land

outer corner
Outer

4
Nominal rdef

corner

Ch~sel edge
corner

angle

3. Chips are removed from the hole by the screw action of the helical flutes. 4. The drill is guided by lands or margins that rub against the walls of the hole.

In recent years, new drill-point geometries and TiN coatings have resulted in improved hole accuracy, longer life, self-centering action, and increased feed-rate capabilities. However, the great majority of drills manufactured are twist drills. One estimate has U.S. manufacturing companies consuming 250 million twist drills per year. When high-speed steel (HSS) drills wear out, the drill can be reground lo restore its original geometry. If regrinding is not done properly, the original drill geometry may be lost, and so will drill accuracy and precision. Drill performance also depends on the drilling machine tool, the workholding device, the drill holder, and the surface of the workpiece. Poor surface conditions (sand pockets andior chilled hard spots on castings, or hard oxide scale on hot-rolled metal) can accelerate early tool failure and degrade the hole-drilling process.

The process of drilling creates two chips.A conventional two-flute drill. with drill of diameter D, has two principal cutting edges rotating at an rpm rate of N and feeding axially.The rpm of the drill is established by the selected cutting velocity or cutling speed

with Vin surface feet per minute and D in inches (mm).This equation assumes that Vis the cutting speed at the outer corner of the cutting lip (point X in Figure 23-2).

- --- -- -..-.-.------

Outer corner force

N=

Rake angle

BURL23-2 Conventional

I geometry v~ewed from the nt showlng how the rake jle varles from the chlsel edge he outer corner along the Itp !thrust force Increases as the b a approached.

630

CHAPTER 23

Dr~ll~ng and Related Hole-Making Processes

The feed, f,, is given in inches per revolution. The depth of cut in driningis to half the feed rate, or t = f,/2 (see section A-A in Figure 23-2). The feed rate in per minute,fm, is The length of cut in drilling equals the depth of the hole, L, an allowance for approach and for the tip of dril1,usually A = D/2. In drilling, thes and feed depend upon the material being machined, the cutting tool material, ang' size of drill.Table 23-1 gives some typical values for V and for carbide indexablei' drills, a type of drill shown later in the chapter. After selecting the cutting speed and feed for drilling the hole, the rpm value spindle of the machine is determined from equation 23.1, the maximum velocity ring at the extreme ends of the drill Lips.The velocity is very small near thecenter chisel end of the drill. For drilling, cutting time is

fa,.

1 9
1
"i

Matertal Group
Cast iron. modular, duct~le, or malleable

S~ze Range

Cutting Speed

sfpm

peed& (m.14

Low-carbon unallayed cae-hardcmng steels

Hlgh-carbon alloyed heat-treated steels

Hlgh-tcnslle steels

Stainless ?tee15

ntaniurn Steels

A cast iron plate is 2 in. thick and nerds 4-&diameter holes drilled in it. An indexable-insert drill has been selecled. Looking atTable23-1, we select a cutting speed of 200 Cpm and a feed of 0.005 ipr. nie spindle rpm = 12VhD = 12 X 200 / 3.14 X 1= 764rpm. What if the machine does not have this specific rpm? Pick the closest value: Let's say it is 750 rpm. The penetration rate or feed rate (in.lmin) = feed (ipr). The rpn = 0.005 X 750 = 3.75 in./min.The maximum chip load = Feed (ipr)/Z = 0,00512 = 0.002iin./rev. What if themachine does not have thespecific feed rate? Pick the next lowest value as a startingvalue. Let's say it is 3.5 in./min. The material removal rate (in?/min) = (1114) X (0)' x feed (ipr) X rpm = (nl4) 1' X feed rate = 3.1414 X I' X 3.50 = 2.75 in?/min. n i e MRR can be used with the uilit power for cast iron (see Chapter 20) to estimate the HP needed to drill the hole. Let HP, = 0.33 for this CI:

HP = HP,

MRR

0.33 X 2.75 = .90

This value would typically represent 80% of the total motor horsepower (HP) needcd,so in this case, a horsepower motor greater than 1.5 or 2 would be sufficient. In estimating the cost of a job,it is often necessary to determine the tims to drill a hole. drill tunelhole = length drilled + allowance feed rate (I" /min) traverse length of wilhdrawal + rap~d rapid traverse rate

+ prqrated downtlme to change drllls per hole

The 1 s t term prorated downt~mc is

drlll change downtune holes drilled per drlll (tool llfe) And the costihole 1s

dnlhng tlmelholes X (labor

+ rnaclune rate) + prorated cost of dnll/hole

The metal removal rate is

MRR

=-

volume Tm .irDZLi4 -(omitting allowances) L/$NS

which reduces to

MRR

(vD2/4)f,N, in.'

Substituting for N with equation 23-1, we obtain an approximate form

MRR

-- 3DVf,

The most common types of drills are twist drills.These have three basic parts: the body, the point, and the shank, shown in Figures 23-1 and 23-2.The body contains two or more spiral or helical grooves, calledflutes, separated by lands.To reduce the friction between the drill and the hole, each land is reduced in diameter except at the leading edge, leaving a narrow margin of full diameter to aid in supporting and guiding the drill and thus aiding in obtaining an accurate hole. The lands terminate in the point, with the leading edge of each land forming a cutting edge. The flutes serve as channels through which the chips are withdrawn from the hole and coolant gets to the cutting edges. Although most drills have two flutes, some, as shown in Figure 23-3, have three, and some have only one.

632

CHAMLR 23

Dr~lling and Related Hole-Mak~ng Processes

FIGURE 23-3 Types of twist drills and shanks. Bottom to top: Straight-shank, three-flute core drill; straight-shank; taper-shank; bit-shank; straight-shank, high-helix angle; straight-shank, straight-flute; taper-shank, subland drill.

~p

The principal rake angles behind the cutting edges are formed by the relati the flute helix angle to the work.This means that the rake angle of a drill vari the cutting edges (or lips), being negative close to the point and equal to the helix out at the lip.Because the helix angle is built into the twist drill, the primary rake cannot b e changed by normal grinding.The helix angle of most drills is 24", withlarger helix angles-often more than 30"-are used for materials that can be vely rapidly, resulting in a large volutne of chips. Helix angles ranging from Oo to used for soft materials, such as plastics and copper. Straight-flute drills (zero rake angles) are also used for drilling thin sheets of soft materials. It is possib change the rake angle adjacent to the cutting edge by a special grinding proce called dubbing. The cone-shaped point on a drill contains the cutting edges and the various c ance angles.This cone angle affects the direction of flow of the chips across th and into the flute. The 118" cone angle that is used most often has been fou vide good cutting conditions and reasonable tool life when drilling mild steel, thu ing it suitable for much general-purpose drilling. Smaller cone angles-from 118"-are sometimes used for drilling more brittle materials,such as gray ca magnesium alloys. Cone angles from 118"to 135" are often used for the more duc terials, such as aluminum alloys. Cone angles less than 90" frequently are used for plastics. Many methods of grinding drills have been developed that produc gles other than 118'. The drill produces a thrust force T and a torque M. Drill torque increases feed (in./rev) and drill diameter, while the thrust force is influenced greatly by the or chisel end design, as shown in Figures 23-2 and 23-4.

The relatively thin web between the flutes forms a metal column or backbone. If aplain conical point is ground on the drill, the intersection of the web and the cone produces a straight-limechisel end, which can be seen in the end view of Figure 23-2.The chisel point, which also must act as a cutting edge, forms a 56" negative rake angle with the conical surface. Such a large negative rake angle does not cut efficiently,causing excessive deformation of the metal.This results in high thrust forces and excessive heat being developed at the point. In addition, the cutting speed at the drill center is low, approaching zero. As a consequence, drill failure on a standard drill occurs both at the center, where the cutting speed is lowest, and at the outer tips of the cutting edges, where the speed is highest. When the rotating, straight-line chisel point comes in contact with the workpiece, it has a tendency to slide or "walk" along the surface, thus moving the drill away from the desired location. The conventional point drill, when used on machining centers or high-speed automatics, will require additional supporting operations like center drilling, burr removal, and tool change, all of which increase total production time and reduce productivity. Many special methods of grinding drill points have been developed to eliminate or minimize the difficulties caused by the chisel point and to obtain better cutting action and tool life (see Figure 23-4 for some examples). The center core or slot-point drill shown in Figure 23-5 has twin carbide tips brazed on a steel shank and a hole (or slot) in the center.The work material in the slot is not machined but, rather, fractured away. The center core drill has a self-centering action and greatly relieves the thrust force produced by the chisel edge of conventional twist drills. This drill operates at about 30 to 50% less thrust than that of

Helical IS-shape chisel point)

Can eliminate center drilling on NC machining centers

Exeellenf hole geometry Close relationship between drill size and hole sire Increased too1 life Lower thrust requirements Leaven burr on breakthrough

Secondary angle 30'4O'Itrue! Prlmary a n g i e 4 n 8 L l t r u e )

Rellsved helcal Reduces thrust force EllmlnateSchlSel end ~ q u a l rake . angle

Four-facet ~ o o selfcemering d abiliw Bresks up chipsfor deep-hole drilling Can be genersred in a single grinding operation: reduces thrust. Eliminate8 ceniei drilling in NC

Bl~hfOrd Combenallon of helcal and Racon p~intfeafuies Self centering and reduced burrs Excellent hale geomeny Increased tool llfe

Racon iradlused convsntlonai polnf! increased feed rates Increased tool lhfe 1&10 times in C 1 ! Reduced burrs at breakthrough Not self centerrng

634

CHAPTER 23

Dr~ll~ng and Related Hole-Mak~ng Processes

Nancutflng zone

C o n ~ e n l ~ o n dr8Il a l wfh largefhrurt force st

web

@k
Thrunforce
~hrustforce

center core drli or

stat palm drill ~ 8 t h gready reduced thrust Center core removed by d~m8lefra~ture itensson/

Center core dnlls can greatly reduce the thrust force.

FIGURE 23-5

conventional drills All rake angles of the cutting edge are positive, which fur reduces the cutting force. The conventional point also has a tendency to produce a burr on the exlt si a hole Some type of chrp breaker is often incorporated Into drllls One procedur grind a small groove in the rake face, parallel wlth and a short distance back fro cuttrng edge. DnIls wrth a special chip-breaker rib as an ~utegral part of the flu aviulable The rib interrupts the flow of the chip, causing it to break into short len The spht-point drill is a form of web thinning to shorten the chisel edge.This reduces thrust and allows for higher feed rates. Web thinning uses a narrow grtn wheel toremovea portion of the web near the point of the drill. Such methods have varying degrees of success, and they require special drill-grinding eqwpment Also shown in Figure 23-4 is a four-facet self-centenng point that works w tougher materials.The facets refer to the number of edges on the clearance susfac posed to the cuttlng action. The self-centering drill lasts longer and saves ma time on numerical control (NC) centers as they can ehminate the need for center A common aspect in drill-point term~nology 1s total Indicator runout (TIR). measure of the cutting lips' relative side-to-side accuracy The orig~nal drill poln duced by the manufacturer lasts only until the first regrind, thereafter performanc life depend upon the quality of regrind. Proper regrinding (recondit~oning) of a a complex and important operation. If satisfactory cutting and hole size are t achieved, it 1s essentral that the polnt angle, lip clearance,lip length, and web be correct As illustrated in Figure 23-6, lneorrect sharpening often results in unbal cutting forces at the tipA causing misalignment and oversized holcs. Drills, cvcn drills, should always be machine ground, never hand ground. Drill grinders, often puter controlled, should be used to ensure exact reproductron of the geometry lished by the manufacturer of the drlll.This is extremely important when drllls are on mass-production or numerically controlled machmes. Companies Invest huge in NC machming centers but overlook the value of a top-quality drill-gnndingmac Drrll shanks are made in several types.% two most common types are the str and the taper. Straight-shank drills are usually used for sizes up to :-in. diameter and be held in some type of drill chuck. Taper shanks are available on larger dnUs and are mon on dnlls above 1in. Morse tapers are used on taper-shank drills, ranging fr number 1taper to a number 6. Taper-shank drills are held in a female taper in the end of the machlne tool If the taper on the drill is different from the spindle taper, adapter sleeves are a The taper assures the drill's being accurately centered in the spmdle.The rang at t

SECTION23.3

Types of Drills

635
.

FIGURE 23-6 Typical causes of

drilling problems.

Outer corners break down C .I:II~!I SJC:I:J :no y' r I , % # ,~ot'. r g . mp: . . r J r t i l r ~ o n I m , 3 : . . 1 a. n t r .tar c . ggeu :IS chips

Cutting lips chip:Too much feed: lip relieftoo great

r cracks in cutting lips: Overheated or too quickly coaled Checks o while sharpening or drilling
Chipped margin: Oversize jig bushing Drill breaks: Paint improperly ground; feed too heavy; spring or backlash in drill press, fixture, or wore drill is dull; flute$ clogged with chips
Tang break$: Imperfect fit between taper shank and socket caused b v d i r i or chipso? by burred or badly worn sockets

Drill breaks when drilling brass or wood: Wrong type drill;flutes closged with chips Drillapiks up center: Lip relief too small: too much feed Drill will not enter work: Drill is dull; web too heavy; lip relief too small

H d e rough: Paint improperly graund or duii; no cutting compounds at drill point; improper cutting compound; feed too great: fixture not rigid . Hole oversize: Unequal angle o f i h e cutting edges; unequal length of the cutting edges; see part ( a )
Chip shape changer while drilling: Dull drill or cutting lips chipped

(a) A n g l e u n e q u a l

(b) L e n g t h u n e q u a l

Large chip ooming from one flute, small chip from t h e other: Point improperly ground, one lip doing ali the cutting

of the taper shank fits loosely in a slot at the end of the tapered hole in the spindle.The drill may be loosened for removal by driving a metal wedge, called a drip, through a hole in the side of the spindle and against the end of the tang.It also acts as asafety device to prevent the drill from rotating in the spindle hole under heavy loads. However,if the tapers on the drill and in the spindle are proper, no slipping should occur.The driving force to the drill is carried by the friction between the two tapered members. Standard drills are available in four size series, the size indicating the diameter of the drill body:

Millimrterserla: 0.01- to 0.50-mmincrements,accordingto size,in diameters from 0.015 mm Numerical series: no. 80 to no. 1 (0.0135 to 0.228 in.) Lettered serie.s:A to Z (0.234 to 0.413 in.) Factionalseries: to 4in. (and over) by 64th~.

TiN coating of conventional drills greatly improves drilling performance. The increase in tool life ofTiN-coated drills over uncoated drills in machining steel is more than 200 to 1000%.

DEPTH-TO-DIAMETER RATIO
The depth of the hole to be drilled divided by the diameter of the drill is the depth-todiameter ratio. Most machinists consider a ratio of 3 to 1 to be deep-hole drilling, after which hole accuracy (location) drilling speed and tool life will he reduced.The bores of rifle barrels were once drilled using conventional drills. Today, deep-hole drills, or gundrills, are used when deep holes are to be drilled. The oldest of these deep-hole techniques is gundrilling.The original gundrills were very likely half-round drills, drilled axially with a coolant hole to deliver cutting fluids to the cutting edge (see Figure 23-7). Modern gundrills typically consist of an alloy-steeltubing shank with a solid carbide or carbide-edged tip brazed or mechanically fixed to it. Guide pads following the cutting edge by about 90' to 180" are also standard. The gundrill is a single-lipped tool, and its major feature is the delivery of coolant through the tool at extremely high pressures-typically from 300 psi to 1800 psi, depending on diameter-to force chips back down the flute. Successful application of a gundrill depends almost entirely on the formation of small chips that can be effectively evacuated by the flow of cuttingfluids. Standard gundrills are made in diameters from 0.0078 in. (2 mm) to 2 in. or more. Depth-to-diameter ratios of 100 to 1 or more are possible.

'ER 23

Processes Drilling and Related Hole-Mak~ng

Gundrill

Secondary rel~ef

FIGURE 23-7 The gundrill from geometry is very d~fferent that of convent~onal drills

In gundrilling tolerances for diameters of drilled holes under 1in. c O.0005-in. total tolerance, andshould not exceed 0.001 in-overall.According toone so 'koundness accuracies of 0.00098inch can be atta~ned." Because of the bu of the guide pads, excellent surface finishes can he produced. Hole straightness is uniformity of workpiece lnaterial,ulndifion of the machine, sharpness feeds and speeds used, and the specific techniqueused (rotahon of the t or both), but deviation should not exceed about 0.002 in.TIR in a 4-in, depth diameter, and it can be held to 0.002 in. per h o t . Bas~csetup tor a gundrilling operation, which is generally horizontal, requir bushmg very close to the work entry surface a d may involve rotating the wo tool, or both, Best concentricity and straightness are achieved by thewark and t rotating in opposite directions. Other deep-hole driIls are called BTA (BoringTrepanning Association) ddls qector drills. A deep hole is one in which the length (or depth) of the mare times the hameter, Coolants can be fed internally through these driI ting edges. See Figure 23-8 for schematic of an ejector drill and the machi for gundriUing.The coolants flush the chips out the lutesThe special design reduces the tendency of the drill to drift, thus producing amore accurately The typical ETA deep-hole drilling tools are designed for single-lip end in a single pass. Solid-deep-hole drills have alloy-steel shanks with a carb that is fixed t o i t mechanically.The cutting edge cuts through the centef on one the hole, leaving no area of material to be extrnded.The cuttmg is done by the ou inner cutfing angles, which meet at a point Theoreticall3 the depth of the hole limlt, but practically, it is restricted by the torsional rigldlty of the shank. Gundnlls have a single-lip cutting action. Bearing areas and lifting forms ated by the coolant pressure counteract the radial and tangential loads. The consrruction forces the edge to cut in a true circular pattern.The tip Thus follow direction of its own axts.The trepatlnitzggundrill leaves a solid core.

Supporting pad

Radial cutting clearance

Circle land or margin

Primary cutting edge or outer edge r pad recess or lag

Counterboring cutting tool

(bl

(ct

URE 23-8 BTA drills for (a) boring, (b) trepanning, (c) counterboring, (d) deep-hole drilling with ejector drill, (e) hoi phole-drilling machine. (5. Azadand 5. Chandeoshekar, Mechanical Engineering, Sept 1985, pp. 62, 63.)

1s 01 se alemme s!ley1 aloq 1OL-EL3an1)1i u!elqo0

.s~os aloy

pallea ualjo are sialln3 a[oH

a1 MES ~ V M ralp3 q)oo)-ald!~[nu~ 'pa[[e~-u!ql aql L q pasnpold sr aloq upur ayl Lq q~'(6-22 a~n%!a) iann3 aloy E q l ! apeu ~ aq L-em 1epaleuI u!yl u! sa~oy ~a%re7

spaaj 'sa%pa Bu!unz

$0 S S ~ U ~ J B 'au!q3eur ~ S

aqljo uoq!puo3 'le!~aleuca 3 a : d y ~ o ~

o L~!m~oj!un'q~dap'~alaure!p se q3ns salqe!len L q papage SF ssauiy%~sris aIoH

S E ~ T I23.3 ~N

Types

of Drills

639

diameters

countersink countersink and counterbore

counterbore

chamfer

countersink. and counterbore

matic
s.

vhereteeth.

large in diameter, or if it is sufficiently short, satisfactory accuracy may often be obtained without center drilling. Special drill holders are available that permit drills to be held with only a very short length protruding. Because of its flexibility and endpoint geometly, a drill may start or drit off centerline during drilling. The use of a center (start) drill will help to ensure that a drill will start drilling at the desired location. Nonhomogeneities in the workpiece and imperfect drill geometries may also cause the hole to be oversize or off-line.For accuracy, it is necessary to follow center drilling and drilling by boring and reaming.Boring corrects the hole alignment, and reaming brings the hole to accurate size and improves the surface finish. Special combination drills can drill two or more diameters, or drill and countersink andlor counterbore, in a single operation (Figure 23-11). Countersinking and counterboring usually follow drilling.These operations are described in more detail later in this chapter. A step drill has a single set of flutes and is ground to two or more diameters. Subland drills have a separate set of flutes on a single body for each diameter or operation; they provide better chip flow, and the cutting edges can be ground to give proper cutting conditions for each operation. Combination drills are expensive and may be difficult to regrind but can be economical for production-type operations if they reduce work handling, setups, or separate machines and operations. Spade drills (Figure 23-12) are widely used for making holes 1 in. or larger in diameter at low speeds or with high feeds (Table 23-3). The workpicce usually has an

Regular spade drill

Spade drill with oil holes

6$0

Cmmw 23 Drilling and Related Hole-Making Processes

Matenal
Mild macbnsq' steel 0 2 aad 03whr1 S~el,annealet10.4 tu0.%artWn Tool stad, l.Zra&m Steel f~rglnn Altny stezl StarnlesssteeLfre~rnachintnn %aialess %@el, hard
Cast irnn,soCf Cat iron, medium hnd Cast irw,herd chi1IWi

SurTaffiSpeed (ft per min)

fia-llD

%8t
45-60
3.5-SO

45-70
5C-70

2540 & ! . S O
SlOn
23-40 7%9U

Mall&la tmn Brass anfl branre, ordinary Brodte,hightensile

Monai msaf Alumaum and its allays hkgms~lltn andats alloys

Zoo-300 70-150 35-50 2QLL30D

2504

Feed Rater for Spade Drilling (ihchm per revolution)

Malleable Iron Drill Size
Brass

& ~ Iron t

Medium Steel stiud&s &el Mand Metal

existing hole, but a spade drill can drill deep holes in solids or staoke drills are la$ expensive becwse the long supporting bar can be made The drillpoint ean be ground witha minimum chisel p&The main mme rigid &use no flutes are required,and it Gau have a ca~tml h a fluid can be eirculawd cco aid in conling and in chip removal. ' b e cut ier to sharpen; only fh& blades a1:ed tta be TiN-coated. Spade drill8 are often used to machine a shallow locating GO smal1e1 drill and at the same tIme to provide a &mall bwel aroma later 'tapping or assembly ope~ations. Such a bevel also f for debwing. This p m c t b is particulwly useful on mas ~onholled machines, Carbide-tipped drills and drills with indexable inserts are Figwe 23-11) wrth one- and two-piece inserts for drilling shallow pieces indaabie h e r <driltn ertn produce a hole fouf times fast because they rm at h i &pw&/10wfeeds and are really more of a b a (IriUing procesa. However, to use indexabIe drill$ you must havean exkernel setup,adequatahorepcmier,adlots of cuCtingfl eratinghole tolerancesof and surface finish of2 the inboardinseft t a wt past the centerline of dially below the cenaer SeeTable 2.34 b r a n indexable driIling

dexable insert drills are ed for holes over 1 inch

Improper seafing of tool in tool holder, spindle, or turret Deflection because of too much overhand and lack of rigidity Improper seating of inscrts in pockets

Damagcd insert screws Improper speeds and feeds Insufficient coolant supply Improper carbide grade in inboard station Off-centor drill

Check parting line betwcen tool shank ilnd socket with feeler gagc. Check tosee if tool is locked lightly. With indicator, check if tool can be movcd by hand. Check if tool can he held shorter. Clean packets whenever indrxing or changing inserts. Check pockets for nicks and burrs. Check if inserts rest co~npletely on pocket hottoms. Check head and t h r e ~ d for nicks and burns. D o not overtorque screws. Check recommended guidelines for given materials. Check coolant flow. Recommend straight grade for mullipie-insert drills. Maintain proper alignment and conecnlriciq. Chcck bottomof hoir ur disk far center stub. Check setup rigidity. Check speed and feed guidelines. Chcek setup and part rigidily. Check seat in spindle vr tool holder. Check Speeds and feeds. Increase coolant pressure and flow. Is coolant flow constant? Make sure coolant reachcs inserts a1 all times Increasc coolant flow. Add codant grooves Mostly speedor feed. Mostly feed rate. Lower feed or increase speed. Increase Feed rate or decrease specd. Use dimple inserts.

moving 04 back stroke: drill body rubbing hole wall: over- or undersize holes or hole surface finish

Deflection \ribrations

lnsuffiden t cooiant pressure and volume

ry short,thick, flat chips

ng and stringy chips

Recutting chips, causing drill to jump P w r chip control; chips trapped in hole Chatter Feed rate too high in relation to cutting speed Feed rate too low in relation to cutting

izerl threads, caused by coolant or

onstant chippingacrcurr especially on an inner inserf and conditions are optimum.try an uncoated-carbide insert or a grade h.ith higherlransveisz rapture strength.

Processes Drilling and Related Hole-Mak~ng

-

& ' @
,

fi

6S-Q
I

#
1

I I 2 3 4 Hole depth to drameter (40)

FIGURE 23-14

Dnll select~on depends on hole dtameter and hole depth.

A high-pressure,pulsating coolant system can generate pressures up to 300 p works well with indexable drilling. It can have disadvantages, however. High pre w~th pulsatmg action can decrease chip control and cause drill deflection. A high-pr coolant stream can flatten chips at the pomt of forming and forces them Into th causlng recuttmg, insert chipping, and poor hole finishes.7lle pressure can force between the drill body and hole diameter, wrapping them around the drill Fn then will weld the ch~ps to the tool body or hole The diameter of the hole and the lengthidia kind of drill to use, Figure 23-14 explores how drill sel hole and the d~ameter of the drill: Section A shows t low holes and small diameters.About half of all the drilling process falls w ~ t h ~ thn egory of thls section. It is the section for which the m drills and a very few cemented carb~de drills Section B is the drillmg of deep h s that of shallow holes which cemented carbide gundnlls are used. Section C 1 large diameters, for wh~ch spade dr~lls are used. Sec large d~ameters,for whrch BTA tools are used.

MICRODRILLING As the term suggests, m~crodrilling involves very-small-diameter cutting tool&i c drills, end mills, routers, and other spenal tools. Drills from 0 0 2 in (0.05 rnm) and 0.005 in. ~ndiameter ate used to produce geometries involving dimens~ons at whic workp~ece mater~als no longer exh~bit uniform~ty and homogeneity. Grain border sions, alloy or carbide segregates,and microscop~c vo~ds are problems in microdnlling, holes of 0.02 to 0.0001 in have been dnlled using pivot drills, as shown in Frgure 23-1 Pivot dr~lls are two-lipped (two-fluted), end-cutting tools of relatively s geometry. Web thckness tapers toward the point, and a generous back-taper is porated. For softer workpiece materials, point angles are typically 118' and lip ance 1s 15".For steels and general use in harder metals, 135' pomts and 8" clearane recommended The chisel edge is simllar to that of a twist dr~ll. Pivot drills are ma tungsten-alloy tool steel in standard sizes from 0.0001 in. to 0.125 in. and of sm tungsten carbide Gom 0.001 in. t o 0.125 in Small drills easily deflect,and gettmg accurate and precise holes requlres ama with a high-qualay spindle and very sens~tive feedmg pressure. In the medical nents field, much of this machming work is performed on Computer Numerical (CNC) Swiss-type turning machines. Speeds and feeds are greatly reduced w~thfre pecking to clear the chips. Use a light, lard-based, sulfurized cutting oil.

23.4 Tool Holders for Drills SECTION

643

Microdrill Note back taper

D @ D
Point angle,
.15

Pivot rnicrodrili ery-small-diameter

/Clearance
8"-15O

angle

Shank

Straighl-shank drills must be held in some type of drill chuck (Figure 23-16). Chucks are adjustable over a considerable size range and have radial steel fingers. When the chuck is tightened by means of a chuck key, these fingers are forced inward against the drill. On smaller drill presses, the chuck often is permanently attached to the machine

la)

FIGURE 23-16 Two of the most commonly used types of dr~ll chucks are the 3-jaw Jacobs chuck (above) and the collet chuck with synthetic rubber support for jaws. (Courtesy of Jacobs Manufacturing Company)

spindle, whpreas on Larger drilling machines the chucks have a tapered shank into the female Morse taper of the machine spindle. Special types of chucks in tomatic or fully automaticmachines permit quite a wide range of sizes of drills to in a single chuck. Chucks mmg c h u ~ k keys cequlre that the mac change a drill. To reduce the downtime when drills change chucks are used. Each drtll is fastened in a Into the chuck hole while it is turning by merely r body. With the use of this type of chuck,center so on can be manu$lly changed in quick suecession. For carbide drills, collet-ry ers with thrust bearings are recommended (Figure 23-17), For drtlls using an coolant supply, a very rigid chuck with e ~ t b e an r inducer or through-spmdle s recommended. source i i Conventio&l holders such askeyless chucks cannot be used became the g strength is lnnited. Collet holders should be cleaned peridicaly with oil tor small clups. The enhre flute length must protrude fr length of flute protrudingfr~m the hole must be at least 1to 1.5 times the drill di Rad~al mnout at the drill tip must not exceed 0.001 i n .

Correct chucking w~th spring coiiet.

Chipscannof be removed if the flute is chucked.

Olmens~on Ashou be 1 to 1.5 times d

Bad

The drlil chuck r~gidity i s Thicker shanks can offer

Slip

hand one rev

Rotate by

The runout of the

drlll when held in

Dial indicator 0 OOl in. is w~th~n

G a p J

the chuck should be less than 0.001 in. Total ~ndicator Nnuut FIR)

when the work i$ totatlng.

Wnhin 0.004in. maxim

PlGURE 23-17 Here are some suggestions for correct chucking of carbide drills.

SECTION 23.6

Machrne Tools for Drilling

645

Work that 1s to be drilled is ordinarily heldin a vise or in specially designed workholdof crs called jigs. Workholding dev~ces are the subject of Chapter 25, where the des~gn workholdmg demces IS discussed. Many examples of drill jigs are shown. With regard to safety, the work should not be held on the table by hand unless adequate leverage is available, even in light drdhng operahons Thls is a dangerous pracbce and can lead to serious accidents, because rhe drill has a tendency to catch on the workplece and cause it to rotate, especially when the drill exits the workpiece Work that is too large to be held In a jig can be clamped directly to the mach~ne table usmg suitable bolts and clamps and the slots or holes in the table. Jigs and workhold~ng devices on indexing machines must be free lrom play and firmly seated.

Toc11,s FOR DRILLING

The bas~c work and tool mohons requlred for dr~lhng-relative rotation between the workpiece and the tool, with relative longtudmal feeding-also occur in a number of other machining operattons. Thus drilling can be done on a variety of mach~ne tools such as lathes, horizontal and vert~cal m~lling machmes, bormg machmes, and machming centers. This section will focus on those machines that are designed, constructed, and used primardy for drilling. First of all, themachine tools must have sufficient power (torque) and thrustto perform the cut. It is the task of the engineer to select the correct machine or select the cutting parameters (speed and feed) based on the drill d~ameter, drill marerial, and work material (hardness). Because of the complex geometry of the dr~ll, empirical equations are widely used. Flgure 23-18 shows the type of informat~on provided by cutting-tool manufacturers to calculate (estimate) thrust in drilling. Data with K, (specific cutting force) and Xand Y (empirical constants) are obtainedfrom cnttmg-tool manufacturers. Much of t h s kind of data has been developed for hlgh-speed-steel tools. When using sohd carbide tools, rigid machines such as machining centers or NC turning machmes are recommended, whereas a radial drllling machine is not recommended (not r~gid enough). Rigid~ty a especially Important in molding chatter. A lack of rigldity in the cutting tool, the workpiece, or the machime tool permits the affected members to deflect due to the cnttlng forces devekop~ng the conditions for chatter (see Chapter 20), with the result that the cutting lips have a hammering action against the work. So, use the shortest tool possible. In add~tion, backlash in the feed mechanism should be kept at a minimum to reduce strain on the drlll when it breaks through the bottom of the hole. The common namc for the machine tool used for drdling is the drill press. Dnll presses conslst of a base, a column that supports apowerhead, a spmdle, and a worktable. On small machines, the base rests on a workbench, whereas on larger machines it rests on the floor (Figure 23-19).The column may be either round or of box-type construction, the latter bemg used on larger, heavy-duty machines, except m radial types. The powerhead contains an electric motor and means for driving the spindle in rotation at several speeds On small drslling machines this may be accomphshed by shlfting a belt on a step-cone pulley, but on larger machines a geared transmlsslon 1s used. The heart of any drillmg machine is its spindle. In order to drdl satisfactorsly, the sp~ndle must rotate accurately and also resist whatever side forces result from the drillingprocess. In virtually all machines the spindle rotates 1n preloaded hall or taperroller hearings. In addltion to powered rotation, provision is made so that the spindle can be moved ax~ally to feed the dr~ll into the work. On small machines the spindle is fed by hand, using thehandles extending from the capstan wheel; on larger machines power feed 1s prov~ded. Except for some small bench types, the spindle contams a hole w t h a Morse taper in its lower end into which taper-shank dr~lls or drill chucks can b e inserted. The worktables on drihng machimes may be moved up and down on the column to accommodate work of various sizes. On round-column machines the table can usually

Axial Thrust Force

7

steel 14140) Hardness: 240 H B Cutting speed: 400 SFM

Mat.: Alloy

I
1

0
0

0
X

2 " 7

1'I2

Drill diameter

F, = Axial thrust
= 01.15 X K, X

f$*

where
D s Drill diameter ( ~ ~ C R B S ) K, = Specific cutting energy from table (in-lblin21 f, = Feed (in./rev)

'I
'I
I

I
Values in in.-lb/in2.
FIGURE 13-18 Estimating the thrust force in drilling; example from Waukesha Cutting Tools.

be rotated out of the way so that workp~eces can be mounted d~rectly on the bas some box-column machines the table 1s mounted on a subbase so that it can b In two directions ~na hor~zontal plane by means of feed screws. Figure 23-19 shows examples of common types of drilling machines used duction environments. Drilllng machines usually are classmed as bench, upri single spindle, turret or NC turret, gang, multispindle, deep-hole, and uansfer.

, ,
L

Power

head

Spindle-

Table

(dl

FIGURE 23-19 Examples of drilling machines: (a) upright column drilling machine; (b) CNC turret drilling machine; (c) gang-drilling machine; (d) radial drill press; (e) rnultiplespindle drilling machine.

(el

CHAPTER 23

Drilling and Related Hole-Maklng Processes Wtth bench dtill presses, holes up to in. in dtameter can be drilled.The same of machine can be obtained w ~ t h a long column so that it can stand on the floor. of bench and upright drilling machines is designated by twrce the distance from teriine of the spindle to the nearest point on the column, t h s belng an indication o maximum size of the work that can be drilled in the machines.For exarnple,a 15press w~ll permit a hole to be drilled at the center of a workp~ece 15 in. in diam Sensit~ve drilling machines are essentially smaller, plan bench-type machlnes more accurate splndles and bearings.They are capable of operating at higher to 30,000 rpm. Very sensitive hand-operated feeding mecham~ms are provld in drilllng small holes Such machines are used for tool and die work and for very small holes, often less than a few thousandrhs of an inch in diameter, when hi dle speeds are necessary to obta~n proper cuttlng speed and sensitive feel to dehcate feeding to avo~d breakage of the very small dnlls. Upright drilling machines usually have spindle speed ranges from 60 to 350 and power feed rates,from 4 to 12 steps, trom about 0 004 to 0.025 in. rev. Most m machines use a singlespeed motor and a geared transmission to provide the ran speeds and feedsTlae feed clutch disengages automahcally when the spindle reac preset depth. Worktables on most upright drilhng mach~nes contain holes and slots for clamping work and nearly always have a channel around the edges to collec fluid, when it s used. On box-column machines, the table is mounted on verti on the front of the column and can be raised or lowered by means of a crank-oper elevat~ng screw. In mass production gang-drilling machines are orten used when several re1 operations, such as dr~ll~ng holes of different sizes, reaming, or wunterboring, mus done on a single part.These consist essentially of several independent columns, he and spindles mounted on a common base and having a smgle tabIe.The work can be into position for the operation at each spmdle.Tney are available with or withoutpo feed. One or several operators may be used.This machine would be an example of a ple small cell except that the maclunes are usually not single-cycle automatics. Turret-type, uprighr drrllzng machrnes are used when a senes of holes of diffe sizes, or a series of operations (such as center dr~llmg,drilling, reaming, and spot fa must be done repeatedly m succession. The selected tools are mounted in the t Each tool can qu~ckly be brought ~ n t o posttion merely by rotation of the turret machines automatically provide mdividual feed rates for each spindle and are o numerically controlled. Radial drilling machrne tools are used on large workpieces that cannot be ea handled manual1y.A~ shown III Figure 23-19, these machmes have a large, heavy,ro vertical column supported on a large base.The column supports a radlal arm tha be raised and lowered by power and rotated over the hase.The spindle head, wit speed- and feed-changing mechamsm, IS mounted on the radial arm.It can be move ~zontally to any des~red position on the arm.Thus the spindle can quickly be posi properly for drilling holes at any point on a large workpiece mounted e~ther on the of the machine or even sitting on the floor, Plan radial drilhug machines provide only a vertical sptndle motion. On se niversal machmes, the sp~ndle head can be pivoted at an angle to a vertical plane universal machines, the radial arm 1s rotated about a horizontal axis to permit at any angle. Radial drilling machmes are des~gnated by the radius of the largest dlsk in a center hole can be dr~lled when the spindle head is at its outermost position. Sizes 3 to 12 ft are available. Radial drill~ng machines have a wide range of speeds and f can do boring, and include provisions for tapping (mternal threading) (see Chapter Multiple-spindle drrllzng machines (F~gure 23-19) are mass-production ma with as many as 50 spindles driven by a smgle powerhead and fed simultaneous y the work Figure 23-20 shows an adjustable multiple sp~ndle head that can be mou on a regular single spindle-drill press. Ftgure 23-20 shows the methods of dr~ving posit~oning the spmdles, which permit them to be adjusted so that holes can be dri

Adjustable drill head 6 Production 50 pleces Sp~ndle.

Geared d r ~ lhead l Spindle: 8 Production: 80,000 pleces

Gearless d r ~ lhead l 30,000 pleces Spindle. 16 Product~on:

[,GJ, (
0
3A
ipm

lOSo Rim

An adjustable drill head should be cons~dered for low production jobs. However, many short run jobs such as this would be requlred to justify a multiple-spindle head.

Ageafed dull head is most appropriate in t h ~ s s~tuation, where there 1 s a large d~fference in srzes and a htgh dally producttan

Only a gearless head can perform this operation in one pass, due to the close proximity of the spindle centers

FIGURE 23-20 Three bas~c types of mult~ple-sp~ndle drlll heads: (left) adjustable; (middle) geared; (right) gearless (Courtesy of Zagar Incorpomted.)

at any location within the overall capaclty of the head. Special drill jigs are often desrgned and built for each job t o provide accurate guidance for each drill. Although such machmes and workholders are quite costly, they can be cost-justified when the quanhty to be produced will justlfy the setup cost and the cost of the jig. Reducing setup on these machines is d~fficult.Numericallycontrolled drill presses other than turret drill presses are not common because drilling and all its related processes can be done on vert~cal or horizontal NCmachining centers equipped with automatic toolchangers (see Chapter 31). Special machines are used for drilling long (deep) holes, such as are found in rifle barrels, connecting rods, and long spmdles H ~ g h cuttlng speeds, very light feeds, and a copious flow of cutting fluid ensure rapid chipremoval.Adequate support for the long, slender drills is required. In most cases horizontal mach~nes are used.The work is rotated in a chuck with steady rests providing support along its length, as required. The drill does not rotate and is fed into the work.Vertica1 machines are also available for shorter workpieces. Notice the wmilariry between this process and boring.

For shallow holes, the general rules relating to cutting fluids, as given in Chapter 21, are applicable. When the depth of the hole exceeds one diameter, it is desirable to increase the lubricating quahty of the fluid because of the rubbing between the drillmargins and the wall of the hole. The effecbveness of a cuttmg fluid as a coolant is quite variable in drilling. While (he rapid exit of the ch~ps is a primary factor in heat removal, this action

Work Mat~nal
Aluminurnandits alloys Brass Copper Cast 1rm Malleable lron Monel metal Slamless steel Steel, ordmary Steel,very hatd Wrought !ran

Cutting Fluid
Soluble 011, kerosene, and lwd-oil cgmpaunds,ltght. nouv~smus neutral oil, kerosene andseiuhleod m Dry or asoluble oil: kerosene and lard-or1 compounds: I~ght,noneiscousneutral o i l Soluble oil,strained lard 091, olwc-wtd compouna Dry or with sjet of comprP:scd arr for o o o h g Soluble 011,non~saousneutral o i l Solubk 011, snlfut~zed mmeral OIL Soluble ail,sulfur~zed mlneral 011 SoIuhIe oil, sullurized all, htgh extreme-preasure-value minerdl 0 1 1 Soluble oilsulfunaed oil, turpEnhne Soluble 0 1 1 , sulfunzedod, mineml-amal oil compound

. .

Nest oil sen be used effeBve1y with fhe solrd oarbtde drillsfor low.spaed drilling (up to 130 &film). If thewwksurkce becorns hardv b l u e m c o i o r , d e ~ fhvrpm ~ ~ and use nest MI For heavy duty cpttmgemulsroptype at1contmntngso~ne extreme pressure addidveis rammmqnded A volumeof 3 0 gallat a pressufeof37-62 I W ~ I ' Xrommlnended ~ A double ?<Team supply ofilulda rccomended

also tends to restfict entry of the cutting fluid.Th~s is of particular nnportance in dri mater~als that have poor heat canduct~vity. Recommendations for cutting fluid dgilling arc given in Table 23-5. I f the hole depth exceeds two or three diameters, it is usually advantag to withdraw the drill each time i t has drilled about one diametar of depth, to chips from the hole. Some machines are equipped to provide this "pecking" automatically. W'ere cooling is desired, the f l u ~ d should be applied copiously, For severe c tions,drills containing coolant holes have a considerable advantage. Not only is th supplied near the cutting edges, but the coolant flow aids in flushing the chipsfro hole.Where feasible, drilling horjzontaliy has hstinct advantages over drilling verti downward.

23.8 COUNTERBORING, COUNTERSINKING, AND SPOT FACING

Dnllimg is often followed by counterbarzug, ~ountersmking~or spotfacmng. As sh Figure '2321, each provides a bearing surface at one end of a drilled hole.They a ally done with a special tool having from three to six cutting edges. Caunterboring provides an enlarged cylindrical hole with a flat bottom so bolt head, or a nut, will have a smooth bearing surface that is normal to fhe q x ~ hole; the depth may be sufficient so that the entlre bolt head or nut wiil be below face of the part, Thepilot on the end of the tool fits into the drilled hole and help sure conoentricity with the original hole-Two or more diameters may be produce single counterboring operation. Counterboring also can be done with a single-p although this method ordinarily is used only on large holes and essentially is aperation. Some counterboring tools are shown in Figure 23-21b. Counrersank2ng makes a beveled section at the end of a drilled hole to pro proper seat for a flat-head screw or rivet. The most common angles are 6n", 82 90'. Countersinking tools are similar to wuuterbonug tools except that the cutting are elements of a cone, and they usually do not have a pilot because the bevel of th causes them to be self-centering. Spotfacfng is done to provide a smooth bearing area on an othemse r face at the openlng of a hole and normal to its axis. Machining is limited to the depth that will provide a smooth, urnform surface. Spot faces thus are somew andmore economical S o produce than counterbores They areusually made with tiedged end-cutting tool that does not have a pilot although counterboring to frequently used.

Reaming

Counterbore

Counters~nk

Spot face

Reaming removes a small amount of material hom the surface of holes. It is done for two purposes: to bnng holes to a more exact size and to Improve the finish of an existing hole. Multiedge cutting tools are used. as shown in F i u r e 23-22. No special machines are built for reaming.The same machine that was employed for drilling the hole can be used for reaming by changing the cutting tool. To obtain proper results, only a mlnimum amount of materials should be left for removal by reaming. As little as 0.005 in. is desirable, and in no case should the amount exceed 0.015 1n.A properly reamed hole will be within 0 001 in of correct size and have a fine fin~sh. The pr~ncipal types of reamers are shown in Flgures 23-22 and 23-23. Hand reamerr are intended to be turned and fed by hand and to remove only a few thousandths of an inch of metal.They have a straight shank with a square tang for a wrench.They can have straight or spiral flutes and be solid or expandable. The teeth have relief along their edges and thus may cut along their entire length. However, the reamer is tapered from 0.005 to 0.010 in. in the first third of its length to assist in starting it in the hole, and most of the cutting therefore takes place in this portion. Alachine or chucking reamers are for use with various machine tools at slow speeds. The best feed is usually two to three hmes the drilling feed. Machme reamers have chamfers on the front end of the cutting edges The chamfer causes the reamer to seat firmly and concentrically in the drilled hole, allowing the reamer to cut at full diameter.De longitudinal cuttlng edges do little or no cuttmg. Chamfer angles are usually 45" Reamers have straight or tapered shanks and straight or spiral flutes Rase-chucking reamers are ground cylindrical and have no relief behind the outer edges of the teeth. All cuttlng is done on the beveled ends of the teeth. Fluted-chucktag reamers, on the other hand, have relief behind the edges of the teeth as well as beveled ends. They can therefore cut on all portions of the teeth Their flutes are relatively short, and they are intended for light finishing cuts. For best results they should not be held rigidly but permitted to float and be aligned by the hole

Chucking

reamer Helical flutes ir h. helix shown) Reamer Chamfer Chamfer dlameter

Shank length Overall length Hand reamer, pilot and guide

-1

Flute length Chamfer length

I Guide1 -

Cuner sweep
i

Reamer d~arneter Stra~ght

"Up I

Pilot

1 /

. - t

k-L

Shank length

4 ------tw
Neck Overall length

Flute length

<

' Cuner sweep

FIGURE 23-22 Standard nomenclature for hand and chuck~ng reamers.

Shell reamers often are used for larger sizes in order to save cutting-toal ma al.The shell, made of tool steel for smaller sizes and with carbide edges for largerd or for mass-production work, is held on an arbor that is made of ad&@ 6 arbor may be used with any number of sheIls. Only the shell is ~ u b j e & . k @ ~ ~ to be replaced when worn.They may be ground as rose or fluted rmmws. Expansion reamers can be adjusted over a few thousandth@$ & & ; I& pensate for wear or to permit some variation in hole size to be o b t a i n - e 8 . T k ~ ' y ~ : able in both hand and machine types.

FIGURE 23-23 Types of reamers. (top to bottom) Straightfluted rose reamer, stra~ght-fluted chucklng reamer, straight-fluted taper reamer, straight-fluted hand reamer, expansion reamer, shell reamer, adjustable 1nsel.tblade reamer.

Adjustable reamers have cutting edges in the form of blades that are locked in a body. The blades can be adjusted over a greater range than expansion reamers. This permits adjustment for size and to compensate for regrinding. When the blades become too small from regrinding, they can be replaced. Both tool steel and carbide blades are used. Taper reamers are used for finishing holes to an exact taper.They may have up to eight straight or spiral flutes. Standard tapers,such as Morse, Jarno, or Brown & Sharpe, come in sets of two.The roughing reamer has nicks along the cutting edge to break up the heavy chips that result as a cylindrical hole is cut to a taper.Thefinishing reamer has smooth cutting edges.

REAMING PRACTICE If the material to be removed is free-cutting, reamers of fairly light construction will give satisfactory results. However, if the material is hard, then tough, solid-type reamers are recommended, even for fairly large holes. To meet quality requirements, including both finish and accuracy (tolerances on diameter, roundness, straightness, and absence of bell-mouth at ends of holes), reamers must have adequate support for the cutting edges, and reamer deflection must be minimal. Reaming speed is usually two-thirds the speed for drilling the same materials. However,for close tolerances and fine finish, speeds should be slower. Feeds are usually much higher than those for drilling and depend upon material. A feed of between 0.0015 and 0.004 in. per flute is recommended as a starting point. Use the highest feed that will still produce the required finish and accuracy. Recommended cutting fluids are the same as those for drilling. Reamers, like drills, should not be allowed to become dull. The chamfer must be reground long before it exhibits excessive wear. Sharpening is usually restricted to the starting taper or chamfer. Each flute must be ground exactly even, or the tool will cut oversize. Reamers tend to chatter when not held securely, when the work or workholder is loose, or when the reamer is not properly ground. Irregularly spaced teeth may help reduce chatter. Other cures for chatter in reaming are to reduce the speed, vary the feed rate, chamfer the hole opening, use a piloted reamer, reduce the relief angle on the chamfer, or change cutting fluid. Any misalignment between the workpiece and the reamer will cause chatter and improper reaming.

I l i Key Words
lter core drill
sel end rck

rnterhoring ~ntersinking :p-bole drilling I1 press

lling te

gang-drilling machine gundrill helix angle indexable insert drill jig lip multiple-spindle drilling machine

radial drillingmachine reaming, hand reaming, machine shell reamer spade drill spot facing suhland drill
tang

thrust force lrepannlng turret drlll~ng machine tw~st drill web

ill Review Questions

What functionsarc performed by the flutcs on a standard twist drill? What determines thc rake angle of a drill'? See Figure 23-2. Basically. what determines what helix angle a drill should have? , When a large-diameter hole is to be drilled. why is a smaller,

diameter hulc oltcn drilled first? Equation 21-4lor the MRR for drilling can be thought of as where f,N3 is the leed rate of the drill bit. -tilnes Are thc rccomrnended surface speeds for spade drills given

in Table 23-3 typically liighcr or lower than those reconlmcnded lor lwisl drills? I-low about the fccds? Why? 7. What can happen when an improperly ground drill is used to drill a hole? 8. Why are most drilled holes oversize with icspea to the nominally specified diamctcr'? 9. What arc thc two primal? lunctions of a comhination ccntcr
drill? 10. What is the function of lhc margins on a twist drill'!

11. What factors tend to cause a drill to "drift" off the centerline

of a hole? 12. The drills shown in Figure 23-13 have coolant passages in the flutes. What is the purpose of these holes? 13. In drilling, the deeper the hole, the greater the torque. Why? 14. Why do cutting fluids for drillingusually havemorelubricating qualities than those for most other machining operations? 15. How does a gang-drilling machine differ from a multiplespindle drilling machine? 16. How does a multiple-spindle drilling machine differ from a NC drilling machine with atool changer that would hold all the drills found in the multiple-spindle machine? 17. How does the thrust force vary with feed? Why? 18. Holding the workpiece by hand when drilling is not a good idea. Why? 19. What is the rationale behind the operation sequence shown in Figure 23-10? 20. In terms of thrust, what is unusual about the slot-point drill compared to other drills?

21. What is the purpose of spot facing? 22. How does the purpose of counterboring differ from t spot facing? 23. What are the primary purposes of reaming? 24. What are the advantages of shell reamers? 25. A drill that operated satisfactorily for drilling cast iro very short life when used for drilling a plastic. What ml the reason for this? 26. What precautionary procedures should be used when ddl a deep,vertical hole in mild steel when using an ordinan t drill? 27. What is the advantage of a spade drill? Is it really a drill7 28. What is a "peeking" action in drilling? 29. Why does drill feed increase with drill size? 30. Suppose you specified a drilling feed rate that was too,l What kinds of problems do you think this might cause Figure 23-6 and Table 23-4 for help.

Problems
1. Suppose you wanted to drill a 1.5-in.-diameter hole through a piece of 1020 cold-rolled steel that is 2 in. thick, using an indexable insert drill.What values of feed and cutting speed will you specify, along with an appropriate allowance. Is this the correct tool? What other drill types could he used? 2. How much time will he required to drill the hole in Problem 1 using the insert drill? 3. What is the metal removal rate when a 1.5-in.-diameter hole, 2 in. deep,is drilled in 1020 steel at a cutting speed of 200 fpm with a feed of 0.010 ipr? What is the cutting time? 4. If the specific horsepower for the steel in Problem 3 is 0.9, what horsepower would be required, assuming 80% efficiency in the machine tool? 5. If the specific power of an AISI 1020 steel of 0.9, and 80% of the output of the 1.0-kW motor of a drilling machine is available at the tool, what is the maximum feed that can be used in drilling a 1-in.-diameter hole with a carbide drill? (Use the cutting speed suggested in Problem 3.) 6. Show how the approximate equation 23-5 for MRRin drilling was obtained. What assumption was needed? 7. A workpiece must have 10 holes finished in it. Manual layout time is 'I2hripiece.To driil and ream all the holes requires 1 hour on themachine foreachpiece,not counting layout or setup.Tne laborrate is$lO!hr and the machine rate is $20ihr. If a jig isused, the labor cost to lay out each piece can he saved. Both methods give the same-quality product, but this jig saves 40 min in processing time on the machine. How large alotjustifies the use of a jig that costs $150 to make (labor and materials)? 8. A part has two holes located for drilling by manual layout. If a drill jig is used, 0.5 min in processing time is saved for each piece.Re labor rate is $9!hr,The overhead rate on the labor saved is 100%. Setup time is no more with than without the jig. The combined rate for interest, insurance, taxes, and maintenance is 35%.The cost of the jig is $500. a. How many pieces must be made in one lot to make the jig worthwhile? h. How many pieces must be made on the jig in one lot each month to earn the cost of the jig in two years? 9. Manufacturer's charts will help determine the best feed and

speed to run the drills. For example, a 1.5-in holeis to be in 4140 steel annealed to Bhn 275. For the spade drill, is 80 sfm; feed, 0.009 ipr: and spindle rotation, 204 rpm. the indexable insert drill, speed is 358 sfm,feed,0.007 ipr; spindle rotation, 891 rpm. Typically, an indexable insert can produce a hole four times faster than a spade drill bu cost (with inserts) 50% to 75% more than the equivalents blade and holder. For making only a couple of holes, the cost is not usually justified. Determine the number of h needed to justify the extra cost of the indexable insert Some additional cost data are given below. Ignore tool life and assume that the blades audi indexable drills make about the same number of holes is this a reasonable assumption?) The holes are 3 in. deep n o allowance needed. Cost of drills: Spade drill $139.00holder +21.90 per blade $160.90 Indexable-insert Dril $273.00drill +12.80 per two inserts $285.80

Assume for this example that a machine rate of $ 4 5 h in the cost of labor and machine burden. 10. Assume that you are drilling eight holes, equally spac bolt-hole circle. That is, there would be holes at 12,3, 9 o'clock and four more holes equally spaced between The diameter of the bolt hole circle is 6 in. R e designe that the holes must be 45" 2 1" from each other aroun circle. a. Compute the tolerance between hole centers. b. Do you think a typical multiple-spindle drill setup be used to make this bolt circle-using eight drills: once? Why or why not? c. D o you think that the use of a jig may help improv situation? d. D o you think a CNC drilling process could do the best? 1 1 . A part with seven holes can be machined on a numec controlled turret drill press in 3 min (estimated time bas

Case study lsunllar parts). The rate on the CNC machine for labor is $34/hr.Currently, the part is being machined on a gang drill press with a special jig in 10 minutes per piece.The jigfor the gang drill costs $300;the combined rate for depreciation, interest, insurance, and taxes is 135%; and the hourly rate for the gang drill and operator is $16. Setup time is about the same for both machines. For how many pieces is it economical to switch to the CNCY

655

12. It is estimated that a jigfor machining a part with three holes costs $400 and with it the operation takes 15 min per part. The operation can be done without a jig on a numerically controlled drill press in 5 min. Assume that any other conditions are the same as in Prohlem 11. How many pieces are needed to cost-justify the use of a jig?

Bolt-down Leg on a Casting

teve Humer is a consulting engineer and h a s lust recelved . the drawlng shown in Figure CS-23 This ISone of four legs on a castrng made by the CRS Company. These legs are used to attach the devlce to the floor.The ~ectlon drawlno to the right shows the typical loadlkl to s sublected.The company is currently whrch the leg I drilling the bdt hole and then counterboring the land, but rnanwfacturlng h a s experienced some difficulty in machrn~ng the four holes. They report a lo! of drill breakage Qualrty control reports that distances between the four holes a r e frequently too large. Sales has recently reported that a substantial number of in-servicefallures

have occurred w~th these legs. Steve ha$ obtained a sketch from sales show~ng where the legs typically tali. This castlng IS manufactured from gray cast Iron us~ng the sand casting proces's. I. Vthat machining difflcuit~es should Steve suspect thrs leg to have? 2 why were the distances hetween the holes too large? 3 What should Steve recommend for solving these probJems m the future m terms of materials, desrgn, and manufacturing methods? 4. What should Steve recommend be done with the UnltS in the freld to stop the failures?

Counterbored

Ugure CS-23 shows the design of one of four legs on a casting made by t h e BRC Company.

MILLING
24.1 INTRODUC~ON 24.2 FUNDAMENTALS OF MILLING PROCESSES Eace Milling Example End Milling Example Up versus Down Milling Milling Surface Finish

11 24.1 INTRODUCTION
Milling is a basic machining process by which a surface is generated by progressiv

in a single p-ass of the cutter or work and, because very good surface finish can

11 24.2 FUNDAMENTALS OF MILLING PROCESSES

depends on the feed per tooth,f,

inches according to

N,

=-

12v
7rD

The depth of cut, called DOC or d in Figure 24-1, is simply t h e distance be, the old and new machined surface.

Cun~ng edge

In
d = DOC

cutter

7
I

Workp~ece

----+ f",

( b )Slab m~lllng-multiple tooth

(a) Horizontal-spindlemilling

machine

Work
(c)Allowances for cutter approach FIGURE 2 4 1

Mater~al removed by one tooth

(dl Feed

- ---*

per tooth

Peripheral milling can be performed on a horizontal-spindle milling machine. The cutter rotates at rpm N,, removing metal at cutting speed \/.Theallowance for starting and finishing the cut depends on the cutter diameter and depth of cut, d. The feed per tooth, f, and cutting speed are selected by the operator or process planner.

The width of cut is the width of the cutter or the work,m inches,and is glven the symbol W.The length of the cut L is the length of the work plus some allowance La for approach and overtravel. The feed of the table f,, in inches per minute, is related to the amount of metal each tooth removes during a revolution, the feed per tooth f,, accordmy to

f", = fdN*fl
where n is the number of teeth in the cutter (teeth rev.). The cutrzng time rr

(24-2)

f i e length of approach is

L , = J ~ = v 2 m
The metal removal rate is

MRR

== ---- =

volume
Tm

LWd
Tm

Wf , d in.?/mm

(24-5)

ignoring LA.Values forf, are given inTable 24-1, along with recommended cutting speeds in feet per minute.

Carbide Cutter8

High-Speed-Steel Cutters Full and

-

Feed -- (~n./tooth)

Full Form Mills

005-010 175-27s 005-010 150-3E
.U%.NO 350-6(10 004-010 311MS0 W3- 008 1%3W 001- a08 100-150 005- 010 1W-175 00&.012 175-250 008-,015 250-350 .00a 015 200-8M) 00&.015 500-1500 .008- 015 2000 UP

Material
MalIeable Iron SofL'hard Cast steel SofUhard 100-150 BHh' %eel 15Q-250 BHN w l 2M-3~0 B8N steel 350-450 BHN steel Cast Iron,hard BHN WMZS imn, medium BHN 180-225 Csst lron,sofr B a . 4 i50-180 hnze %/hard

Speed
/fpm)

Face

Mills

Slab

Mills

End M~lls
00>010 2UIt3.B ,003-,010 15%350 005-.OIO 450-600 O W (110 30W50 ,0051010 150-300 904-008 100-150 .W3-,008 125-2W 005-.010 200-275 005- 012 275400

Half S~de M~lls Saws

.WS-010 200-300 005-010 150-350 008-012 450-800 007-012 31M45O O C 1 5 . 0 U 160-300 005-,012 125-180 001-010 125-200 .QOS-.012 ZOO-275 0 0 8 015 275-dW ,008-,012 300-10(10 ,008- 012 500-1500 008- 025 2WD W

Face Mills
095-015 a100 010-,015 40-60 ,015-,030 80-I30 010-020 50-70 OG- 010 354 003-008 20-35 ,005- 012 4040 . 1 ) 1 0 , 0 2 0
6 W

Slab Mis
005-,015 6 W 0

End Mills 003-015 661W oo5-a10 40-60 003-010 80-140 003-010 6D-8U ,001-.010 4 .0031.101 XMl 003-008 40-50 003- 010

and Half-Slde

E
Form Saws 003-006 60-100 ou2-00s

Mtlls
006-,012 6160 nos- 010 40-60 010-.020 W30 010-015
3-76

M~lls
.00>010 60-80

9
h l 01

Brm
*fvh%rd AIurmnum alloy SoPl/hard

~ e e per d tooth Speed, fpm Feed per iaoth Spced, (pm Feed per tooa Speei(,tpm Feed per twth S.peed,fprn Feed per tooth Speed,fpm Feed per tooth Speed fpm Feed per tooth Spaed,fpm ~ e z per d tooth Speed,fpm Feed per toofb Speed,fpm Feedperloath Speed,fpm Feed per tooth Speed,fpm F e d per moth Speed, fpm

005-.015 200-300 .008-01s 150-350 01&015 4.50-Roo 010-,015 300-450

005-015
200-300

005-015 150-350 008- 015 450-600 ,008-,015 3W450 .008-008-015 007- 012 180-309 150-300 .00%.015 007- 012 125-180 Il&tS0 .00S-,Bl0 005-010 12S20rJ 100-175 00% 015 00%.015 290-275 175-250 015--.O25 010- 020 2754Ni 250-350 Ql&.OZO 01%.02,0 300-100 30a-800 .010-.M0 010- 020 50b1500 500-15011 OIU-.LMQ ,010- 030 20W) Up 2MF)UP

a05-.016
30p1000 m- 010 500-1500

0 0 % 015 2MUP

U0%004 200-350 002- 004 150-300 OM006 356600 003-.W6 W 5 0 .002-005 150-300 001- 004 100-150 002-003 125-200 . O M OD4 200-250 Aft-004 25@-350 OM-.W 3U0-10W 0 0 3 004 50U-lSfM 03- 006 2 W O UP

.olrrrns
40-60 008-015 80-130 aD8-015 50-70 005- 010 35-50 005-008

008-012

3 =

4060
003-W6 70-100 R03-W6 50-70 W2- 005 35.50 ,001-.OM 20-35 002- W4

40-m
.OM-010 70-100 .006-0tO

SMO
005-,010 3 . % % W3-.008

$05- 010

35-JO
.W3-,008 2%35 035- U10 40-60 008- 015

20-35
,005- 010 3540 008- 015 50-70 ,010- 025 70-110 ,008-,020 50-200 ,008-020 100-300 015-,040 3W1200

60-90
004-010 80-120 WS.010 50-224 .00.L.015 150-3% 015- M 3W!-1200

013-030 80-I20 010-,025 50-225

.010-M5
150-300 010-,040 3W1200

20-35 005- 010 35-60 35-50 00'3-005 .008- 012 60-70 5040 f2-80 O 3 ,010- 015 ,010- 020 002- G 6 W 70-110 $&I20 003-,005 008- 015 .008-015 50-250 50-200 5G'E .My8-,015 ,003-W5 .WE-013 150-300 1W3O 150-350 0113-,020 004- 008 010- 0% 30M2W 3CG1200 300-1000

"G*nerafl~, lower end of urnge uwd far imuled blade mtters hqhshor end olrangefwinliexable i n s ~ icutters t

Inface mrllrng and end mlll~ng, the generated surface is at r~ght angles to the cutter axis (F~gure 24-2). Most of the cutting 1s done by the penpheral portions of the teeth, with the face portions providing some finishing actton. Face milling 1s done on both horizontal- and vertical-spindle machines. The tool rotates (face mills) at some rpm ( N , ) while the work feeds past the tool. The rpm is related to the surface cuttmg speed V and the cutting tool diameter D, accord~ng to equation 24-1.The depth of cut 1s d, in inches, as shown in Figure 24-2b.The width of cut is W,in inches, and may be width of the workpiece or width of the cutter, depending on the setup. The length of cut is the length of the workpiece L plus an allowance LAfor approach and overtravel Lo, in inches.The feed rate of the table f , , m inches per minute,is related to the amount of metal each tooth removes dunng apass over the work, called the feed per toothf, so f;,, = f,N,n where the number of teeth in the cutter is n.The cuthng time is

T,
The metal removal rate is M

+ LA + Lo mm
frn
-

volume R R
' m

- -1
" 1

LWd

.r

f,,,Wd i ~ ~ . ~ / m i n

When calculating the MRR, ignore Lo and LA The length of approach is usually equal to the length of overtravel, which usually equals D/2 in. For a setup where the tool does not completely pass over the workpiece,

FACE MILLING EXAMPLE

A 4-m.-diameter, six-tooth face m~ll is selected, usmg carb~de inserts (F~gure 24-3). The material being machimed is low-alloy steel, annealed Usmg culting data recommendations, the cutting speed cho~en is 400 sfpm with a feed of 0.008 in.1tooth at a d of 0.12 inches. Determining rpm at the spindle,

N , = - 12V =

TD

12 X 400 3.14 X 4

392 rpm

Determining the feed rate of the table, f , = nN,f,,

f,

0.008

6 x 392

191n./min

If slab or side milling were being performed, as shown in F~gure 24-4, with the same parameters being selected as above, the setup would be d~fferent but the spindle rpm and table feed rate the same.The cutting tlme would be different because the allowances for face milling are greater than for slab nulling. In milling, power consumption is usually the limiting factor. A th~ck chip 1s more power efficient than a thin ch~p.

END MILLING EXAMPLE

End milling is a very common operation performed on both vertlcal- and horizontalspindle milling machines or machinmg centers. Figure 24-5 shows a vertlcal splndle end milling process, cutting a step in the workplece.This cutter can cut on both the sides and ends of the tool. If you were performing this operation on a block of metal (for example, 430F stamless steel), you (the manufacturing engineer) would select a specific machine tool. You would have to determine how many passes (rough and finish cuts) were needed to produce the geometry specified in the deslgn Why? The number of passes deternines the total cutting tlme for the job. Using a vertical-spindle milling machine, an end mill can produce a step in the workplece. In Figure 24-5, an end mill with six teeth on a 2-in. diameter is used to cut a step in

I
Colum SIX-toothface mill

(a1Vert~calsprndle mllllng mach~na Top vlews

(bi Face milling over part of surface

Machrned surface

(c) Allowance

for part~al coverage

Id) Allowance for full coverage

FIGURE 2 4 2 Face milling is often performed on a spindle mllllng machlne uslng a mulbple-tooth cutter (n = 6 teeth) rotatlng N,at rpm t o produce cuttlng speed K The workp~ece feeds at rate f,,, In inches per minute past the tool. The allowance depends on the tool diameter and the w ~ d t h of cut.

Top view

Workpiece

---f,

feed rate = 19 ln./min

=

FIGURE 24-3 Face milling viewed from above with vertical spindle-machine.

i&/
7 ----+ f , , Workpiece
IrM<ned surface

FIGURE 2 4 4 Slab orside milling being done as a down milling process with horizontal sp~ndle-machine.

End View

(not to scale)

Side V~ew

Tool - End Mill = 2 in. diam.. 6 teeth

7 ror V I S ~

feed rate Iinimin)

Vlse

I
I

Warkp~ece stalnless 4 3 0 .

Depth of cut I DO1 = Depth o f immersmn

d=

CURE 24-5

End milling a step feature in a block using a flat-bottomed, end mill cutter in a vertical spindle-milling machine. left, photo. In middle, end view, table moving the block into the cutter. On right, side view, workpiece feeding right to left tool. 430F stainless. The d (depth of cut) is 0.375 in., and the depth of immersion (DOI) is 1.25 in. The tool deflects due to the cutting iorces, so the cut needs to be made a t full immersion; but there may not be enough power lor a full DOC. Can the step be cut in one pass or will multiple cuts be necessary? The vertical milling machinc tool available has a 5-hp motor with an 80% efficiency.The specific horsepower for 430F stainless is 1.3hp/in.'/min. The maximum amount of material that call be removed per pass is usually limited by the available power. Using Lhe hp equation from Chapter 21, hp
=

HP,

MRR

HP,

f;,,WD = HP,/;,, X DO1 X d

(24-9)

Select5 = 0.005 ipt and V = 250 fpm from Table 24-1. Calculate the spindle rpm:

N,

12 X 250 3.14 X 2

477 rpm of cutter

Next, assuming the machine tool has this rpm available, calculale the table feed rate:

h,

f, X rr

N,

0.005 X 6

477

14.31 in./min

But the actual table feedrates for the selected machine are 11 in./min or 16 in./min, so, being conservative, select

f,, = table feed rate

11.00 in./min

Next,assuming 8076 of the available power IS usedfor cutting,calcnlate the depth of cut from equation 24-9:
d = DOC

5 x 0.8 1.3 X 11.00 X 1.25

0 225

-- 0.225 in. maximum

Therefore, two passes are needed because (0 37510 225 = 1.6). 0.375
-

=
=

0.150 m. second pass D O C

2 parses DOC DOC

0.225 rough cut = 0 150 finrsh cut 0.375 total D O C

Note that f o r d 0.0051 ipt:

ft

0 150, the feed per tooth would be only slightly increase

0.5
X

= 1.3 X 6 X 477 X 0.150 X 1.25 = 0.0051 tn./tooth

0.8

You may want lo change f, to improve the surface finish. With a smaller f,, a ter surface finish is usually obtamed. However, there are other factors to consider, machinmg tlme. In general (for face, slab, or end m~lling), rf machme power is lacking the follo actlons may help.

1. Use a cutter with a positlve rake as this can be more efficient than one with a ne
rake. 2. Use a cutter with a coarser pitch (fewer teeth). 3. Use a smaller cutter and take several passes (reduce d or D01).

UP VERSUS DOWN MILLING

For either slab or end or face &ng,
surfaces can be gen~rated by two dishnctly methods (F~gure 24-6). Up mzll~ng is the traditional way to mill and is called co tronal milling.The cutter rotates aga~nst the d~rection of feed of the workp~ece. In or down millmg. the cutter rotation is in the same direction as the feed rate.The me of chip formation is completely different in the two cases. In up milling, the chip is very thin at the beginning, where the tooth first c the work; then it increases in thickness, becomlng a maximum where the toothleave work.The cutter tends to push the work along and hft it upward from the tah1e.n tion tends to eliminate any effect of looseness in the feed screw and nut of the machme table and results in a smooth cut. However, the actlon tends to loosen the from the fixture.Therefore, greater clamping forces must be employed, w ~ t h the d of deflecting the part. In addition, the smoothness of the generated surface de greatly on the sharpness of the cuttmg edges. In up mlling, chips can be carried m newly machined surface, causing the surface fin~sh to be poorer (rougher) than m milling and causmg damage to the insert. In down maing, maxunum chip thickness occurs close to the pomt at wh' tooth contacts the work. Because the relative motion tends to pull the workpie the cutter, any poss~bility of looseness m the table feed screw must be eliminatedif

Cltrnb cut

and ease out

~~~d

- - - --+

pull

toward cutter

workpiece

Conventional cut

FIGURE 244 Climb cut or down milling versus conventional cut or up milling for slab or face or end milling.

Cuttlng

forces oppose feed

Perlpherisl or slab ~nllling

Face or end

mllling

mtlling is to be used. It should never be attempted on machines that are not designed for this type of milling Virtually all modern m h g machmes are capable of down milling, and it 1s a most favorable applicatioa for carblde cutting edges. Because the material ylelds in approximately a tangential direction at the end of the tooth engagement, there is less tendency (than when up milling is used) for the machined surface to show toothmarks, and the cuttxng process is smoother,with less chatter. Another advantageof down milling is that the cuttlng force tends to hold the work against the machine table, permitting lower clamping forces. However, the fact that the cutter teeth strike against the surface of the work at the beginning of each chip can be a disadvantage if the workpiece has a hard surface, as castings sometimes do.This may cause the teeth to dull rapidly. Metals that readily workharden should be down milled, and many toolmakers recommend that down milling should always be the first chotce.

MILLING SURFACE FINISH The average surface finishes that can be expected on free-machining materials range from 60 t o 150 pin. Conditions exlst, however, that can produce wide vartations on either side of these ranges. For example, some inserts are destgned \nth wiper flats (short parallel surface behind the tool tip). If the feed per revolution [feed per tooth X number of teeth] of the cutter is smaller than the length of the wiper flat (the land on the tool), then the surface finish on the workpiece wtll be generated by the highest insert. In finishing cuts, keepmg the depth ot cut small wrll limit the axial cutting force, reducing vibrations and producmg a superlor finish. See Chapter 35 for discussions on measuring surface fimsh. Milling is an interrupted cuttxng process wherein entering and leaving the cut subjects the tool to Impact loading, cyclic heating, and cycle cuttingforces.As shownm Figure 24-7, the cutting force, F,, bullds rapidly as the tool enters the work at A and progresses to B, peaks as the blade crosses the directton of feed at C, decreases to D, and then drops to zero abruptly upon exit.The diagram does not indicate the impulse loads caused by mpacts.The interrupted-cut phenomenon explams tn large part why milling cutter teeth are designed to have small positive or negative rakes, partrcularly when the tool material is carbide or ceramic.These brittle materials tend to be very strong in compression. and negatrve rake results in the cutting edges being placed in compression by the cutting forces rather than tension. Cutters made from high-speed steel (HSS) are made with posltive rakes, in the main. but must be nm at lower speeds. Positive rake tends to lift the workptece, whiie negative rakes compress the workpiece and allow heavier cuts to be made. Table 24-2 summarizes some additional mllling problems.

.S AND

CUTTERS

Most milling work today is done with face mills and end mills. The face mills use indexable carbide insert tooling, while the end mills are either solid HSS or insert tooling (Figure 24-8). Basically, m i l l s are shank-type cutters liaving teeth on the circumferential surface and one end.They thus can be used for facing, profiling, and end milling. The teeth

Rotat~on 0"to 360"of cutter

SECTION 24.3 Milling Tools and Cutters

665

may be either straight or hehcal, but the latter 1s more common. Small end m~lls have 24-8). straight shanks, whereas taper shanks are wed on larger slzes (F~gure Plain end mrlls have mult~ple teeth that extend only about halfway toward the center on the end.They are used in mllllng slots, profiling, and fac~ng narrow surfaces. TwoIrp mzlls have two straight or helical teeth that extend to the center Thus they may be sunk into material, like a dr~ll, and then fed lengthwise to form a groove, a slot, or a pocket. Shell endmills are solid multrple-tooth cutters,similar to plam end m~lls but without a shank.The center of the face 1s recessed to recelve a screw head or nut for mouuting the cutter on a separate shank or a stub arbor. One shank can hold any of several cutters and thus provides great economy for larger-s~zed end mills. Hollow end mzllr are tubular rn cross sect~on, with tecth only on the end but havlng mtemal clearance. They are used pr~manly on automat~c screw machmes for slzlng cyhndrical stock, producing a short cylindrical surface of accurate diameter. Face mzlls have a center hole so that they can be arbor mounted Face mill~ng cutters are widely used in both hor~zontaland vertical-sp~ndle machine tools and come 1n a w~de vaslety of sizes (diameters and height?) and geometrles (round, square, triangular, etc.), as shown in F~gure 24-9 The Insert can usually be Indexed four timer and must be well supported. Either the power or the rlgldlty of the machme tool will be the lnnitrng factor. although sometimes setup can be the limiting factor.

Face mill

URE 24-9 Face mills come iany d~fferent des~gns uslng

ty differentInsert geometrles

d~fferent mountlng arbors.

Stub arbor mountlng

L Lock

L~ock

screw

screw

666

CHAPTER 24

Milling Radial relref angle

Peripheral cutting edge-

(staggeredtooth s~deshown)

FIGURE 24-10 The side-m~lling cutter can cut on slots or grooves However, only a few teeth are engaged at any one polnt in tlme, causlng I torsional v~brations. The average ch~p thickness, h,, will be less than the feed per tooth, - 6.1

feed per tooth ,f will be less than feed per tooth selected, F,, accord~ng to f , =

h,

JS

Another common type of arbor-mounted m~lling cutter is called a s~de cause it cuts on the ends and sides of the cutters Figure 24-10 shows the geom staggered-tooth side milling cutter. Staggered-tooth mrllzng cutters are narrow cylindrical cutters having st teeth, and with alternate teeth having opposite hellx angles.They are ground t on the periphery, but each tooth also has c h ~ p clearance ground on the protru These cutters have a free cutting action that makes them particularly effective I deep slots. Staggered-tooth cutters are really special scde-milling cutters,whic lar to plain milling cutters except that the teeth extend radially part way ac both ends of the cyl~nder toward the center. The teeth may be ellher straigh Frequently, these cutters are relatwely narrow, being disklike in shape Two or milling cutters often are paced on an arbor to straddle the workpleee (calle mllhng), and two or more parallel surfaces are machined at once. In Figure 24-11 msert-tooth side m~lls are arranged in a gang-milling sen three slots in the workpiece s~multaneously.Thus the desired part geometry i edly produced by the setup as the position of the cutter? IS fired H o v e v e r , , n s a face-milling operations only a few teeth are engaged at any point in tlme, rer heavy torsional vibrations detrimental to the result~ng machined pr0duct.A $a solve t h ~ problem s and in many cases be the key t o improved product~vity.

Soacers

nut

Arbor bearini surface for outboard support

FIGURE 24-11

Arbor (two views) used on a horizontal-spindle milling machine on left. On right, a gangmilling setup showing three side-milling cutters mounted on an arbor (A) with an outboard flywheel (B). For the gang milling as shown in Figure 24-11 the diameter of the flywheel should be as large as possible. (The moment of inertia increases with the square of the radius.) The best position of the flywheel is inboard on the arbor a1 A, but depending on the setup, this may not be possible, so then position B should be chosen. It is important that the distance between the cutters and flywheel be as small as possible. A flywheel can be built up from a number of carbon steel disks, each having a center hole and keyway to fit the arbor, so the weight can be easily varied. Interlocking slotting cutters consist of two cutters similar to side mills but made to operate as a unit for milling slots. The two cutters arc adjusted to the desired width by inserting shims between them. Slittingsaws are thin, plain milling cutters, usually from 'I,, to 3 / , 6 in. thick, which have their sides slightly "dished" to provide clearance and prevent binding. They usually have more teeth per unit of diameter than ordinary plain milling cutters and are used for milling deep narrow slots and cutting-off operations. In milling, the average chip thickness ( h J is not the same as the feed per tooth. For example, a thickness ( hJ of 0.004 in. corresponds to 0.012 in. feed per tooth in most side and face milling operations. If the radial depth of cut, d, is very small compared to the cutter diameter, D, use this formula:

(Note:For calculating the table feed, use halt the number ol inserts in a full side and face miU to arnve at the effective number of teeth.) table feed rate (ipm)
=

rpm X number of effective teeth x feed per tooth

Another method of classification for face and end mill cutters relates to the direction of rotation. A right-hand cutter must rotate counterclockwise when viewed from the front end of the machine spindle. Similarly, a lefr-hand cutter must rotate clockwise.All other cutters can be reversed on the arbor to change them from one hand to the other. Positive rake angles are used on general-purpose HSS millimg cutters. Negative rake angles are commonly used on carbide- and ceramic-tipped cutters employed in mass-production milling in order to obtain the greater strength and cooling capacity. TiN coating of these tools is quite common, resulting in significant increases in tool life. Plain milling cutters used for plain or slab milling have straight or helical teeth on the periphery and are used for milling flat surfaces. Helical mills (Figure 24-12) engage the work gradually, and usually more khan one tooth cuts at a given time.This reduces shock and chattering tendencies and promotes a smoother surface. Consequently, this type of cutter usually is preferred over one with straight teeth. Angle milling cutters are made in two types: single angle and double angle. Angle cutters are used for milling slots of various angles or for milling the edges of workpieces to a desired angle. Single-angle cutters have teeth on the conical surface, usually at an

Salid helical

;FIGURE 24-12 The chips are formed progressively by the ,teeth of a plain helical-tooth milling cutter during up milling.

angle of 45' to 60' to the plane face. Double-angle cutters have V-shaped teeth, both conical surfaces at an angle t o the end faces but not necessarily a t the same a The V-angle usually is 45", 6U", or 90". Furm milling curtm have the teeth ground to a special shape-usually an ir lar contour-to produce a surface having a desired transverse contour. They m sharpened by grinding only the tooth face, thereby retaining the original contou as the plane of the face remains unchanged with respect to the axis of rotation. concave, corner-rounding, and gear-tooth cutters arc common examples (Figure Solid HSS cutlers of simple shape and reasonably small size are usually moree ical in initial cost than inserted-blade cutters. However, inserted-blade cutters lowest in overall cost on large production jobs. Form-relieved cutters can be cost effective where intricately shaped cu needed. Solid or carbide insert tool cutters may need large volumes to be cost-jus by high-production requirements. Most larger-sized milling cutters are oI the inrerted-foofP1 type. The cutter bo made of steel, with the teeth made of high-speed steel, carbides, or TiN carbides, tened to the body by various methods,An insert tooth cutter uses indexable carb ceramic inserts, as shown in Figure 24-9. This type of construction reduces the a of costly material that is required and can be used for any type of cutter, but it i ofren used with face mllls.

Radiai

r t

I1

Radial rake angle (pos~t!ve shown)

Ax~al rellef angle

Tooth

Flute

FIGURE 2413

Sol~d form milling cutter rel~eved

Solid form relieved milling cutter, would be mounted on an arbor in a horizontal milling machine;

T-slot cutters are integral-shank cutters with teeth on the periphery and both sides. They are used for milling the wide groove of a T-slot,To use them, the vertical groove must flrst be made w ~ t h a slotting mlll or an end mlll to provide clearance for the shank. Because the T-slot cutter cuts on five surfaces simultaneously,it must be fed w ~ t h care Woodruff keyseat cutters are made for the single purpose of mllllng the semlcylindrical seats required in shafts for Woodruff keys. They come in standard sizes corresponding to Woodruff key sizes.Those below 2 in. in diameter have rntegral shanks; the larger sues may be arbor mounted. Occasionally,fly cutters may be used for face milling 01 bonng. Both operations may be done with a single tool at one setup.A single-point cuttmg tool 1s attached to a special shank,usually with provision Eor adjusting the effectweradlus of the cutting tool with respect to the axis of-rotation,The cutting edge can be made in any desired shape and, because it is a smgle-point tooi, 1s very easy to gr~nd.

24.4 MACHINES FOR MILLING

The four most common types of manually controlled mllllng mach~nes are listed below in order of Increasing power (and therefore metal removal capabillly):

1. Ram-type milling machlnes

2. Column-and-knee-type milling machines a. Horizontal spindle b. Vertical spindle 3. Fixed-bed-type millrng machines 4. Planer-type m~lling machlnes

Mill~ng machlnes whose motions are electronically controllcd are h t e d in order of increanngproduct~on capaclty and decreasing flexibil~ty:

1. Manual data input milling mach~nes 2. Programmable CNC nulling machines

3. Mach~ning centers (tool changer and pallet exchange capability) 4. Flexible Manufacturing Cell and Flexible Manufacturing System 5. Transfer lines

BASIC MILLING MACHINE CONSTRUCTION Most basic nulhng machines are of calwnn-and-knee constluctron, employing the components and motions shown 111 F~gure 24-14. The column, mounted on thc bace, is the main supportmg frame for all the other part&and contams the splndle with its drlv~ng mechanrsm. This construction provides controlled motion of the worktable in three mutually perpendicular directions. ( I ) through the knee, movmg vertically on ways on the front ol the column, (2) through the saddle, moving Lransversely on ways on the knee; and (3) through the table,moving long~tudinally on ways on the saddle. All these motions can be imparted by elther manual or powered means. In most cases, a powered rap~d traverse is provlded in addition to the regular feed rates for use in qettlng up wcrk and in returning the table at the end of a cut The ram-type mlllmg machine is one of the most versatile and popular milhng mach~nes,usmg the knee and column design. Ram-type machines have a head eqmpped with a motor-stopped pulley and belt drrve as well as a spindle.The ram, mounted on horizontal ways at the top of the column, supports the head and p e n t s positioning of the spindle wlth respect to the table. Ram-type milling mach~nes are normally 10 hp or less and suitable for l~ghtiluty milling, drrlling,reamlng, and so on (F~gure 24-15). Milling machines havlng only the three mutually perpendicular table motlons are calledplaln colwnn-and-knee type.These are available with both horizontal and vertical spindles (Figure 24-14). On the older horizontal-sp~ndle-type machmes, an adjustable overarm provldes an outboard bearing support for the end of the cutter arbor, which 1s shown in Flgure 24-11.These machines are well surted for slab, side, or straddle rnillmg.

\ f h '

-~levatin(l*-.

Dl

Knee-

FIGURE 24-14 Malor components of a pla~n column-and-knee-type milling machine, whlch can have horizontal gplndle a vert~cal sp~ndle, shown on the nght. The workp~ece and workholder on the tab11 on the left, or a turret type machine w~th with respect to the tool. be translated m X, Y, and Zd~rections

In some vertical-spindle machtnes the sptndle can be fed up and down, eitl power or by hand. Vertical-spindle machines are especially well suited for fac end milling operations. They also are very useful for drtlling and boring, partic where holes must be spaced accurately in a horizontal plane, because of thecont table mohon. Turret-typecolwnn-and-kneemilling machtnes have dual heads that can be sw about a horizontal axis on the endof a horizontally adjustable ram.Th~s permits n to be done horizontally, vertically, or at any angle.This added flexrbllity is advanta when a variety of work has to be done, as in tool and die or experimental shops are available with either plarn or universal tables. Unrversul column-and-knee mlll~ng mach~nes differ from plain column-an? machines in that the table is mounted on a housrng that can beswiveled in a hori plane, thereby increasrng its flexibility.Helices, as found m twist driUs,millingcutte hel~cal gear teeth, can be m~lled on universal machines.

FIGURE 24-15 The ram-type knee-and-columr

Vertical sptndle

ram-type

rn il'ng machine s one of the nlo,t versatile ana

popt. ar m:.lny rndcnine tools evzr de,ignerl.

Sxm~ 24.4

Machinex for Milling

671

BED-TYPE MILLING MACHINES In production manufacturing operafions, ruggedness and the capability of making h e a ~ cuts &re of more importancb than versatility Bed-type milling ma&ines (Figure 24-16)
are made for these eonditionsThe table i8 mounted directly on the bed and has only Iongitrrhal motion. The spindla head ean be moved vertically in order to set up the machine for a given operation. Normally, once the setup is completed, t h spindle ~ head is clamped in position m d no further motion of it occurs durmgmachining.Howwet, on some machines vertical motion df the sspindle occurs during each ~ycle. After such millingma~hines are set up, hUle skill ifirequlred to operate them, permitting faster learning time for the operators Some machines of this type am equipped with automatieeont~ols so that all the operator has to do is load and unload workpieces into thefkture and set the machinainto operation.Eor stand-alonemachines,afixture can be Io@& at each end of the table so that one workpiece can be loaded while another is being maobined. Bed-type milling machines with single spindles are sometimes called s~mplex mdling mac%nes; they are made with both horizontal and vertical spindles Bed-type machines alse are made in dllples and mplm typeq having two or thrm spindles r t h e e surfacesat a Single pass. respectively, permitting the simuItaneousmrlling of two o

PLANER-TYPE MILLING MACHINES

Planer-ipppe milling machrnes (Figwe 24-17) utilize several milling heads, which can removelarge amonnbof metal while permitting the table and workpiece to feed quite slowly, Often only a single pass of the workpiece past the cutters is required.Thmugh the use of different types of milling heads and cutters, a wide variety of surfaces can i t h a single setup of ihe worltpiece. This is an advantage when heavy be machined W workpieces are involvtd.

ROTARY-TAME MILLING MACHINES Sorne types o f face mllling in mass-production manufacturing are often done on rotaiyr t z b d b l e milling maehink8. Roughing and finishing cuts can be made in succession as the workpieces aremoved past The 'severalmilling cutterswhiie held in fiwtureson the rotating table The operafor can loadand unload the work without stopping the maehme.

FIGURE 24-16 Bed-type vertical-sprndle heavy-duty productron machine tools for m~llhngusually have three axes o f rnotlon.

FIGURE 24-17 Large planer-

type milling machine, Inset shows 90" head being used. (Courtesy of Cosa Corporation.)

PROFILERS AND DUPLICATORS

nected lo a precision servomechanism for each axis of control.The servos are co onto the workpiece. Dupliccsrars produce forms in three dimensions and are widely used to molds and dies. Sometimes these machines are called die-sinking machines. used extensively i n the aerospace i n d e w to machfhe paits Ram m @ p t stock as substitutes for forgings when the small number of parts required wo the cost of forging dies uneconomical. Mdny of these kinds of jobs are now NC- and CNC-type machines; their applications are discussed in Chapters 26.

MILLING MACHINE SELECTION When purchasing or using a milling machine, consider the following issues:
1 . Spindle orientation and rpm 2. Machine capability (accuracy and precision) 3. Machine capacity (size of workpieces)

5. Automatic tool changing

The choice of spindle orientation, horizontal or vertical, depends on th he machined. Relatively flat parts are usually done on vertical machines. Cu are usually done on a horizontal machine, where chips tend to fall free of

and the power available.

SECIION 24.4

Machines for Milling

673

Universal milling

FIGURE 24-18 End milling a

helical groove on a horizontalspindle milling machine using a universal dividing head and a universal milling attachment. (Courtesy of Cincinnati Miiacron, Inc.)

ks with all tooling applications,the tolerances that can be marntaincd in milhng are dependent upon the rig~dity of the workpiece, the accuracy and rigrdrty of the machine spmdle, the precision and accuracy of the workholding device, and the quahty of the cuthng tool itself. Mdling produces forces that contribute to chatter and v~bration because of the intermittent cutting action. Soft materials tend to adhere to the cutter teeth and make it more difficult to hold tolerances. Materials such as cast iron and aluminum are easy to rmll. With~n these criteria, properly marntained cutters used m rigid sp~ndles on properly fixtured workpieces can expect to machine w~thin tolerances with surface flatness tolerances of 0.001 in./ft Such tolerances are also possrble on "slotting" operations w ~ t h milling cutters, but +0.001 m. to +0.002in.ismoreprobable.Flatness specifications are more dlfttcult to maintain in steel and easier to maintain in some types of alummum, cast Iron, and other nonferrous mater~al Part size 1s the prrmary factor in select~ng the machine size, but the length of the tooling as mounted in the spindle must be cowidered. Horsepower requ~red at the spindle depends on the MRR and the materials ( u n ~horsepower. t hp,). Remember, coated inserts allow the MRR (the cuttlng speed) to be increased and available power may be exceeded Finally, the capaclty of the tool changers on machrning centers k limted by the number, size, and werght of the tools, especially if large-diameter tools are being employed.These often have to be stored m every other space in the storage mechanism.
ACCESSORIES FOR MILLING MACHINES The usefulness of ordinary m~lllng machines can be greatly extended by employing various accessories or attachments. Here are some examples A hor~zontal m~llrng machine can be equipped with a vertical milling attachment to permrt vertical m~lling to be done. Ordinarily, heavy cuts cannot be made with such an attachment. The un~versal millrng attachment (Figure 24-18) IS similar to the vertical attachment but can be sw~veled about both the axis of the milling machine spindle and a second, perpendicular axis to permit mllling to be done at any angle. The unrversal dividrng head is by far the most widely used m~llrng machine accessory,providing a means for holding and rndexing work through any des~red arc of rotation.The work may be mounted betweencenters (Figure 24-18) or held in a chuck that is mounted in the spindle hole of the dividing hcad.The splndle can be tiled from about 5" below horizontal to beyond the vertlcal position Basically, a dwiding head IS a rugged, accurate, 40.1worm-gear reduction unrt The spindle of the dividing head 1s rotated one revolutron by turning the mput crank 40 turns. An Index plate mounted beneath the crank contams a number of holes, arranged in concentrrc circles and equally spaced, with each circle having a different number of holes. A plunger pin on the crank handle can be adjusted to engage the holes of any

674

C H A ~ 24 R Milling

circle. This permits the crank to be turned an accurate, fractional part of a complete cle as represented by the increment between any two holes of a given circle on the i plate. Utilizing the 40:l gear ratio and the proper hole circle on the index plate, the dle can be rotated a precise amount by the application of either of the following rul number of turns of crank = holes to be indexed

40 cuts per revolution of work

40 X holes index circle cuts per revolution of work

If the first rule is used, an index circle must be selected that has the proper numb holes to be divisible by the denominator of any resultingfractional portion of a t the crank. In using the second rule, the number of holes in the index circle must be s that the numerator of the fraction is an even multiple of the denominator. For exam if 24 euts are to be taken about the circumference of a workpiece, the number oft of the crank required would be 121p A n index circle having 12 holes could be used one full turn plus eight additional holes. The second rule would give the same re Adjustable sector arms are provided on the index plate that can be set to a desired n her of holes, less than a full turn, so that Eractional turns can be made readily withou necessity for counting holes each time. Dividing heads are made having ratios o than 4O:l.The ratio should be checked before using. Because each full turn of the crank on a standard dividing head represe or 9" of rotation of the spindle, indexing to a fraction of a degree can be ob dexingcan be done in three ways. Plain indexing is done solely by theuse of the in the dividing head. In compound indexing, the index plate is moved forward orb ward a number of hole spaces each time the crank handle is advanced. For differe indexing the spindle and the index plate are connected by suitable gearing so that a spindle is turned by means of the crank, the index plate is rotated a proportional The dividing head can also be connected to the feed screw of the milling table by means of gearing. This procedure is used to provide a definite rotation o workpiece with respect to the longitudinal movement of the table, as in cutting he gears.This procedure is illustrated in Chapter 29.

I I Key Words
:climb (down) milling kolumn-and-knee milling machine conventional (up) milling cutting time end milling face milling insert-tooth milling cutter interrupted cutting machining center metal removal rate milling milling cutters milling machines peripheral milling slab milling staggered-tooth milling cu straddle milling Woodrull keysea

II Review Questions
1. Suppose you wanted to machine a cast iron with BHN of 275. The process to be used is face milling and an HSS cutter is 5. For producing Qatsurfaces in mass production machinin

going to be used. What feed and speed values would you select? 2. Explain how table feed (ipm) and spindle rpm are specified or computed for a milling machine after speed and feed per tooth are selected. 3. Why must the number of teeth on the cutter be known when calculating milling machine table reed, in in./min? 4. Why is the question of up or down milling more critical in horizontal slab milling than in vertical-spindle (end or face) milling?

does face milling differ basically from peripheral mill 6. Milling has a higher metal removal rate than planing. 7. Which type of milling (up or down) is being d Figures 24-1, Figure 24-2, and 24-77 8. Why does down milling dull the cutter more rapidly th milling when machining sand castings? 9. What parameters do you need to specify in order to cal MRR in milling? 10. In Figure 24-2b the tool material is carhide.What woul change in the process?

,11 What 1s the advantage of a helical-tooth cutter over a stratghttooth cutter for slab m~lhng? '12 What would the cutung force diagram for F, look hke if the cutter were perform~ng climb nulling? 13 Could the stub arbor-mounted face mlll shown m F~gure 24-9 he used to mach~ne aT-slot7 Why or why not? 14 In a typ~cal sohd arbor milhng cutter shown in Figure 24-10, why are the teeth staggered? (Check in Chapter 19 for d~scusslon of dynamics ) 15 Make some sketches to show how you would youset up a plaln column-and-knee mllhng machine to make it suitable for rnllhng the top and s~des of a large block.
I

16. Make somesketches to show how you would set upa horizontal m~llingrnach~ne to cut both sldes of a block of metal s~multaneously. 17. Explaln how controlled movements of the work in three mutually perpend~eulardirections are obtalned in columnand-knee-type milhng machmw 18. What n the bas~c pr~nc~ple of a un~versal divldmg head? 19. What is the purpose of the hole-clrcle plate on a universal dividrng head?

I Problems
milling machne. The &rect time to machine each piece on the shaper is 14min and on the d e r 166 rnin Labor costs $lOihr. facem~lling process. Reasonable values forfeed and speed are lkecharge for the use of the shaper 1s $lOhr and for the n?ilhng 0.010 in per tooth and 200 sfpm.The cuttet is 8 in. in diameter, as shown in Rgure 24-9. Campure the input values for the machine $2Whr-What is the breakeven quantity,below which the shaper 1s more econormcal fhan themill. machme tool. 2 How much time will be required to face mlll an AISI 1 020 8. In Figure24-12, the feed is 0 O O 6 in.per tooth The cutter is rotating at an rpm that will produce the desired surface cutting steel surface Bhn, 150, that is 12 m long and 5 in. wlde, using a 6-m-diameter, e~ght-tooth tungsten carb~de inserted-tooth speed of 125sfpm.The cutter diameter is 5 in.The depth of cut face mill cutter? Select values of feed per tooth and cutting is 0.5 in. The block is 2 in. wid%. a. What is the feed, in inches per mnnte, of the milling speed fromTable 24-1. machine table? 3 If the dspth of cut rs 035 in., what n the metal removal rate m b. What 3the MRR for this situation? Prnblem 2? 4 Estimate the power required for the operation of Problem 3. c. What is horsepower (HP) consumed by this process, assumingan 80% effiaeney and a HP,value for this maten& Do not forget to consider F~gure 24-7. 5 Examme the part shown in Figure 16.The slot on the left end al of l.8? must be produced by machining. Provide a process plan (a de9. Suppose you want to do the job described m Problem 6 by slab milling. You have selected a 6-in.-diameter cutter M k h scrrption (sketch] of how the part would set up in the machine eight TiN-coated earbide teeth.Thc cutting speed will be 500 for machining the slot and the details regarding cutting tools, sfpm and the feed pet' tooth will be 0.010 m.per tooth, Detersuch as material, slzes, and so on) Speafy (select) the type of mrllrng machme, the cutnng parameters, and any other informine theinputparameters for the machine (rpm of arbor and matlon needed to make this component. table feed), then calculate the T , and MRR. Compare these 6. A gray cast tron surface 6 in. wide and 18 m. long may be answers with what you got for face milling t h eblockwith HSS machmed on either a vertical milling machine, uszng en 8-in: teeth. diameter face millhavrug e~ght inserted HSS teeth,or on a hor- 10. The Bridgeport vertical-spindle milling machimz is perhaps wrth eight teeth the sin& most popular machine tool. Viually every factory zontal mllhng machms using an HSS slab m~ll on a4-in. diameter.Whlch machine has the faster cumng nme? (or shop) that does machining has one or more of these type 7 An operation is to be performed to machme $beegrooves on madme8 Goto your nearest machine shop and fmda Bridgea number of parts shown in Rgure 2411. Setup time is 20 minport, make a sketch ta show how n works, and explain what makes it so popular. utes an a shapm (not shown) and 30 rnrnutes on the hor~ontal
1 You have selected a feed per tooth and a cuttmgspeed for a

HSS versus Tungsten Carbide Milling

he K & C Machine works, which does job shop mach~n~ng, . , has received an order to make 40 duplicate pieces, made of AlSl 4140 steel, which will require 1 hour per piece of actual cutting time if a high-speed-steel (MI) milling cutter is used. Abigail Langley, a new machinist, says the cutting time could be reduced significantly if the company would purchase a suitable tungsten carbide milling cutter. Hugh Fellows, the foreman for the milling area, says he does not believe that Abigail's estimate is realistic, and he is not going to spend $450 (the current price from the vendor) of the company's money on a carbide

cutter that probably would not be used again.The ma hour rate, including labor for your shop is $40 per hot Abigail and Hugh have come to you, the supervisor o shop, for a decision on whether or not to buy the cutte which is readily available from a local supplier. What factors should you consider in this situation? much faster could the carbide cutter cut compared to HSS cutter? See reference table. Based on your best guess as to the savings in actual cutting time per piet who do you think is correct, Abigail or Hugh?

Material
Work
AIS14140 A1S14140 AIS14145

i"am
.
-

Tool

Back Rake (deg.)

Feed (ipt)
0.0104 0.011 0.015

Width (in.)
(1.100 0.15 0.25

Velocity (fpm)
100 540 560

Cutting (Ih).
360 540 1190

'&mi
190
156 560

HSS 0 Carbidc 0 Carhidi: O

JORKHOLDING DEVICES FOR

1 INTRODUCTION

25.9 C O ~ V ~ T I ~ O ;~ ~ ?~ UR AEL S
FIXTURE VERSUS NO

2 CONVENTIONAL FIXTURE DESTGN 25.10 MQDULARFIXIURING

3 DESIGN STEPS 3-2-1 Locatton Principle

4 5 6 7 8

25.11 SETUP AND C%~GEQVER Intermediate fig Concept CLAMPING CONSIDERATIONS 25.12 CLAMPS WORKHOLDING DEV~CES 25.13 OTHER CHIP DISPOSAL AssembIy Iigs UNLOADING AND LOADING TIME Magnetic Workholders EXA~~P OF LJIG E DESIGN ElectrostatlcWorkholders TYPES OF JIGS Vacuum Chucks

. .

25.1 INTRODUCTION
In the chapters on machining processes, the manner in which workp~eces are mounted and held in the vanous machlne tools was discussed. Workholding devlces, often called j~gs and fixtures, are critical components in the manufacturing of interchangeable parts. Workholders hold and locate the work m the machine tool w ~ t h respect to the cutting tool. For example, Figure 25-1 shows an engme lathe and a CNC turning center with two spindles, each holding a chuck that holds the workpiece III the correct location with respect to the cutting tools For many machine lools,f~xtures hold the workpieceswh~le prov~ding location with respect to the cutting tools. With workholders, process accuracy and precision (repeatability) can be achieved that o t h e m s e would be impossible wlth a glven combination of cutting tools and machine tools. In this chapter, workholdlng devlces (jigs and fixtures) wlll be considered as important production tools or adjuncts, with prlmdry attention bemg dlrected toward their functional characterist~cs, their relatronship to the machne tools, and the manufacturmg processes In recent years, workholding devices have become more flex~ble; that is, (1) able to hold more than one part and (2) able to be qulckly exchanged Flexlble workholders are critical elements in lean manufacturing cells, where components are made in families of parts (groups o i parts of slmilar design). Furthel, belng able to change from one devlce to anolher qu~ckly to accommodate different parts means smaller lot sizes can be run, wh~ch reduces inventory levels in plants.These flexib~lity requirements add signiftcantly to the complexity of convennonal jig and fixture design Let's begln with a discussion of the has~cs of jig and fixture deslgn

In the convent~onal method of fixture design, tool designers rely on their expenence and intu~tion to design s~mple, single-purpose fixtures for specific machining operation& often using a tr~al-and-error method untll the workholders perform ~atisfactor~ly. Of course, these designers should calculate the clamping forces or stress distributions in the fixtunng elements to make sure that the loads will not deform the fixtures or the workpieces elastically or plastically. In the design of the workholdmg devicas. two primary Functions must be considered: locating and clampzrlg Locatlngrefers to onenting and positioning the part in the machine tool wtth respect to the cuttmg tools to achieve

678

CHA~E 25 R Workholding Devices for Machine Tools

Spindle orientation can be fixed during cutting

FIGURE 25-1 Two kinds of lathes. Above, a conventional engine lathe with principal parts named, including the workholder device, the chuck. Below, a CNC turning center with two chucks, turrets for cutting tools, and C-axis control for the main spindle. The C-axis control, on the spindle, can stop it in any orientation so the powered tools can operate on the workpiece.

SEcnoN 25.2 Conventional Fixture Design

679

C
i i i
AL

I \
6
C

-iaF

II

.g
. . --. .

-gh

. -

. -

-If1

locuiiotz. Size dimensions denote the slze of geometrical shapes-holes, cubes, paralleleprpeds, and so on-of which oblects are composed. Locat~on dimensions, on the other hand, determine thc posillon or location of these geometrical shapes wit11 respect to each other.Thus a and c in Figure 25-2 are lotiatlon dlmensions, whereas e andg are size dimensions With location dimellsions in mind, one can precisely define a jig as follows: A jig o a specla1 workholding device that, through built-in features, determmes operatzonr The key locat~on dcmenscom that are produced by machincng of fasten~ng requirement of a j ~ g is that rt determine a location dimension. Thus, jigs accomplish layout by means of their deslgn In order to establ~sh location dimensions,~ i g may s do a number of other thmgs. They frequently guide tools, as in drill jigs. and thus determne the location of a component geometrical shape.However, they do not always gulde tools In the case of welding jigs, component parts are held (located) in a desired relationship w ~ t h respect to each other while an ungulded tool accompl~shes s the fastening. The guiding of a tool 1 not a necessary requirement of a jrg. Sim~larly, jlgs usually hold the work that IS to be machmed,fastened,or assembled. However, in certain cases, the work actually supports the jig. Thus, although a ~ i g may ~ncidentally perfom other functions, the basic requirement is that, through qualities that arc built into it, certain cntical dimenmons of the workp~ece are determmed. A fixture 1s a speczal worl<hoM~ng devrce that holds work during machcning or assembly operatrons and establzshes size dzmenslonr. The key characteristic is that it is a special workholdlng device, designed and constructed for a particular part or shape. A general-purpose device, such as a chuck in a lathe or a clamp on a milling machine table, is usually not considered to be a f~xture. Thus a fixture has as its specific objective the fac~litating of ~ e t u por , mak~ng the part holding easier. Because many jigs hold the work whlle determining critrcal location dlmensions, they osually meet all the requlrement9 ot a Gwture Alternati~ely~many bxtures are used in NCmachines holdmg parts where holes are located and drilled according to a program. So the strlct definition of jigs and fixtures has been blurred by the changes ID technology. In designing workholders the des~gner must conslder whether the part IS a casting, forging, or bar stock. With castnlgs and forgings variations in shape and size must be accommodated m the deslgn, and usually a machinmg operationis required to establish a reference surface (called the datum rurface) to aid initlal fixturing. In parts cut from bar stock, allowances must he made for inaccurac~es and irrega summary of design ularitles produced by the cutoff operations. Table 25-1 prov~des criteria for v~orkholders for you to review. Obviously, it is impossible to meet all these design criteria for workholders. Compromise is mevitable. Stil1,it IS useful to know the optimal deslgn objectives to illustrate the positioning, holdmng, and supporting functions that fixtures must fulfill.

680

C m m 2s Workholding Devicer for Machine Tools

Reparability identreal workp~eces should be lo~ated by t h workholder ~ in precrsely the same space on repeated loadmng and unloadng cycles should be mpossible to load the warkpieccmcarrectly This is called ''t~dproofrng'' rhe jig or bxture Adeqnuce clampingformThe workholder must hold the workpleee 1mm6bde against the forces of gravity, centr~fuyal forces,mertial t~rces,

tain and lubncate.

Lawprofile Workholder elements must be clear of the cuttlng tool path. D e s i p n g l u g s m the part for clamp~ng can slmpllfy the fixture andall proQer tool olearancc

Erlpommim undsafezy Clamps should he seleQed and poslnoned to elim~nate pinch points and facibtate ease of operation The workholder meets shnuld not obsm~ct the loadma or unloadmg of workpieces In manual operations tlle operator should nat have to reach past the tool load or unload pmts A rule xlmeumes used is that the operator m repeatedly exert a force nf 30 to 40 1b to open or close a clamp but gee fo~ses than this can cause ergonomic problems

The classical deslgn of a workholder (e.g., a dnll pg) mvolves the followmg stepx

1 . Analyze the drawing of the workpiece and determine (visualize) the machining eratlons requ~red to machine it. Note the critical (slze and location) dimensions
2
3.

4.

5 .

6.

tolerances. Determine the orlentatlous of the workp~ece in relation to the cutting movements of the tools and tables. Perform an analysis to estimate the magnitude and direction of the 21). (see Chapte~ Study the standard dev~ces ava~lable for workholders and for the clam Can an off-the-shelf device be modified? What standard elements can be used? Form a mental picture of the workplece in posttlon in the workholder in th tool with the cutting tools performing the requited operationls). See the chapters on machining for examples Make a three-d~mensional sketch of the workp~ece m the workholder in its re position to determme the location of all the elements. clamps, locator buttons, mgs, and so on

3-2-1 LOCATION PRINCIPLE After determirung the onentation of the workp~ece in the workholder, the next st locate it in that position.Thls location IS also used for all slmilar workpieces.'Ihe d must select or deszgn locating dev~ces (supports) t h t ensure that ever), workp~ece

SECTION 25.3 Deslgn Steps

681

Three-polnt support locatlon in any one plane

Two additional

pornts determ~ne the second plane

One add~t~onal

pornt determines

Clamp close-near support iocatlons

FIGURE 25-3

Workpiece location is based on the 3-2-1 princ~ple. Three points will define a base surface, positively two polnts in a vertical plane will establish an end reference, and one point in a third plane w~ll locate most parts.
1~ the devlce o~cupzrs the same position wlth respect to the cutting tools. Thus, when the machlnrng operation is performed, the workpieces are processed identically. This is, of course, the key to making interchangeable parts In locating the workpiece, the bas~c 3-2-1 prcnclple of location 1s used (Flgure 25-3). For posltive locat~on, the fixture must posltton the workpiece in each of three perpendicular planes. Positlonlng processes can vary greatly, but workholder design always begins by defining the frrst plane of reference with three points. Once the object is defined in a single plane, supported at three points (hke a three-legged stool on a floor), a second plane can be ass~gned that is perpendicular to the first.To do this, the object is brought up against any two points in the second plane. To continue the example,the stool IS slid along the floor until two legs touch a wall. A third plane, perpendicular to each of the other two, is then defrned by designating one po~nt on it. As long as an object is IU contact with three points on the first plane, two po~nts on the second, and a single point on the third, it is positively located in space.7he location polnts within each plane should be selected as far apart as possible for maxlmum stabihty In practice, it is often necessary to support a workpiece on more points than this 3-2-1 formula dictates.The machining of a large rectangular plate, for example, typically requires support at four or more points. However, any extra polnts must be established carefully to support the workp~ecein a plane defined by three-and only three-points. Appropriate clamping dewces are selected so that the clamping forces hold the workpiece ~nthe proper location and reslst the effects of the cutting forces, centrifugal forces, and vibrations. Epossible, the machining forces should act into the l o a t ~ o n points, not Into the clamps, so that smaller clamps can be used. In reality, the worker often determines clamplng force when loading the part lnto the workholder. Fixtures are usually fastened to the table of the machme tool.Although used primarily on mllliig and

'ed 3h-

682

Devices for Machine Tools C m a r e ~ 2 5 Workhc~Idtng

I. Dec~de an i ~ c a t ~ points n g and clamping arrangemmh 3. For mating parts, use corresponding locatmg pornts or surfaces to ensure proper ahgnment assembled 4 . 2 ta ~ uuse 3-2-1 lucal~on, w~th 3 asslgned to large71 surrace Addrt~onal polnts should be adjusr
S.Locat~ng points should be vlsible so that the aperator can see rf they areclean Can they be r If worn? 6. Prov~de clamp& thal are asquck actmg and easy to use as a eoouomtcally jushfiable fur fap~dl and unloading 7. Clamps should not require undue effort hy the opcrdtor to douse or to open, nor should they hands and fingers during use 8 Clamps ,shouldbe mntegral parts O f devEce.Avoid loose parts that can get lost

11 T a l c e the t h w t of the outung forcesos thelacaters (if poss~ble), not on thedamps 12&range the workholder sothat theworkpiececan easily be loaded and unloaded from thede so that rt can beloaded onlyin the carradtrnarmar (mistake-praof) and m such a way that the can be found quickly (vmually) 13 Comutent with strrnglh dnd rig~d~Iy,make the workholder aslrght as possible 14 Provide ample room fat chrp clearance and rerswal 15 Pwvide accesstbdftyfor clqaning 16 Provldefor entrdhce andexlt of cuttingflutd (which may carry offchtp?) iconex to beused 17 Provlde four feet an all movableworkholders 18.Provide hold.down lugs on all fixed workholders.

DNSIDERATIONS

regularities that may produce force components in an undesired direction. Co clamp~ng forces should be applied in directionsthat will assure that the work in the desired position. Whenever possible,jigs and Muresshould be designed so that the forces i by the cutting process act to hold the wotkpiece in pasitlon against the supports.

Exaggerated ~llustrat~on o f the manner in \%n~ch excesvve r amplnq force, can alfertthe I naI dlniensions of a workpiece.
FIGURE 254

Overclamped before m a c h ~ n ~ n g . part distorted

After m a c h ~ n ~ n g (st111 clamped)

S F C ~ 25.5 I ~ N Ch~p Disposal

683

F~xture
Good
(a)

Not good

Good

Not good

(b)

Down mlll~ng good

Good

(cl

Better

FIGURE 25-5 In (a) and (b), proper work support to reslst the forces imposed by cuttlng tools is forthe work to rest on. demonstrated. In (c), three buttons form a tr~angle

forces are predictable, and proper utilization of them can materially aid in reducing the magnitude of the clamping stresses required. In addition to locating the work properly, the stops or work-supporting areas must be arranged so as to provide adequate support against the cutting forces.As shown in Figure 25-5a, having the cutting force act against a fixed portion of the jig or fixture and not against a movable section permits lower clamping forces to be used. Figure 25-5b illustrates the principle of keeping the points of clamping as nearly as possible in line with the action forces of the cutting tool so as to reduce their tendency to pull the work from the clamping jaws. Compliance with this principle results both in lower clamping stresses and less massive clamping devices. Don't forget that down milling produces different forces than up milling. The location points should be as far apart as possible but positioned so as not to allow the cutting force to distort the work, as shown in Figure 25-5c. The cutting forces may distort the work, with resulting inaccuracy or broken tools. These design suggestions materially reduce vibration and chatter during the cutting process. As many operations as are possible and practical should be performed with each clamping of the workpiece.This principle has both physical and economic aspects. Because some stresses result from each clamping, with the possibility of accompanying distortion, greater accuracy is achieved if multiple operations are performed with each clamping.From the economic viewpoint, if the number of jigs or h t u r e s is reduced, less capital will be required and less time will be spent handling the workpiece loading and unloading.

or fixtures are used in connection w ~ t h chi-makmg operations, adequate When j~gs provision must be made for the easy removal of the chips This 1s essential for several reasons. Frrst, if chips become packed around the tool, heat will not be carned away and tool life can be decreased. Figure 25-6 Illustrates how insufficient clearance between the end of a drill bushing and the workpiece can prevent the chips from escaping, whereas too much clearance may not provide accurate drill guidance and can result in broken dr~lls.

684

CHAPTER 25

Workholding Devices for Machine TOOIS

hing

Drill

Proper clearance between drill bushing and tool of workpiece is important.

FIGURE 25-6

Too much clearance permits tool drift

F s s G T m

too little clearance

A second reason chips must be removed is so that they do not interfere proper seating of the work in the jig or fixture (Figul-e25-7). Even though chips always have to bc cleaned from the locating and supporting surfaces by a worke automatic means, such as an air blast, the design details should be such that c other debris will not readily adhere to,or be caught in or on, the locating sur or overhanging elements and thereby prevenl the work from seating properly condition results in distortion, high clamping stresses,and incorrect workpiccc dim

1%1 25.6 UNLOADING AND LOADING TIME

The cost of the workholders must be justified by the q and their primary purpose is to increase productivity and quality. While put into or being taken out of jigs and fixtures the are not making chips. The loading and unloading called the run time) plus any delay times equals th and unloading time is greatly influenced by the choice of clamps. There are many ways in which jigs and fixtures can be made easi unload.Some clamping methods can be operated more readily than ot ple, in the drilljig shown in Figure 25-8, a knurled clampingscrew is used t against the buttons at the end of thejig.To clamp or unclamp the block in this dir requires several motions. On the other hand, a cum lurch is used to close the jig an the workpiece against the rear locating buttons.This type of latch can b a single motion. Certainly, the deviceshould be designed so that the part cannot rectly. Defect prevention is often accomplished by the clamping dcvi loaded improperly cannot be clamped. Ease of operation of workholders not o rectly increases the productivity of such equipment but also results indirectly in quality and fewer lost-time accidents. The workholder is as critical as the machine tool and the cutting tool tot quality of the part.Thc use of the workholder eliminates manual layout of the features of the part on the raw material. Manual layout requires a highly skilled and is vely time consuming. The workholder permits a less skilled person to a quality and repeatable production with far greater efficiency.

ILI 25.7 EXAMPLE OF JIG DESIGN

Several principles of work location and tool guidance are illustrated in two mounting holes in the base of the bearing block are to be located and drilled. mensions A, B, and C are determined by the jig. While it is not specified on the dr

Methods of prov~d~ng ch~p clearance to ensure proper seatlng of the work.

FIGURE 25-7

.
Chips

,
Good

Notgood

SECTION 25.7

Example of jig Design

685

Mounting hole

FIGURE 25-8

(Lower left) Part to be drilled; (lower r~ght) box drill jig for drilling two holes; (upper left) j~gin drill press; (upper r~ght) drill being guided by drill bushing.

there is one other location dimension that must be controlled.?he axes of the mounting holes must be at right angles to the bottom surface of the block. so the bottom must be machined (milled) prior to this drilling step. The way in which the part dimensions are obtained in the finished workpiece is as follows.The surfaces marked with a large carats b' are reference (or location) surfaces and are finished (machined) prior to insertion of the part into the drill jig.The part rests in the jig on four buttons marked X in figure 25-8.These buttons, made of hardened steel, are set into the bottom plate of the jig and are accurately ground so that their surfaces are in a single plane.The left-hand end of the part is held against another button Y.This locating button is built into the jig so that its surface is at right angles to the plane of the

686

for Machine Tools CHAITER 25 Workholding Dev~ces X buttons. When the block is placed in the jig, its rear surface rests against three mo buttons marked Z.These buttons are located and ground so that their surfaces lie I plane that is at right angles to the planes of both the X and Y buttons. The part is h in its located posltion by the two clamps marked C. The use of four buttons on the bottom of t h s jig (X buttons) appears not to here to the 3-2-1 pciuclple stated previously. However, although only two X butt would have been required for complete location, the use of only two buttons would have provided adequate support during drilling.The thrust from the drills would h d~slodged the part from the 1ocatb1sThus the 3-2-1principle is a minimum concept often must be exceeded. To ensure that the mounting holes are drilled in their proper locations, the d must be located and then guided during the drilling process. This is accomplished by two drill bushmgs marked K. Such drill bushmgs are accurately made of hardenedst with their inner and outer cylindrical surfaces concentric. The inner diameter is slightly larger than the dnll-usually 0.0005 to 0.002 in.-so that the drill can turn but not shift appreciably. The bushings are accurately mounted in the upper plate jig and positioned so that their axes are exactly perpendicular to the plane of the X tons, at a distance A from the Z buttons and at distances B and C, respectively,fr plane of the Y button. Note that the bushings are sufficiently long that the drill is close to the surface where it will start drilling. Consequently, when the properly placed and clamped m the jig, the dnll will be located and guided by the so that the critical lmensions on the workpiece will be correct. The right hole drilled in a vertical-spmdle drill press (not ~unning), and then the jrg will be (manually by the operator) to the right, and the left hole will be drilled. The construction is rigid but open for chip removal.

Jigs are made in several basic farms and carry names that are descriptive of their configurations or predominant features Several of these are illustrated m F~gure Aplatelig is one of the simplest types, consisting only of a plate that contai drill bushings and a simple means of clamping the work in the jig or the jig work. In the latter case, wherein the jig is clamped to the work the device is so called a clamp-on jig. Such jigs are frequently used on large parts, where it is to drill one or more holes that must be spaced accurately wtth respect to e or to a corner of the part, but that need not have an exact relationship w ~ t portions of the work. Channel jigs also are simple and derive their name from the cross-seckonal of the main member.They can be used only with parts having fairly simple shape Rtng jigs are used only for drilling round parts, such as pipe flanges. The cla force must be sufficient to prevent the part from rotating in the jig. Dtameterjrgs p ameans of locating a dnlled hole exactly on adiameter of a cylindrical or spheric Leafjigs derive their name from the hinged leaf or cover that can be swun to permit the workpiece to be inserted and then closed to clamp the work in positio bushings may be locatedin the leaf as well asin the body of the jig to permit locatin drilling holes on more than one side of the workpiece. Such jigs are called rollover tumbleligs when they require turning to pennit drilling from more than one side. Box jigs are very common, deriving their name from their boxlike constr They have five fixed sides and a hinged cover or leaf, which opens to perm workpiece, and a cam that locks the workpiece in place. Usually, the drill bu cated in the fixed sides to ensure retention of their accuracy.The fixed sides of are usually fastened by means of dowel pins and screws so that they can bet and reassembled without loss of accuracy. Because of their more complex co box jigs are costly, but their inherent accuracy and strength can be justified when is sufficient volume of production. They have two obvious disadvantages: (1)1 ly more difficult to put work into them than into simpler types, and (2) there is tendency for chips to accumulate within them. Figure 25-8 shows a box-type ]ig

SECTION 25.8
4 Bushings

Types of Jigs

687

Bushing

Work

Plate jig

. . -

----

Diameter jig

Channel jig

Leaf jig

Because jigs must be constructed very accurately and be made sufficiently rugged so as to maintain their accuracy despite the use (and abuse) to which they inevitably are subjected, they are expensive. Consequently, several methods have been devised to aid in lowering the cost of manufacturing jigs. One way to reduce this cost is to use simple, standardized plate and clamping mechanisms called universaljlgs (Figure 25-10),

CHA~E 25 R Workholding Devices for Machine Tools

FIGURE 25-10 Two types of universal jigs are manual (bottom) (center). A completed jig (on the top) made from unit right below. Universallig Division, The Industrial Machine Company.)

*~(~+-~--=-+"=+-,-

These can easily be equipped with suitable locating buttons and drill bushing; struct a jig for a particular job. Such universal jigs are available in a variety of rations and sizes, and because they can be produced in quantities, their cost is r low. However, the variety of work that can be accommodated by such jigs obr, limited. While the drill bushing should be spaced far enough from the work chips to escape without entering the bushing, when drilling into an angled sur bushing should be very close. Once the drill has penetrated to at least one-half ol diameter, the bushing should be retracted to provide chip clearance. Design of jig must not obstruct coolant flow to where it is needed. Bushing length shoul~ to 2.5 times the drill diameter.

11 25.9 CONVENTIONAL FIXTURES

Many examples of conventional fixtures have appeared in the text. Productron broaching, and boring processes as performed on NC machmes, conventronal ment, or machining centers routinely use fixtures to locate and hold the part pr with respect to the cuthng tools on the machine tool. Like cutting tools, toolingfor holding is sold separately and IS not usually supplied by the machine tool builde dit~onally,beg~nningwlth Eli Whitney, manufacturers have deslgned and custom-made, dedicated fixtures. Because of the pressure of shorter production and smaller lot sizes,many companies are turning to modified hturing approache greatest advantage of these systems is that the fixture can be constructed quickly. Perhaps the most common fixture uses the vrse as its base element. Figure shows a schemat~c and photo of a typical commercially available vise that can be ad for use as a fixture. As shown, the vise jaws are readily modified t o conform 3-2-1 location principle and provlde adequate clamplng forces for almost machining operation. Four vises (also shown in Figure 25-11) can be mounte subplate for rapid insertion and location in the machine, or four vices can be mo on a tombstone for milling parts in a CNC machne.

Fixed jaw

Movable jaw

Conventional or standard vise

Grmoued jaws (standard jaws) wiVlimaIier surfam for inoreasing the

Gpecificsumoe

sensitive clamping surfaces.J a n

roundworkpieoes. Rolleijaw "Dullb'

with removable jaw plates to FIGURE 25-11 The conventional orstandard vlse (top left and nght) can be mod~f~ed adapt to d~fferent part geometries. These vices can be integrated Into ~nllllng ftxtures (right middle and bottom)

The chucks used in lathes are really general-purpose f~xtures for rotational parts. Newer chuck designs have greatly improved their flexibility (the range of diameters the chuck can accommodate in a glven setup and speed of setup). F~gure 25-12 ~ h o w a s complete change of top jaws for a three-jaw chuck belng done In less than 5 mnutes.The normal time for thls part of the setup might exceed 15 muutey New qu~ck-change ~nsert top jaws may even snap in by hand with no jaw nuts, keys, screws, or tools. Jaws that can be exchanged by robots can also be designed

C H ~25R Work holding Devices for Machine Tools

1.

Pre-asembled Mint-System

2.

and top jaws.

Assembly being inserted into the Master Jaw.

3.

Quickly retighten 'cap screw.

4.

A11 3 jaws changed in.

minutes or

less.

~gof the top jaws on a three-jaw chuck. (Courtesy of Huron Machine Pfodarcts.)

Movt pruducers of' chucks use some variation of equation 25-1 to comput maximum rpm rate at which the chuck can run:
=

F,,,
3 X (2.84 X 10-0 X W X D

where
S,
=

maximum rpm value at which gripping force equals '13 F,, rate gripping force, at rest (lb) weight of jaws (Ib)

F,
W
D

= maximum = combined = distance

from spindle centerline to centeT of jaw mass (in.)

Thus,with this equation,a 10-in. power chuck with a published rating F,,, of 13, would retain one-third of its initial gripping force at 2507 rpm. (Check this calcul using W = 8 lh, D = 3.1 in.) The higher the rpm value, the greater the centrifugal factor.This is an important factor in high-speed machining operations in which the is rotating.

FIXTURING
Modulur,firturrs have all the same design criteria as those of conventional fixtures, more-versatility. Modular fixture elements must be useful for a variety of m applications and easily adaptable to different workpiece geometries. Individua designs can be photographed or entered into a computer-aided design (CAD) lib future reference.After the job is done, the fixture itself can be dismanlled and the e returned to the toolroom.The erector-set approachuses eitherT-slot or dowel-pin Figures 25-13. and 25-14 show two examples of modular fixturing.The designs be base plates. Elements for locating and clamping are added to the subplate. Rect square, and round are the typical patterns for the subplates. Also shown are th components for modular fixturing systems used for mounting points,locators, atta and so on.The standard elements needed to construct the fixture include riser W blocks, angle plates,cubes,box parallels,and thelike.Smaller elements such as loc supports, pads, and clamps are added to the subplate on the Larger structural Mechanical clamping devices are shown,but power-assisted clamps are available. and fixturingelementsaremade to tolerances of +0.0002 to 0.0004in.in flatness,para and size. Figure 25-14 shows a part in a dedicated fixture compared to a modular The dedicated fixture represents a capital investment that must be absorbed by the must be maintained after the job is complete.The modular fixture is disassembled elements reused later in fixtures for other parts Modular fixtures are commonly use prototype tooling and small-batch production runs. They are being incorporated' frequently into regular production as users gain confidence in this approach.

Modular ns with a subplate i d adds locators

Dedicated left versus modula right. (From g Engineering,

4
.)

Every part coming out of the workholder should be the same. resulting in interchangeable parts. But what al)out the first part? Does it meet specilications? What about the initial setup of the workholdcr into the machine tool? In nlany cases, the setup operation takes hours and the machine is not producing anything during this time. Kapid exchanges of workholding devices is a key technique in modes11 manufacturing systems. (See Shingo, 1085 for more discussion of the elimination of setup and SMED.) Reducing setup times p e r ~ l ~ isl~orter ts production runs (smaller lot sizes). D o not confuse initial setup (of workholders) with part loading and unloading or tool changing.The trick with initial setup is to do it quickly and to get the first pal-t out of the process as a good part, with n o adjustment of the machine, the tooling, 01.the workholder. Quick tool and die exchange is a critical component in the strategy for the factory with a future.

CHAPTER 25

Workholding Devices for Machine Tools

FIGURE 25-15

Master jig designed for a family of similar components. (a) Part family of rounds plates (six parts, A-F); (b) group jig for drilling, showing adapter and part A.

Ibl

Another approach to rapid setup is shown in Figure 25-15, where instea different jigs, a master jig is made (also called a group jig) for a family of components and then a set of adapters is made that customizes the jig for eac the part family.This concept of group jigs and fixtures originated from the gro nology (GT) concept for master jigs as a method to form cellular manufacturing by determining a family of parts where an imaginary part, called the composite. designed that has all the key features of all the parts in the family. In other wo. composite part is an envelope, the shape of which encompasses the shapes of parts in the family. The theory is that if the tooling is designed for the compos' any part that fits within the envelope could be machined without any tooling c This part is used for designing the workholder.The workholding devices should to accommodate all the parts within the parts family. For manufacturing cells, th holders will also have to compensate for variation in cutting forces, centrifugal

and so on. Group workholders are designed to accept every part-family member, with or without adapters, that accommodate minor part variations.

INTERMEDIATE JIGCONCEPT One way to achieve rapid fixture exchange is to employ the intermediate jig concept.This means that the workholding devices are designed so that they all appear the same to the machine tool but different to the partsThis usually requires one to construct intermediate jig or fixture plates to which the jig or fixture is attached. The jigs or fixtures are all different, but the plates are all identical. The cassette tape for your VCR is an example of an intermediate workholder.To the VCR, every cassette appears to be the same and can be quickly loaded and unloaded with one bandling-that is, one touch. From the outside, every tape appears to be the same, but on the inside, every tape is different. I f you think about the workholding devices in terms of the intermediate jig concept, you can quickly achieve one-touch setups Figure 25-16 shows an example of the intermediate jig concept, applied to lathes and chucks. An adapter or intermediate fixture is bolted l o the lathe's spindle and is a permanent part of the machine tool. The intermediate fixture will acccpt mating chucks that have been preset for the workpiece prior to insertion. Different chuck designs mount interchangeably on the common actuator. This method greatly reduces setup time and permits the operator to perform chuck maintenance and retooling (setup) while the machine is running.The chucks can be exchanged automatically. Quick-change fixtures for CNC milling machines and machining centers (using the intermediate jig concept) are now available commercially.

Lathe

s~indle

Intermediate fixture-actuator

Chuck

Workpiece

Three types

Actuator

Diaphragm chuck

Coliet

Jaw

chuck

chuck

FIGURE 25-16 Example of the intermediate jig concept applied to lathe chucks. The actuator is mounted on the lathe and can quickly adapt to three different chuck types. (Courtesy of Sheffer Collet Company.)

694

CHArrLR

25 Workholding Devices for Mach~ne Tools

(b)Strap clamps

Handle

Spherical nut

( c )Toggle clamp

fast

ol Od\lC type$ FIGURE 25-17 f k ~ n l p ! r s ot clalnar dsrd for t\ordnolrllnq. Trle clamp elements come in a wide variety of sues.
'

111 25.12 CLAMPS

When designing a jig or fixture, there are many choices to be made regarding the cl Manual clamps which include screw, strap, swing, edge, cam, toggle, and C-clam with certain strengths and weaknesses, are usually cheaper but slower. In Figu typical types of clamps that are used in fixtu many forms and sizes and is simple, low cost, a hand knob, a cam, or a wrench turning do commodates only small thickness variation consistent clamping force. Power-actuated clamps (shown in Figure 25-18) provide more consistent forces than do manual clamps, especially in applications that promote operat The higher cost must be weighed against the capability for consistent and r operation, automatic adjustment of holding forces, remote actuations, and au sequencing of clamping actions. Extending of a manual clamp-strap assembly.They ext Edge clamps have a very low profile.They

1O 25.13 OTHER WORKHOLDING DEVICES

ASSEMBLY JIGS
Because assembly jigs usually must provide for the introduction of several co parts and the use of some type of fastening equipment, such as welding or ri calnmonly are of the open-frame type. Such jigs are widely used in autom welding and aircraft assembly. Large jigs of the type are shown in Figure 25for the assembly and usually feature automatic clips. This jig is constructed m reinforced concrete.

SECTION 25.13 Other Workholding Devices

End mill

695

Workpiece
Clamp

Face mill

Edge clamp
I I
%

Fixed

, I,--, I
\ . _ ,

U!4E 25-18 Examples of er-clamping dev~ces:

Workpiece

MAGNETIC WORKHQLDERS
Because of the light cuts and low cutting forces, workpieces can be held in a different manner on surface grinders than on other machine tooIa Magnerrc chucks are used for ferromagnetic materials.To obtain high accuraq,it is desirable to reduce clampingforces and distribute them over the entire area of the workpiece.Also, grinding is eequently done on quite thin or reIative1y delicate workpieces, which would be difficult to damp by normal methods In addition, there is often the problem of grinding a number of small, duplicate wv~kpieces. Magnehc chucks solve all these problems very satisfactorily. Magnetic chucks are available in disk or rectangular shape6.D~-disk rectifiers are used to provide the necessary direct-current power. Some magnetic chucks utilize permanent magnets, as shown in Figure 25-20, and can be tilted so that angles can be ground. Magnetic chucks provlde an excellent means of holding workpiece$ providedzhat the

696

G m m 25 Wo~hold~ng Device:, lor Machine To&

FIGURE 25-20 Example o f magnetic chuck. (Courtesy o f 0 5. Walker)

cutting or incrtial forces are not too grcat.'lh~, holding lorce IS distributed over I lire contacl ~ ~ ~ r f of i~ thc cc work, thz clanlping strcqscs arc low,,md rhcrefore {here tie tenden? for the work tc bedistarted. Consequently, pieces can be held and accurately.Aha, a number of small pieces can be mounted on a ehutk and gmund same time. Mametic chucks pruvi.de areat mt-to-part repeatability became ing pirwer fro& one pwt to the next isthe same. initial setup i$ usukly fast, sitnpl relatively inexpensive. Pats lmdlng and unJoading is alsa relaitwly easy. It oflen isnecwsaq to demag~etiza work that has been held ona magnetic Some electrically pawered chwks provide satisfactory demagnetization by rev the direct currmt briefly whm the gower isshut off.

ELECTRaSTATIC WORKHOLDERS Magnetic chucks daq be used only with ferromagnetic materials: Electrostatr can be used with any electrically conductive material lhis principle (Fi rwts that work be held by mutually attracting eiectmtatic Yields in the
workpiece. Thrse pmide a holding force of up ta 20 psi (21,WO Pa). u,ually bc held it thes are ~lnshed(i.c.,conted) witha thin layer of rnetal.lhise hare thc added :icivmtagc of not inJucing rcsidual mngn-tisrn in thr. work. VfiCUUM CHUCKS Vacuum ckueRs are alsa available. In nnetype, illnstratsd in Figure Z-22, the h the work plate aremnnecred to a vacuum pump and can be opened cx closed bp of valve serrnsThnvalUes are opened in the &ea on which the *ark is to rest.Tht type has a porgus plate on which the work rest%The w~rkpiece and plate aref
Prastic~asin

coating

Hous~ng

Base

SECTION 25.13 Other Workhdding Devices

Sealed

697

Ogen Atmospheric pressure

O-ring seals in

25-22 Cutaway view uum chuck (Courtesy of i Tool Company, Inc.)

Push valve to release part

Adiust knurled nu2 to

vary holdlng power

with a polyethylene sheet. When the vacuum is turned on. the f i l m forms around the workpiece, covehug and sealing the holes not covered by the workptece and thus producing a s e a . The film covering the workp~ece 1s removed or the hrst cut removes the film covering the workplece. Vacuum chuck6 have the advantage that they can be used on both nonmetals and metals and can provtde an easily variable force. Magnetic, electrastatic,and vacuum chucks are used for some light milling and turning operations. As shown in Figure 25-23, T-slots are provided on milling machine tables so that workpieces can be clamped directly to the table, More often various workholding devices, called vcces or &&re$, are utilized. Smaller workpieces are usually held m a vlse mounted on the table. Fitures des~gned to spec~fically hold a part ~nthe correet location with respect to the tool are used for larger volumes. Fixtures reduce the time it takes to put the part in the machtne and assure repeatable location with respect to the cutting tools Fixtures provide clamping forces that counteract the cutting forces.

Workpiece to $6 rnllled

the work can be clamped dlrectly to the mllltng machines table (on the left), the FlGUfE 25-23 Wh~le workpiece on the fight is located in a flxturemounted in a clrcular attachment located on the table, so a circular slot can be mtlled in the workpiece

698

CHAPTER 25 Workholding Devices for Machine Tools

11 25.14 ECONOMIC JUSTIFICATION

OF JIGS AND

FIXTURES

As discussed previously, workholders are expensive: even when designed and const by using standard components. Obviously, their cost is a part of the total cost o f duction, and one must determine whether they can be justified economically by the ings in labor and machine cost and improvements use. Often it is only through the use of such devices met and sustained from part to part.To determine the economic cia1 tooling, the following factors must be considered:

1. The cost of the tooling 2. Interest or profit charges on the tooling cost 3. The savings resulting from the use of the tooling; can result Gom reduced cycle or improved quality or lower-cost labor 4. The savings in machine cost due to increased productivity 5. The number of units that will be produced using the tooling
The economic relationship between these factors can he expressed in the follo manner:
~avings per piece (exclusive of tooling costs)
2

additio
1 tooling

total cost per piece without tooting

A

I-{

total cost per piece using tooling (exclusive of tooling cost) labor cost per piece with tooling

costperpi

labor cost per piece without tooling

machine and cost + overhead per piece without tooling

machine and with tooling

C,

[(R)(t) + ( R , ) ( t ) l - [Rt)(ft) + (R,,,)(t,)l 2 where R

= labor

+ (C,/Z)(n)'(q

rate per hour, without tooling

R, = labor rate per hour, using tooling

t = hours per piece, without tooling
t, = hours per piece, using tooling

R,, =machine cost per hour, including all overhead

C, = cost of the special tooling

n
= number

of years tooling will be used

i = interest rate (or what invested capital is worth)

N = number of pieces that will be produced with the tooling
Equation 25-2 can be expressed in a simpler form:

(R

+ R,)t

(R!

+R

This equation assumes straight-line depreciation and computes interest on the age amount of capital invested throughout the life of the tooling.' When the tim
For the use of more sophisLiciried economic analysis,see C . S. Park, Conirmpory Engineering Econnmics,Znd cd., New York, Wiky. 1999...

which the tooling is to be used is less than one year, companies often do not include an interest cost. If this factor is neglected, the right-hand term of equatlon 2 5 3 reduces to C,/N The equations assume that thematerial cost will be the same regardless of whether or not special toolingis used.This is not always true.Although these equations are not completely accurate for all cases, they are satisfactory far determining toobng justiication i n most cases, because the life of tool~ng for machine tools seldom exceeds five years and often does nut exceed two yearsThe equation does not include ihe cost of poor quality.This can beincluded by estr&ting the decrease in the number of defective parts when the wu~kholder is med versus when it is not The following example,illustrates the use of equation 25-3 to determine tooling jusMication for a dedicated jig. In drilIing a series of holes on a radial drill, the use of a drill jig will reduce the time from 'iz hour per prece to 15 minutes per piece. If a jig is not used, a machinist,whose hourly rate is $18/hr,must be used. II the jig is used, the job ean be done by a machinist whose rate is $12/hr. The hourly rate for the radial drill is $3Z/hr. g jig would include $350 for design, $150 for material, and 50 The cast of m ~ n the hours of toolmaker's labor, which i s charged at the rate of $2% to include all machine and overhead costs inthe toolmakiig department. Investment capital is worth 16%to the company. It is estimated that the jig would last three years and that it would be used for the production of 300 parts over this period. Is the jig justified? The cost of the jig, C,, is estimated to be

Substituting the values givenin equation 25-3, we find:

So this jig is not justified based on cost savings

One could alxo ask how many parts would havc to he produced with the jig to break even (i.e., increased costs just equal savings). By omittlng the value 300 in the solution above and solving for N, it is found that a t leaat 1627 pieces would have to be produced annually with the jig for it to break even. This analysis assumes that the rime (of the people and rnach~nes) saved by the use of the spec~al tooling can be nsed far other product~ve work. If thisis not the case, the cost analysis should be altered t o take This important fact into account. Otherwise, the tooling ju~t~catianm3;y be substantially ineTror. The applicat'lon of group technology, NC machines, and lean manufacturing techniques may eliminate the need for designing and bulding a neui~ig or fixture every time a new part is designed. New measures of manufacturing productivity that include terms for quality and flexibility are being developed.

and methods on wst aecounhn&see C 3 Park;-Counmg the Cost6"Jafiuary 1987, MecPlontml Enginerring. p. 66

IF Key Words
iciple
JI&

electrostatic workholder future intermediate jig concept

~ig

locatlon magneuc chuck strap clamp

plate jlg ""g 1%

toggle clamp vacuum chuck workholder

'@

Leaf jig

700

CHAPTER 25

Workholding Devices for Machine Tools

IL: Review Questions

I . What are the two primary fu~lctions of a workholding device? 2. What distinguishes a jig from a fixture? 3. An early treatise delined a jig as"a device that holds the work and guides a tool." Why was this definition incorrect? 4. What modifications do you need to make to an ordinary vise so it can he a fixture? 5. What basic criteria should be considered in designingjigs and fixtures'? 6. In any part drawing, what arc! the critical surfaces of a part (i.e., what makes a part surface critical)? (This question requires an understanding of basic part drawings.) 7. What difficulties can result from not keeping clamping stresses low in designing jigs and fixtures? 8. Explain the 3-2-1 concept for workpiece location in a workholder on a machine tool. 9. Which of thc basic design principles relating to jigs and fixtures

15. What are rollover jigs, and what advantages do they offer 16. In the clampsshown in Figure 25-17, what is the purpose of spherical washer? 17. What are other common types of clamps'l 18. What is the purpose oI dimensioning the strap-clamp ass hly in Figure25-17 with letters? 19. Figure 25-8 showed the part sitting on locator buttons not have the part rest on the flat plate'? 20. In Figure 25-8, why aren't there three points put on sion on the part?

10. What are two reasons for not having drill bushings actually thischaplcr violate this rule? It is not uncommon to havc conflicts and trade-offs in fixture design situations. 11. Why does the use of down milling often make it easier to design a milling fixture than if up milling were used? 12. Namc another example of the intermcdiate workholder concept aside from thc video cnssetle. 13. A large assembly jig Cor an airplane-wing component gave difficulty when it rested on four-point support. The

23. For thepartshowninFigure25-8,why don't you drill theb first, then mill?Why mill at aU? 24. Notice that the holesin the part in Figure28-8 need to be tersunk after they are drilled. How must the jig be d a i put the countersinks on the mounting holes while the pa the jig, or would this operation be done afterward?

I I Problems
1. Using the following values, determine the number of pieces that would have to be made to justify the usc of a jig costing $3000. R = $5.75 R, = $4.50
t , = 1 '14 r = 2 'I,

1. How much time is saved with a swing clamp (in the loading and unloading) cycle? 2. How much does a swing clamp cost?

I = 10% R,,, = $4.50 N =3 2. Suppose in the sample problem at the end of thischapter that modular fixturing is used, which reduces the toolmaker's labor to 4 hours and the design cost t o $100 (4 hours at $25 per hour), and that the material cost (modular elements) for the suhplate structural elements, clamps, and so on was $600.What is the break-even quantity for a modular fixture? (Note:The modular fixture is uscd for the job, then disassembled and returned to the tool room.The parts arc reused in other workholders.) 3. Suppose the leaf jig in Figure 25-A could he improved by replacing the screw clamp with swing clamps See Figure 25-A. Many things are needed to be able to cost justify the improvement in the jig.This problem requires that the engineer cstimate or determine the following:

hole is 30 sec, the unloadlload time is 30sec, the operator i ting $12lhr, and the machine cost is $30/hr. 4. Examinc Figure 25-23. l b e part has a circular slot in the How else could you produce this slot?

Case study

701

fixture, 6 minutes to remove part from table and bolt up 2. String milling with three pieces 0.6 in. apart in a fixture. 3, index base milling (0.15 minute required to index the base) on a vertical single spindle mill where the operator is unloadingiloading the index table while the alternate piece is being machined (2 fixtures required). which arrangementshould be selectedbased on your operation time per piece and your estimated fixture cost per piece?

A 16-tooth cemented carbide face mill 150 mm (6 in.) in diameter has been selected for the job. The material is medium hard cast iron (220 to 260 Bhn). This surface is to be milled on a large number of identical pieces.The estimated time to unload and load a piece in a fixture is 0.30 minutes. Here are three possible choices of machine setup you could use to do this job.

NUMERICAL CONTROL (NC) AND THE A(4) LEVELOF AUTOMATION

26.1 I N T R O D U ~ Q N Brief History of Numerical Control and Flexible Manufacturing Systems (FMSs) Flexible Manufacturing Systems 26.2 BASIC PRINCIPLES OF . NUMERICAL CONTROL Part Programming A~TREND -S
~

.li.l

CI 26.1 INTRODUCTION

computer numerical control (CNC) machine tools with onboard microprocessors could be programmed directly

Flexible or programmable autovnatiun is typified by CNC machines or robots be taught or programmed externally by means of computers.The controlis in co software rather than mechanical bardware.

and the part with respect to the axis of the spindle of the cutting tool.Thrce position trol schemes are shown. Ln CNC tumingmachines, the feedback is on the tool tip with respect to thero part creating loo1 paths. Figure 26-3 shows how a part can be turned (machine a round bar in a CNC lathe. A program is written that directs the machine to the necessary roughing and finishing passes. number of rough passes = stock diameter - minimum diameter t finish depth of cut x 2

In this case, eight roughing passes and one finishing cut were specified.

Venlcal rnllltng rnach~ne

Venlcal boring rnachlne

Early NC rnach~ne

Early CNC rnachlne

CNC rnach~ne wlth Integrated NC rotary table, plvotable through 90"

Early CNC machining Center Early NC machine tools were controlled b y paper tape. Soon onboard computers were added, followed b y tool changers and pallet changers.

FIGURE 26-1

Machlnrng center w ~ t h tool changer

(20tools), chip conveyor, pallet changer,

and anti splash booth

Numerical Control (NC) and the A(4) Level of Automation Open loop

Machine
I

Part

control unlt (MCUI

Typical feedback: none

Drive motor

Closed loop-motor feedback

MCU compares

actual posltion to desired

position

a/Leq
Dr~ve motor
I / -

Typical feedback motor has turned 1.003 revolutions

Encoder

Closed loop-ball-screw feedback

Encoder measur positlon of table

a
Drive motor Command
V_

/ ' ~ @ p a f l

Feedback I I / Typ~cal feedback: ball screw hasturned 1003 revolut~ons

. . . .

Closed loop-worktable posltlon feedback

Feedback Pinion encoder

Typical feedback' table has moved 0.012 (0.3mm)

FIGURE 26-2 0 1

control schemes for NC and CNC :versus thre e posit~on

machine tools.

BRIEF HIST'ORY O F IN U M E R I CA L C O N T R O L A N D FLEXIBLE M A N U F A CTURII N G SYSTEMS (FMSs)

The advent and widc scale adoption of numerically (tape- and computer-) controlledmi chine tools has been the most significant development in machine tools in the paski years. These machines raised automation to a new level by providing positional feedbaq as well as prosrammable flexibility to machine tools. Numerical control of machine created entirely new concepts in manufacturing. Certain operations are now rout!! that previously were very difficult, if not impossible. to accomplish. However, NC impacts on machine tools were greater than expected. Machine t had to be redesigned to sustain more wear and tear on the drives, gears, and mote'$ and made better (more precise) because the operator n o longer controlled the positi of thc work in respect to the cutting tool.The machines cost more (for controls andpl cision), with automatic tool changers and pallet changers quickly being added to ma the NC machine a "machining center."

4
I

r\k

-----------Return 1 , - - - - - - - - - - - - Return 2~

- -- --

15 depth of ger pass

0.3 R

Pass 2 ~eturn3t----; Pass ? - - - - - .Return 4 . - -Pass 4 -- - --~etur ,n 5 - - - <-G71 Pass 5 - Return 6 - - +

i1

cycle

rough

G70 f ~ n ~ s h
cycle

*z-P-5-i

FIGURE 26-3 The tool paths necessary to rough and fin~sh turn a part in a CNC lathe are computer generated usinq G codes.

In earlier years, highly trained NC programmers were required.The development of low-cost, solid-state microprocessing chips resulted in machines that can be quickly programmed by skilled machinists after only a few hours of training, using only simple machine shop language. As a consequence, today almost all manufacturing facilities, from the largest to the smallest job shops, have one or more numerically controlled machine tools in routine use. NC came into being to fill a need.The U.S. Air Force (USAF) and the airframe industry were seeking a means to manufacture complex contoured aircraft components to close tolerances on a highly repeatable basis. John Parsons of the Parsons Corporation of Traverse, Michigan, had been working on a project to solve this problem; in 1947 he and his engineers developed a machine that would machine templates to be used for inspecting helicopter blades. He conceived of a machine (a jig borer) that was controlled by numerical data to make these templates and took his proposal to the USAE. Parsons convinced the USAF to fund the development of a machine. Subsequently, MIT was subcontracted to build the first NC machine in 1949.The prototype was a conventional two-axis tracer mill retrofitted with servomechanisms. As luck would have it, the servomechanism lab was located next to a lab where one of the very first digital computers (Whirlwind) was being developed. This computer generated the digital numerical data for the servomechanisms, and in 1952 a modified three-axis Cincinnati Hydrotel milling machine was demonstrated. In 1962,35 billion was spent on machine tools and NC machines accounted for about 10% of total. Early on, NC machines were continuous-path or contouring machines where the entire path of the tool was controlled with close accuracy in regard to position and velocity, and the large aerospace companies had many home-made NC machines. Today, milling machines, machining centers, laser beam and water-jet cutting machines, and lathes are popular applications of continuous-path control requiring feedback control. Next, point-to-point machines were produced in which the path taken between operations was relatively unimportant and therefore not monitored continuously. Point-to-point machines are used primarily for drilling, milling straight cuts, cutoff, and punching. Automatic tool changers, which require that the tools be precisely set to a given length prior to installation in the machines,permitted the merging of many into

CHAmrR 26

Numerical Control (NC) and the A(4) Level of Automation

Control panel and compute1

Tombstone or post-type modular

FIGURE 26-4 Hor~zontal mach~nlng center wlth four-axls control (X, 2, R table) recelves lnputs to the control panel from many sources

Rotary index table on

center. During the early days of NC, the machine tools had to be program machine control language, a difficult, time-consuming task prone to error. panies developed computer languages to aid the NC programming task, each ing a different language to describe tool geometry, tool movements, and m

SECTION 26.1 Introduction

707

(automatically programmed tools), was introduced by MIT in 1959, when computer technology was in its adolescence. Hundreds of thousands of parts have been programmed using APT,running initially on large mainframe computers. The output from the APT program had to be converted to the language of a particular was called postprocessing. Traditional postprocessing yields NC workpiece programs that are not exchangeable. To machine an identical workpiece on another machine, the program must be postprocessed again unless the machine and control are exactly the same. With the arrival of high-resolution computer-aided design (CALl) graphics, many people thought that APT would be phased out. But when complex tool control is required for complex parts, APT or one of its many offspring is still used. The chief problems in NC programming were tool radius compensation and tool path interpolation (discussed later). Computer software with the capability to perform linear, circular, parabolic, and other kinds of interpolations were developed. The capabilities were included in APT, and such software programs are now routinely available on CNC machines. In the 1960s,it was envisioned that a large computer could be used directly to control, in real time, a number of machines. A limited number of direct numerical control (DNC) systems were developed, with the idea that the programs were to be sent directly to the machines (eliminating paper tape handling).The mainframe computer would be shared on a real-time basis by many machine tools. The machine operator would have access to the main computer through a remote terminal at the machine, while management would have up-to-the-minute data on production status and machine utilization.This version of DNC had very few takers. Instead, NC machines became computer numerical control (CNC) machines through the development of small,inexpensive computer microprocessors with large memories. These onboard computers have every sort of input imaginable coupled with functions such as programstorage, tool offset and tool compensation,program-editing capability,various degrees of computation, and the ability to send and receive data from a variety of sources, including remote locations. The computer can store multiple-part programs, recalling them as needed for different parts. As the software and controllers developed, it was immediately found that the machine tool operator could readily learn how to program these machines (manually) for many component parts, often eliminating the need for a part programmer. In recent years the DNC concept has been revived with the small but powerful computers on the machine tools being networked to a larger computer to provide enhanced computer memory and computational capacity.Therefore, DNC now means distributed numerical control, with the distribution of NC programs by a central computer to individual CNC units. And so emerged a special type of manufacturing system called the flexible manufacturing system, or FMS. Historically speaking, the first examples of FMS systems appeared in the late 1960s,but few companies adopted them because of their high initial cost.

FLEXIBLE MANUFACTURING SYSTEMS The development of FMSs began in England and the United States simultaneously in the 1960s. By combining the repeatability and productivity of the transfer Line with the programmable flexibility of the NC machine, a variety of parts could be produced on the same set of machines. In the United States the first systems were called variable mission or flexible manufacturing systems. In the late 1960sSundstrand installed a system for machining aircraft speed drive housings that was used for over 30 years. Overall, however, very few of these systems were sold until the late 1970s and early 1980s,when a worldwide FMS movement began. But even today international trade in FSM is not significant, and there are fewer than 2000 systems in the world (less than 0.1 % of the machine tool population).There is also some evidence that the market for these large, expensive systems became saturated around the mid-1980s. Essentially, the FMS permits (schedules) the products to take random paths through the machines.This system is fundamentally an automated, conveyorized, computerized job shop so the system is complex to schedule. Because the machining time for different parts varies greatly, the FMS is difficult to link to an integrated system and often remains an island of very expensive automation.

708

CHAPTER 26

Numerical Control (NC) and the A(4) Level of Automation

- Remote wire-guided AGV (2 reqd.1

Measuringlinspection station

Two Powerhawk HTC300 Horizontal Tool Changers with 60-tool storage magazines

Pallet acquisition system

Application: An aircratt parts manufacturer needed pans transfer mobiiity, inlout parts queue, cutting tool library, and quality control management for production of high-technology pans. Wire-guided vehicles offer interdepartrnent transfer capability as well as in-cell transport. The qc center manages the machining accuracy for continuous flow of acceptable parts. Parts are scheduled in batch andlor random, controiled by a management computer. The machines ere equipped with telemetry probes, adaptive control, bulk tool storage, and complete tool management.

FIGURE 26-5 An example o f a sophisticated FMS developed for machining aircraft parts. A wire-guided cart called an A@ (automated guided vehicle) is used to transport pallets from the unloadiload station to the machines.

A b o u t 60 t o 70% o f FMS i m p l e m e n t a t i o n s a r e for c o m p o n e n t s c o n s X i n g of r o t a t i o n a l (prismatic) p a r t s such as crankcases a n d transmission housings. Figure M S w i t h t w o m a c h i n e t o o l s serviced by a p a l l e t system a n d a n automat depicts a n F g u i d e d v e h i c l e (AGV). T h e balance of FMS installations a r e for r o t a t i o n a l parts o

typical. A n n u a l production volumes for the systems are usually in the range of 3000 to 10, parts, t h e n u m b e r of d i f f e r e n t k i n d s of p a r t s r a n g i n g from 2 t o 20, w i t h 8 being typi T h e l o t sizes a r e t y p i c a l l y 20 to 100 parts. a n d t h e t y p i c a l p a r t has a m a c h i n i n g time

S~cno~26.1 Introduction . . . ... . , . #,&x&fi?anufacturing

709

. ... . . . ...
.,

. .

Systems

Pallet changers Mult~plernach~ile tmls WCor 6NC Automated m a t u u l handling system (to deitver parts to mach~nes) Computer control f4rsystem: DNC Mulnple pa~ts.Med~urn-sized l ~ t(20&10,0W) s wlth fam~hes of parts Random sequencing of parts to machines (aptional) Automaric tool c h a n ~ n g Inprbcesr mspectian Parts washlug (optronal) Autqrnuted storage dml ralircval (optional, to deliver patis ta system)

about 30 minutes witb arange of 6 to 90 minutes per part. Each part typically needs two or three chuckings or locating positions and 30 or 40 mach~ning operations. Early on, NC machining centers were used, but in recent years, CNC machlne tools have been favored, leadinn to a cofls~derable number of systems being operated under direct numerical control. The machining centers always have tool changers.To overcome the limitation of asingle spindle, some systems are beingbuilt withhead changers.These are sometimes refe-rred to as modular machining centers. Common features of FMSs (see Table 26-11 are pallet chnngnrs, underfloor conveyor for the collection of chips (not shown), and a conveyor system that delivers - system . parts to the machine. This is also an expensive pan of the system, as the convsyor systems are e~ther powered rollers, mechanical pallet transfer conveyors,or AGVs operating on underground towlines or buried guidance cables The carts are more flexible than the conveyors. The AGVs also serve to connect the islands of automation, operating between FMSs and replacing human guided veh~cles (forklift trucks). Pallets are a srgnificant cost item for the FMS because the part must be accurately located on the palIet and the pallet accurately locatedin the machine. Since many pallets are required for each different component, a lot of pallets are needed and they rypically represent anywherefxom 15 to 20% of the total system cost FMSs cast about $1million per machine tool. Thus the seven-machine FMS costs $6 million fox hardware and software, with the transporter costlng over $3 millian
Computer Contml in FM5. The CNC machines receive programs as needed from a host minicomputer, which acts as a supervisory computer for the system,TracKiog the sfatus of any particular machin~ 1n the system. In recent years, iil-process inspection, detectlon, and automatic tool posit~on correction for tool wear and breakage have been added because a very common problem w ~ t h these systems is the monitoring of the tool condition and performance. Most installations also mcur problems in the performance e may take just as long and reliability bf the software and the control systems. ~ g c a u r it to debug software as it does to debug hardware, delays of two to six months in startup are ~ .not ~ uncommon ~~~~. Hu\\oi.r, dpc'r,ltor>drc r!.pi;all! n-xdcd to load u(~rkpiecc\. ur~load finished part,. ch;lnd', u u r n tot~ls, and perform ctluipmr,nt n~.iintcn;~n~c ilr~d repairs. ( Y C imd I)S(: functions are often incorporated into a single FMS.The system can usually monitor piece-part counts, tool chadges, and machine utiluation, with the computer also providing supervisory control of the production. The workp~ecesare launched randomly into the system, which identifies each part In the famlly and routes it to the proper 111achi11es 'Illc s>stcrnsqcncr;llly d~spliry rcJuicJ m;tnul,~cturinc IcaJ ri~nr,ld\v in-l~rocc% in\cntorv, and high mcichinc tool utiliz;ttion,uith rcduc,>rlindtrecr arid direct lahor.'lhc materials-handling system mu$t be able to rouTe any part to any m a c h i i in any order and pmvide each machine with a small queue of "banked parts" waiting to be processed so as to maximize machine utilizauon. Convenient access for loading and unloading parts, compatlb~lity with the control system, and accessibility to the machme tools are other necessary d w ~ g n features for the material handling system
~~ ~ ~

a1 Control (NC) and the A(4) Level of Automat~on

The computer control system for an FMS typ~callyhas three levels. The mast control monitors the entire system for tool fa~lures or machine breakdowns, schedul the work, and routes the parts to the appropriate machine.The DNC computer dlstribut programs lo the CNC inach~nes and supervises their operations, selectmg the requlre programs and transmitt~ng them at the appropriate time It also keeps track of thec to the master computer plelion of the cutting programs and sends this rnformat~on bottom level of computer control is at the machmes themselves. It is difficult to design an FMS because it is, in fact, a very complex assembly ofe ements that must work together. Designing the FMS to be flcxtble 1s also difficult,Man companies have found that between the time they ordered their system and the time they had it ~nstalled and operational, design changes had eliminated a number of partsfram the FMS That IS, the system was not as flexible as they thought. Flgure 26-6 showssome typical FMS de~igns.Today a popular system has only one (or two) mach~ning centen, with an automatic storage and retr~eval system (ASRS) to provide the machine w~tha contlnuaus supply of parts dur~ng all three shifts.This deslgn is called a flexiblemanu. facturing cell (F'MC). FMSs are, in fact, classic examples of supermachines Suchlarge, expensive systems must be exam~ned with careful and complete planning. It i s tant to remember that even though thcy are often marketed and sold as a tumke stallation (the buyer pays a lump sum and receives a system that can be turned o run), this is only rarely possible with asystem that has so many elements thatmus together reliabJy.Taken in the context of integrated manufacturing systems,large may be difficult to synchronize to the reat of the system. The flexib~lity of the requires variable speeds and cycles, numerical control, and a supemsory comput coordinate cell operation. In the long run, smaller manned or unmanned cells ma be the better solution, in terms of system flexlb~lity Perhaps a better name for systems would be ~arzable mzsAron or random-path manuficturcng systems Many companies elect to ldentify the fam~lies of parts around which the FM s~gned, wh~ch greatly mproves the ubli~ation of the FMS. One might say tha were developed before their time, snlce they are bemg more readily accepted group technology has been used (at least conccptually) to ~ d e n t ~families fy of p the system to produce. As an FMS generally needs about three or four workers per shift to load an load parts, change tools, and perform general mamtenance, it cannot really be sat self-operating. FMS systems are rarely left untended, as in th~rd-shift operations than the personnel doing the loadlng and unloading, the workers m the FMS are us h~ghly sk~lled and trained in NC and CNC. Most ~nstallations run fairly reliably ( they are debugged) aver three shifts, w ~ t h uptlme ranglng from 70 to 80%, and man able to run on one sh~ft untended.

LES OF NUMERICAL CONTROL

NC uses a processing language to control the movement of the cutting tool or plece or both.The programs contain informalion about the machme tool and cuttin geometry, the part dimensions (from rough to finish slze), and the machining par ters (speeds and feeds and depth of cut).Thus NC machlnes can duplicate conse parts, and a part made at a later date will be the same as one made today. Repeat and quality are unproved over conventional bob shop) machines. Workholding de can be made more universal, and setup tlme can be reduced, along with tool-cha time, thus making programmable machlnes economical for produc~ng small lots or a slngle plece When combrned with the managenal and organ~zat~onal strateg group technology (GT) and cellular manufacturing, programmable mach~nes le tremendous improvements in quality and productiv~ty. GT bas~cally leads to the atlon of fam~lies of parts made in machining cells containing flexible, programm machines The compat~bility of the components (similarly in process and sequen processes) greatly enhances the productivity (utility) of the programmable equi A side result has been the decrease m the non-ch~p-producing time of m tools. The operator was relieved of the jobs of ehangmg speeds and feeds and 1

Secno~ 26.2
@changeable head-type matching center

Basic Pnnc~ples of Numerical Control

711

@ Matching center with

pallet stored in ASRS

@~ ombinat~o of n multiple @ ~ombinatlon of multlple machin~ng centers

with pallet pool

machln~ng centers and manuallvoDerated conveyor

I
machining centers and automatic conveyor

Loadsiunloads ASRS

d3
of multlple ~ndependent CNC machines . DNC

Q
Conveyor

@ Comblnatlon of multiple @ S~multaneouscontrol

mach~nlng centers and AGV with robot carver

@Transfer line comprised of mult~ple mach~nrng centers

comprised of machining centers and NC special purpose machines NC measuring st.

with special-purpose machines and generalpurpose machines

Used as needed Sequence controller, including

Operator General-purpose machine tool machine

Computer

IXI electrical conrrol box

Computer control of FMS Pallet changer storage and retrieval system AGV-automated guided vehicle NC special-purpose machine

FIGURE 266

Examples of machining centers-FMC

and

FMS designs.

the tool relatlve to the work Even simple forms of NC and digital readout equipment have provided both greater productivity and increased accuracy. Most early NC machine tools were developed for specla1 types of work where accuracies of as much as to provide accuracies of at 0.00005 in were required, and many NC machines are bu~lt least 0.0001 in., regardless of whether or not it 1s needed.This forced many machine tool builders to redesign their machines and improve t h e ~quality r because the operator was not available to compensate for the machine (positiomng)error. Whde most NC machne tools today will provide greater accuracy than 1s required for most jobs, the trend is toward greater accuracy and precis~on (i.e., better quality) but at no Increase in cost.

: a 1 Control (NC) and the A(4) Level of Automation

Therefore,NC machines will continue to be the very backbone of the machine tool ness. Hopefully, all bmlders will one day arrive at a common language, so that if one program one machine, one can program them all. As the name implies, numertcal control is a method of controlling them of mach~ne components by means of numbers or coded instructions. Assume that 1-in. holes in the part shown in Figure 26-7 are to be drllled and bored on a ver sprndle machine, The centers of these hules must be located relatlve to each other wlth respect to the left-hand edge (X direction) and the bottom edge (Y dlrectio the workpiece.The depth of the hole wlll be controlled by the Z- (or W - )axls Fo part, this is the zero reference point for the part. The holes will be produced by center drilling, hole drilhng, boring, reaming counterboring (five tool changes). If this were done conventionally or m manned three or four different machines might be required. On the NC machinrng center,lt requires changing the tool automatically The movements of the table are contro coordinate systems for each direction The machrne tool has a zero point.The ac positioning of the cutting tool wlth respect to the work is established by these points.The center of the table or a point along the edge of the traverse range 1s co used. The workpiece is positioned on the table with respect to this iero poin table. For our example, the lower left-hand comer of the part is placed on the table in the X+ direction,4 in. ln the Y + direction, and 2 in in the Z+ direction with res to the machine zero point A software program 1s wntten that instructs the table to move with respect to axis of the spindle to bring the holes to the correct locat~on for machmmg.The mac shown in Figure 26-7 is called a five-ax~s machine because it has five movements [sh by the dark arrows) under numerical control No fixture is shown in this example, but would typically be used to obtain quick and repeatablelocahon of the part on the

HOW CNC MACHINES WORK NC and CNC machines can be subdivided Into two types, shown in Figure 26-8. Inp to-point machines, the tool path 1 s not controlled but tools can be moved in str lines or parallel traverses at des~red table feed rates, but only one iws drrve is oper at a trme. Contourrng permits two or three axes to be controlled simultaneo permrtting two- or three-drmensianal geometries to be generated Another ter contouring is contrnuous path. Most mill~ngand turning machining centers contouring capability and are closed 1oop.The point-to-point machines can he o loop rather than closed loop. The X-axis ot the three-ax~svertical spindle CNC mach~netool show Figure 26-7 wtll be used to explain how the closed-loop posiaonal control works. Controlling a machine tool using var~aMc input, such as a punched tape or ast program, 1s known as numerical control and is defined by the Electronic Indus Association (EM)as "a system in which actions are controlled by direct insertion numerical data at some point (the measured data is call the parts program).The sys must automat~cally Interpret at least some portlon of the data." Traditionally,NCmachine tool has a machcne control unrt (MCU).The MCUis ther divided into two elements. the data-processing unlt (DPU) and the control-1 unit (CLU). The DPU processes the coded data that are read from thc tape or other input med~um that it glves to the CPU, specifically the position of each axis,i rection of motion feed, and its auxiliary-function control signals The CLU operate dr~ve mechanisms of the machine. The CNC control system,shown schemat~cally in F~gure 26-9, uses a resolver coder to prov~de axis-positron feedback to the MCU A closed-loop control transducer os sensing device to detect machine table position [and velocit touring) and transmit that lnformatlon back to the MCU to compare the current with the desired state If they are different, the control unit produces a s~gnal drive motors to move the table, reduc~ng the error signal and ultimately movi table tothe desired position at the desired veloctty.At thrs polnt, the command co reaches zero, meanrng that the wlrect number of pulses has been sent to move the to the desired position. In a closed-loop system, a comparator 1s used to corn

SECTION 26.2

Basic Principles of Numerical Control

713

j1
I

1 . 0 0 0 5 Bore 1.252 C' bore

k . 7 1 1 8 /
I

1. O O O O

0.2180 1.250 deep (3 holes)

3.000 diam. slot 0.3000 deep. , "0.3000 wide

.7118

1.9445

Zero reference

a,

FIGURE 26-7 The part (above) to be machined on the NC machine (below) has a zero reference point. The machine also has a zero reference point.

FIGURE 26-8 NC and CNC systems are subdivided into two basic categories: point-to-point controls or contouring controls.

Point-to-point control (milling machines, welding)

Contouring control (machining centers)

Measuring scale

Inputs to MCU

Axis drive system ---------------

FIGURE 26-9 The table of the CNC machine (above) is translated with a ball screw mechanism, and its location is detected with a resolver. The schematic below shows how the table is located with respect to the spindle axis of the machine tool.

(actual value)

Resolver or

(drilling a hole) takes place. Closed-loop systems tend t o have greater accu

Srcnom 26.2 Basic Principles of Numerical Control

715

For most NC controls, the feedback signals are supplred by transducers actuated either by the feed screw or by the actual movement of the component. The transducers may provide e~ther digital or analog information (signals).The resolver in Figure 26-9 measures (indirectly) the movement of the table by the rotation of the ball screw. A pulse disk on the end of the ball screw converts analog movement back Into drgttal pulses, which are used to calculate table movement. The tachometer on the dnve motor measures the table velocity Tho basic types of dtgttal transducers are used. One supplies rncremeutul mformation and tells how much motion of the input shaft or table has occurred smce the last tlme The mformatlon supplied is simtlar to telltng newspaper carriers that papers are to be delivered to the first, fourth, and erghth houses from a glven corner on one slde of the block. To follow the instructions, the carriers would need a means of counting the houses (pulses) as they passed them.They would deliver papers as they counted 1,4,and 8.The second type of drgrtal information is absolute in character, with each pulse corresponding to a specd~e location of the machine components.To continue the carrier analogy, this would correspond to telling the carriers to deliver papers to the houses having house numbers 2400,2406, and 2414 In this case it would only be necessary for the caniers (machine component) to be able to read the house numbers (addresses) and stop to deliver a paper when arriving at a proper nddress.This address system is a common one in numerical control systems, because it provides absolute location informatlon relative to a machine zero point. When analog information is used, the signal 1s usually in the form of an electric voltage that varies as the tnput shaft is rotated or the machlne component is moved, the variable output being a function of movement.The movement is evaluated by measuring, or matching, the voltage, or by measuring the ratlos between the applied and feedback voltages, thts eliminates the effect of supply-voltage variations. The Input informatton (r.e., the location of the holes) is given in bmary form to the machine control unit in the form of a punched lape-a magnetlc tape in a cassette on old machmes or disks, as shown, on newer machines. The data are mput drrectly via the control panel or a computer program.Thls command signal is converted into pnlses by the machine control unit, which in turn drives the servomotor or stepper motor. Alternatmg-current servomotors are rapidly replanng direct-current motors on new CNC machine tools.The reasons for change are better reliability, better performance-toweight ratlo, and lower power cvnsumption. If the system is point-to-polnt (or positioning), the control system disregards the paths between points. Some positioning systems provlde for control of straight cuts along the machine axes and produce duganal paths at 45" to the axes by maintaming one-to-one relationships between the n~otions of perpendicular axes. Contouring systems require directional changes at controlled velocity.Any lost motion m the system can &stort the part. In contouring, it is usually necessary to control multiple paths between pouts by lnterpolattng lntermedtate coordinate pos~tions. As many of these systems as destred can be combined to provide control in several axes-two- and three-axis contlols are most common, but some machtnes have as many as seven. In many, conversion to either English or metric measurement is available merely by throwing a switch. The components requ~red for such a numertcal control system are now standardized Items of hardware. In most cases the drwe motor is electric, but hydraulic systems are also used.They are usually capable of moving the machme elements, such as tables, at high rates of speed, up to 200 in.lmin being common. Thus exact positiontug can be achieved more raprdly than by manual means.The transducer can be placed on the drive motor or connected directly to the leadscrew, with specla1precautions being taken, such as the use of extra-large screws and ball nuts, to avoid backlash and to assure accuracy In other systems, the encoder is attached to the machine table, providmg direct measurement of the table positton. Various degrees of accuracy are obtainable. Guaranteed positioning aecnrawes of 0.001 or 0.0001 in. are common, but greater accuracies can be ohtamed at higher cost. Most NC systems are built into the machines, but they can he retrofilted to some machine tools

26 Numerical Control (NC) and the A(4) Level of Automailon

Ball nut attached

A
M

P1
FIGURE 26-10 The ball lead screw shown in &tall provides great accuracy 4nd posltlon to NC and CNC machine tools.

Clt

Eur

'IU

table mavements than is pomible for cooventional machines.

sequence of operations; selects thecutting tools and workholding devices; and speeds, feeds, and depths of cat.Common machining routines such as pocket peck drilling hsve bet% programmed into many CWC machines.These a* call cycles.& shownin Figure26-11, the operator merely supplies theinformation re by the eontrol menu,and the machme fills in the necessary data for the previously program to perfarm the desired maehining routines

Rectangular pocket mill~ng Control menu asks for. +Setupclearance M~ll~n depth g Rough~ng depth -Feed rate for rough~ng F~rst slde length * Second s~de length Feed rate D~rectlon or rotatton

Peck dr~ll~ng Control menu asks for Setup clearance Pecking depth -----.------..---.Total hole depth Pecking depth -------.-----..--Peck~ng depth Dwell time (seconds) Total hole depth -------.-----...Feed rate

Setup clearance

URE 26-11 Canned or programmed machining tines greatly simplify gramming CNC machines. urtesy of Heidenkoin ooration, Elk Grove Village,
)

Total hole depth

Tapping Control menu asks for. Setupclearance .Total hole depth Dwell time isecondsl Feed rate

Source: He~denha~n Corp (Elk Grove Village, IL.)

To ensure accurate machining o f a workpiece o n a CNC machine, the control syst e m has to know certain dimensions of the tools. These tool dimenxiurn are referenced to a fixed settingpoint o n the tool holder. For the milling cutter, the dimensions are length L and cutter radius. F o r the turning tool, the dimensions are length L and transverse overhang.These dimensions are part of the information given to the operator o n the setting sheet (see Figure 26-12). The program assumes that the tools w i l l have the specified dimensions.

Tool setting point Tool setting point

Cutter radius

Traverse overhang

/ S~ngle pomt tool

Length L

FIGURE 26-12 The locatlon of the corner of the end m ~ l(left) l or the t ~ of p a s~ngle-point tool (r~ght)must be known with respect to the tool settlng polnts so that tool d~mensions are accurately set

TER 26

Numer~calControl (NC) and the A(4) Level of Autornat~on

(a1Workpiece drawing IXY-Plane1

j *

(cl Programmtng the outer contour

Ib) Workpiace drawing

(XZ-Plane)

(dl Programming the slot

Example of programming a patt in a vertical-spindle WCmachine.

PART PROGRAMMING
Obviously, the preparation of thz~ontrol program for use in NC machines is a cri step. Many standard languages and programs have been developed by the machine builders, and only the very basicaspect can be presented here.

sioned with respect to its zero reference point. This may r e w r e redrawing and

The workpiece drawing, two views, is redone into a wordinate system. In FIgure 2 workpiece zero point is located at the bottom left-hand corner of the workpiece d

The three axes of the machine coordinate sysfem is established in relation to the workpiece as follows: *Then- and y-axes are used to show the workpiece geometry in the XY-plane. The z-axis shows the down feed or depth.

All the coordinates of the most important pohtv on the part (mthis case the points PI-P11) have been collated in the form of a table:
P1
X

P2
190

P3
lOa
130 20

P4
lOrr

P5
70
105 20

P6
70

P7
10

P8
10

P9
20 10
220

PI0
160 100 20

PI1
130 45 2 0 0

190
10

130
2U

105

130
20

130

20
20

20

2Q

20

Preparation of the program includes the selechon of the cutting tooIs (a 10-mmdiameter end mill) and theselection of desired cutting speeds and feed per tooth.As discussed in Chapter 25, these parameters, along with the depth of cut and depth of immersian, determine the spindle rpm and table feed rates. For determining the outer contour, the tool radius must be considered because the path of the milling cutting along the outer contour is input Into thecontrol program, notthe actual pertmetw of theparts.To ensure that the control system can in fact control the milling cutter center axis correctly,the size of the milling cutter (radius or diameter) has to be k n o w beforehand, as does the side of the finished contour (relative to the machining direction) on which the milling cutter is located. How these two items of information are input is not standardized for all machine tools, and it is therefore assumed here that the control system has already received this information. Machiningthevuter contourbegins with the milling cutter moving at rapid traverse rate-that is, funcr~on G O O Xl90 Y-5 to the starting point adjacent to the workpieceO O 220 after switching on the spinwhereupon down feed takes place with function G dle rotation. Machimng of The outer contour is then accomplished with the following G functions The supplementary Functton F produces the necessary feed rate in mdmin. The down feed depth is taken up by the mtlling cutter at the startmg point.

GO1 G O T GO1 QOZ GO1 GO1 GO1 GD3 GO1

Y130 F200 XlOO XlO5 F150 X70 Y105 R15 Y130 F200 XI0 Y20 X20 Yl0 RIO XI90 F200

S t r a ~ hline t trom starting p m t to F2 Straight line from P2 to P3 Straight line from P3 to P4 Radial arc, clockwise, with 13 radius Straight line from P5 to P6 Straight line from P6 to P7 Straight line from P7 to P8 Fl50 Radial arc, counterclockwise with 10 radius Straight line from P9 to P1

Word

Use

D~mens~orurlcoordmalen data linear and hngular mution commands far the a m athe machine Feedfirnulow szls feed rate Tar thre operation Syeedfunctcowsets cuttxng speed for t h ~ ~ o p e r a t ~ o n Tadlfwctmn. teLlsfhe maahme thelocatwn O f the toolin the tool holder or Lo01 turret M~scellaneousfunctron'turns coalant on or offopeds spmdle,reverses spmdle, tool ohange,etc. End ofblock, indicates to the MCU that a Eull block of informatlirn has been Lrsnsm1tted and the blo& can be executed

Numerical Control (NC) and theA(4) Level of Automation For milling the slot, the milling cutter is first retracted by GO1 235 from the workp to point P1 shown in Figure 26-13d. Because the width of the slot is equal to the cutter diameter, the path of ter is easy to program.The tool radius compensation feature is switches off (by the following traverse instructions refer to the cutter center line axis. Milling of the slot is accomplished by the following instructions: G O O GO1 GO1 GO1 G O O XI60 220 XI30
235

XlOO F150

Y45
F200 Y30O

Z10

X300

Rapid traverse to point PI0 Down feed at point PI0 Straight line from P10 to PI1 Retraction from workpiece Rapid traverse away from workpiece.

After the program has been written, it is verified, which means the steps the part are graphically simulated.The verification step can use the comput which simulates the part being made by tracing all the toolwork paths as they occur on the machine tool. Sometimes a sample part is machined in plastic or mach wax for checking the part specifications from the drawing against the real part. The program can be directly entered into the control panel of a CNC mac Usually, longer programs are entered by tape or disk and short programs are ent manually.The inputs may be in binary code (in older machines) or in arabicn verbal commands that the machine understands. Historically, punched tape was used to control the movements.The ide the old player-piano roll, which was a form of tape control, and punched ca used for many years for controllimg complicated weaving and business machinesTh control of machine tools was an extension of an existing basic concept in which ho resenting information that has been punched into the tape, are read by sensing and used to actuate the devices that control various electrical or mechanical mech Over the years, four basic types of tape format have been used for NC input to municate dimensional and nondimensional information:fixed-sequential address format, tab-sequential format, and word-address format. Most ne systems use the word-address format, which allows the words to be presented order and is the most flexible. Today, CNC machines permit the user t o program the interfac machine tool and the control, greatly reducing the number of components and interconnections. Current CNCs have extensive sel performance-monitoring systems. The greatest advances in CNC techn are in part programming, where easy-to-use,menu-driven software makes progr almost as simple as setting up the machine manually. With older NC mach' operator could override the program when necessary but could not repro machine unless a new program was written. The CNC machine has the cap reading a program into its computer memory, and the program can be modif machine like any other computer program. On CNCmachines, the machine tool operator may perform a steps right at the console of the machine, programming the processing directly into the computer memory. The program can be saved by hav print out a copy of the program, which can be used later for reorders of the s Features such as program edit, canned routines, program storage, diagnostics, surface speed, and tape punch are common on today's CNC machines. As more and more design work is done on the computer (CAD) using dat and software that are compatible to the machime tools, there will be less depende tape for program storage and more utilization of floppy disks, hard disks, and 0th ical computer storage means. For example, a machining cell composed of NCm tools designed for a family of 10 component parts is able to make the 10 differe without needing retooling or refixturing, but it will still have 10 different progr these parts for each machine. If the programs are stored in the computer, they readily accessed, but if they are stored on disk, delays will occur in dumping the ent programs into or out of the control computer.

S E ~ I263 ~ N Mach~ning Center Features and Trends

721

FlGUKE 26-14 Two classic problems in NC programrnlng are the determlnatlon o f cutter offset and

interpolation o f cutter parts.

Cutfer Offset and Int~rpolutIon in Numerical Conhof. Although the majority of NG and tape-controlled machine tools do not provide for machiningcontoured surfaces most CNC and DNC machines do. The requi~ed curves and contours are generated approximately by a series o f very short straight lines or segments of some type of regular curves, such as hyperbolas.This is caIled inte~polalion.The program fed to the machrne IS arranged to approxunate the required curve withm the desired accuracy.Figure 26-18 illustrates how a desired straight-line or curved surface can be approximated by means of short segmenrs. Interpolation refers to the fact that curved surfaces as generated by machine tools must be approximated by a series of very s h ~ r t straight-line , movements in the nay,and z directions. The length of the segment must he varied in accordance wlth the deviatron permitted. Most NC machine tools with contouring capability wtll produce a surface that is within 0.001 in. of the one desued, and many wfil p~ovide considerably better performance. Most contouring machines have elther two- or tbreeaxis capability, but five-=is capabilliy as bhown in Figures 26-7 and 26-15 arereadtly available. In milling machines, the centerline of the cutter is offset from the desired surface b y the radius of the cutter.The path that the cutter needs to take to generate the deriired g e o m e q is not sidlply the per'meter or profie of the part. Thus cutter offset program must be included in thesoftware. Obviously, contour machining (with cutter offset and interpolatron) requxer that complex information be entered ~ n t the o computer because the numberaf straight-lineor curvedsegments may be quite large. Manual programming of the tape ean be quite laborious. Computer programming can translate srmple commands into the complex information required by the machine.
AGHXNING CENTER FEATURES AND TRENDS Computer numerical conu-01is used on a wide variety of machine tools.These rarlge from single-spindle drilling machines, which often have only two-axis control and can be
obtainedfar about $10,000, to machimng centers,such as shown in F w r e 26-15.The machining center can do dnillrng, boring, milling, tapping, and so on, with four- or five-axis control It can automaticalIy select and change 40 to 180preset tools.The table can move lefthight and inhut, and the spindle can move upidown and &out, with pasitionkg accuracy in the raoge: U.00312 in. m.th repeatability to 2 . m in. over 40 in. of travel.

FIGURE 26-15 Modern rnachlning centers will typ~cally have horizontal sp~ndlez w~th rpms up to IS,OOO,dual pallet! 40 to 100 tools and cuttfng-tool rnagazlnes hold~ng

Beyond accuracy and repeatability and the number of controlled axes, CNC mach~nes also categorued by spindle speed, horsepower, and the size of the workpiece.The racy and repeatability (prec~s~on) result from a comb~natlon of factors includin resolution of the control instrumentation and the accuracy of thc hardware.Theco resolut~on is the minimum length distinguishable by the control unit. It is called the length unlt (BLU) and is mainly a factor of the axis transducer and the quality o leadscrew or linear translator.There are many sources of error in a machme, hclu wear m the m a b e sliding elements,machine tool assembly errors, spindle m o u t (w and leddscrew backlash. Tool deflection due to the cutting forces produces d~mensional error and ch marks Thermal error,caused by the thermal expansion of machine elements,is n form and is normally the greatest source of mach~ne error. Methods used t o r heat from a machine include cutting flu~ds, locating drive motors away from the of a machme, redunng fnct~on trom the ways and bearings, and spray coolrng c element of the machine. Most modern machining centers have automatic tool-change and auto work-transfer capability,so that workpieces can be loaded and unloaded while ma 1 s in process. Such a machine can cost over $200,000. Between these extremes merous machine tools that do less varied work than the highly sophisticated ma centers but that combine high output and mmimum setup tlme w~th remarkable fle (large number of tool motions prov~ded).

CNC TURNING CENTERS

The modernlathe has CNC control and tools mounted in turrets on slanted beds.The stock has been replaced by a live, powered spindle and chuck On some lathes the cept of automatic tool changing has been implemented The tools are held on a rota tool magazine, and a gantry-type tool changer is used to change the tools Each mag

FIGURE 26-16

This CNC turnlng center has a multiple-axis capability with two spindles and a 12-tool turret with X, Y, and Z control as well as axls control of the spindles.

holds one type of cutting tool. This is an example of the trend of providing greater versatility along with high productivity in 1athes.Theversatility is being further increased by combining both rotary-work and rotary-tool operations--turning and milling in a single machine.Live, powered, or Qiven tools replace regular tools in the turrets and perform milling and drilling operations when the spindle is stopped. See Figure 26-16.

OTHER NC MACHINES
Numerical control has been applied to a wlde variety of other product~on processes.NC turret punches with X-Y control on the table, CNCwire EUM machmes,laser and waterjet abrasive machinmg, flame cutters, and many other machines are readily ava~lable. Some new trends are be~ng observed in the development of machming centers, such as smaller, compact mach~tung centers with h~gher sp~ndle speeds, Machines with four- and five-axis capability ale readily available Modern machinmg centers have contributed s~gnlficantly to improvedproductivity m many compames.They have eliminated the tlme lost in moving workpieces from machine to machine and the ume needed for workp~ece loading and unloading for separate operations In addition, they have rnlnim~zedthe time lost in changing tools, carrying out gaging operations, and alignmg workpieces on the machine. The latest generation of niachmmg centers is aimed at further nnprovlng utihzatlon by reducing the time when machines are stopped, either dwing pauses in a shift or betweenshifts. Delays are caused by tool breaka&,-unforeseen tool wear, a limited number of tools, or an madequate number of ava~lable workpleces Machine.; are fitted wlth tool breakage monitors, tool wear campensatmg devices, and means for increasing the number of tools and workpieces avrulable. Probes on CNC machines can greatly improve the process capability of the machme tool.There 1s a big d~fference between the claimed program resolution for an NC machlne and the accuracy and preclslon (the process capabihty) of the actual par1s.A~ shown i n Figure 26-17, true posltion~ng accuracy and precision are affected by machine alignment, machine and fixture setup, variations m the workholding device. raw matesla1 variat~ons, workpiece location in the fixture variations. and cutting-taol tolerances.

Workoiece

Table centerline

Product to f~xture\. tolerance

and Mach~ne ftxture setup

FIGURE 26-17 Process capability in NC machines is

+ 0 001" IY)

0 001" (Z)

affected by many factors. Thus the finished workplece may he unacceptable even though the machlne i than capable ot producing the part to the design specifications The part program assurance that the part is properly located m the fixture or that the fixture is PI located on the table of the machine. However, a probe, carrled in the tool s magazine and mounted when needed :dn the spindle hke a cutttng tool, can establ location of the surface features relat~ve to each other and to the spindle axis 0.0005 m. (F~gure 26-18) The machine controller, using the probe data, will the the program reference data accordingly.The probc can be used to determine the a of material on a rough casting, locate a corner of a part, define the centcr of a 1 . check for the presence or absence of a feature. All of the var~ab~lity descn Figure 26-17 can be compensated for except for variations in the cutting-tool gel or tool wear.A probe mounted on the machine tool can be used to autoniatically tool-off~et data in the control computer.Thus the machine tool can function as, dlnate measuring machine By comparing the actual touched locatlon wlth tk gran~med locat~on, the measuring routine determines appropriate compensati~r

FIGURE 26-18 (a) Probe

carried in the tool changer can be mounted in the spindle (b) for checking the location of part features accurately.

SECTION 26.4

Ultra-H~gh-Speed Machining Centers (UHSMCs)

725

Many of the new machining centers are eqmpped wlth novel automatic pallet changers and workpiece loading and unloading devices. Robots are increasingly being applied for workpiece handling in machme groups, including manufacturing cells. In some cases the robot is also used for tool-changing functions.

ULTRA-HIGH-SPEED MACHINING CENTERS (UHSMCs)

Anyone involved in the product development process knows that the longest lead-ilme path from a new product (like a car) design to a finished product coming off the final assembly line always includes the die-making process. For example, m the automotive mdustry, each new car des~gn may have many die sets for forged and sheet metal pwts. Machining of the dies is a key process in the conventional die-making process (see Figure 26-19), where the malor steps are millmg, electronic discharge machining (EDM), polishing, die assembly, tryout, and modificat~on.
Hlgh-speed cutting In des~gning the component (a connecting rod). a dre set to foroe the comDonent rn-st also be n z i ynen A n c nctru,lc's nlilcle (mac:rlined lor tne EDM processc's

Electrode designing

~~h 1
Boring H~gh-speed cutting

I Elcctr0o*r"anLfacr.'r'ng

,
Tool steel for die

Elecrr.>nr~ proo..<cs oy ChC r~ og fo~ owe" by nano f n st>,ng\PO. so;ng

The new process uses hlghly accurate h ~ g h speed machtn~ n to g fabricate the dles set in two steps.

Boring The steel d ~ e block 1 s bored and the basic die shape machtned Into the block Thls 1 s the roughing stage

i
Structure cutting (milllngl

CNC m l l l ~ n g of the basic dle cavity

Electr~c discharge m a c h ~ n ~ n cuts g the die cavlty in the steel.

EDM cavity

Electropollshlng cavlty

The d ~ cavlty e receives a ftn~shed surface by electropollsh~ng.

FIGURE 26-19 The process t o manufacture d ~ e for s forgtng processes a shown on the left. Uslng ultra-h~gh-speedmachining centers reduces the sequence t o two steps

CHAPTER 26

Numerical Control (NC) and the A(4) Level of Automation Ultrd-hlgh-speed machining centers (UHSM mak~ngprocess lead time The new machining centers have h~ghly accurate highspindles capable of rpms of 30,000 to 50,000, cutting feed rates of 60 mlmin, and feed accelerahons of Y.8 m/sec2. The m a c h ~ n e requlres a sp~ndle with high s utilizing ceramic ball bearings with a constant-pressure preload mechanism lubr~catlon for superlor performance. Robust machrnlng centers of t h ~ sort s are capable of very high metal removal particularly m materials like aluminum. However, before investing in a high-spee chining center (HSMC), many issues must b e addre~sedThe new machine will u requlre the purchase of new cuttlng tools and tool rdtes, smaller-diameter tools are ed material. Make sure the insert retention Other major problems will Include chatter and re UHSMCs require exceptionally accurate, st~ff sp~ndles usmg ceramc hall rasing the posible rpm as compared to regu

under control and represent the pinnacle of machine tool development at t h ~ ~

Numerical control machmes, robots,FMS~, and computers are crit~cal elements oft vanced manufacturing technologies available for the next decade, which is being t as the time when computer-integrated manufac

Ultra Htgh Speed Machining Center

Construction of UHSMC with Horizorrtal Spindle

Capabilities of Machining Centers Conventional structure-two ball s

Construction of Spindle

Feed Mechanism of Conventional (x-axil MC vs UHSMC

FIGURE 2 6 2 0 Ultra-high-speed mach~ning centers (UHSMCs) are being developed w~th ceramic synchron~zed ball screws on the X-ax~s to reduce d~stortion (due to inertla) bearlngs in the sp~ndles, Speed Mach~n~ng Center", Toyota Technical the movlng components. (''Developmenf of Ultra H~gh Review, vol 49, No 1, September 1989.)

Review Questions

727

reality Computer techn~logy abounds: computer-a~ded design, computer-aided manufacturing with NC, CNC, o r AIC and DNC, computer-a~ded process plannmg, computera~ded testing and Inspection (CATI); artificial intelligence; smart robot^, and much more. But a word of caut1on:Any company can buy computers, robots, and other pieces of automat~on hardware and software.The secret to manufacturing success lies In the design of a srmple,umque manufacturing system that can achieve superior quality at low cost with on-tme dehvery and still be flexible. Flextbllity means the system can read~ly adapt to changes m the customer demand (both volume and mix) while quickly implementmg design changesThis requlres a visionary management team and a change in culture at all levels, startlng on factory floor with an empowered and mvolved workforce. Chang~ng how people work in a manufacturing system means you have to redesign the system. No better example of this can befound than theToyota Motor Company Led by their viceprestdent for manufacturing,Taiichi Ohno, who conceived, developed, and implemented Toyota's unique manufacturing system, this company has emerged as the world leader in car producQon. The implementation of this system has saved many companies (Harley Davidson,for example) and canied many others to the top posltion in thelr industry.The Toyota system 1s unique and as revolutionary today as were the American Armory System (jobshopb) and the Ford system (flow shops) in their day It IS significant that virtually every manufacturing system or technology citedin this chapter is practiced at Toyota. This new system is now being called lean productcon (to contrast it to mass production).Toyota does not use the word CIM because the computer is only a tool used in their system, a manufacturing system design that recognizes people as the most flexible element (Black and Hunter, 2003). Students of industrial, mechanical and manufacturing engineenng are well advised to be knowledgeable of this new nnlque system.

1II Key Words

contouring controller d~rect numer~cal control (DNC) distnbuted numerical control (DNC) feedback device feedforward flexlble manufacturing system (FMS) mach~ne control unit (MCU) mach~nlng center numer~cal control (NC) open-loop control pallet changer part program point-to-pomt machines transfer mach~ne turning center ultra-high-speed rnachlnmg centers (UHSMCs)

III Review Questions

1 . What human attribute is replaced by an NC machine? 2. Give an everyday cxample of a household device or appliance that exhibits feedback in its control system.

12. Explain what is meant hy inieryolaiion in NC programming. 13. Explain the problem of cutter ollset in NC programming by making a sketch showing an cnd mill cutting a perimctcr on a square platc. 14. Some of the functions perfor~iiedby the operator in piecepart manufacturing are vcry difficult to automate completely. Name the [unctions and explain why. 15. Why do you think there are no NC shapers?
\ ,
~

3. Explain how a toaster could be made a closed-loop device. 4. The first N(1 machines were closcd-loop control. Later some machines were open loop. What change did this require on the part of machine tool builders? '5. Can a continuous-path NC machine he open loop'?Why or

~-

18. 19.

9. Many manufacturers have purchased large machining centers tied to an ASRS. Go on the Internet to find an example of an ASRS, and explain how such a system works. 0. I-low docs fccdforward differ from leedback in a process
leadscrews on their machines when they made them into NC

20. 21. 22.

23.

24.

used on CNC machines. Which method for position lerdhack is the iiiost accurate? What is the dillercnce between the zero reference point and the machine zero point? What is an encoder? What is a peck-drilling subroutine for a CNC machine? What is pocket milling, and what kind of milling cutter is usually used to perform it'? How are probes used in CNC machines to improve process capability? What are G words used for in NC'?

728

CHAFSER 26

Numerical Control (NC) and the A(4) Level of Automation 26. What major changes are being introduced into UHSM

25. flexible manufacturing systems use CNC machines for processing and AGVs, robots, o r conveyors lo transport parts. So what diffcrentiates the FMS from a transfer line?

El Problems
1. What are the X and Y dimensions for the center position of holes 1.2, and 3 in the part shown in Figure 26-7? 2. Configurations obtained from continuous-path machining are the result of a series of straight-line,parabolic-span or higherorder curves.The degree to which curved surfaces correspond to their design depends on how many lines or spans are used. Four equal chords in a circle describe a square (Figure 26-A). Six make a hexagon, and, as the number increases, the lines themselves come closer to a perfect circle.The number of lines needed is determined by a maximum tolerance allowed between the design of the curved section and the actual chord programmed. This is the dimension T. The program for a
origin at the zero rcference point. Move the tool aro plate counterclockwise.

ance o f T = 0.001 in., how long should the span be for a curve with a 5-in. radius.'? Assume that the arc is part of a circle. Whai is the span angle hcre,in degrees?

milling job described in Problem 4.lhns down milling to be used. Rewrite the NC program points to accom

will require turning,facing, grooving,boring, and du is machimed from a casting.See Figure 26-C. Unfortu

prepared, you have developed some cost data for the ture of the part by four different lathe processes (see A). Complete the table by determining the run costp the cost per unit at the various quantities, and the total batch. Answer the following questions regarding this si

FIGURE 26-A

3. In Problem 2,suppose that the acceptable tolerance was 0.0001 in. Determine the span angle. 4. Suppose that the plate shown in Figure 26-B was to b e profile milled around the periphery with a 1-in.-diameter end milling cutter. The dashed line is the cutter path. The programmer must calculate an offset path to allow for cutler diameter. Since the programmed points are followed by the cutter centerline and the profile is made at the tool's periphery, the programmer called for a '/,-in. cutter offset. Working with computer assistance, the programmer would describe the part profile to be machined and specify the cutter.The computer would generate the cutter path. Complete the table below to specify the culler path, starting with the

FIGURE 26-8

FIGURE 26-C

1. Of the four costs listed for each process, which costs a t e fixed and which are variable'? 2. What is included in the cost per unit? Do you need to estimate the machining time per piece and the cycle time per piece, including the time to change parts and setups in order to compute run cost? 3. How would yon go about estimating this time, and what time elements might b e included in the cycle time in addition to the machining time? 4. How would you use this estimate of time in the cost table? Show the calculation. 5. Make a plot of cost in dollars versus quantity, with all four methods on one plot. 6. Make a plot of cost per unit versus quantity, again with all four methods on one plot. Find the break-even quantities. (Hint: Did you plot the data on log paper?) 7. Discuss the break-even quantities that you found in part 5 versus part 6. (They should be the same.) 8. When would you use the NC lathe? The turret lathe? When would you use the turret lathe if you have n o NC lathe?

Make Ouantlb 10,000 Umts

Cost to produce on six-spindle antomatie

1,000 Umts

100 Un~ts

10 Un~ts

1 Unit

Total cost of batch

Eng~neenng 2 5 h r at $40 /hi Toollng (cutting toolsand workholdeis) Setup 8 hr at$15Atr Run cost per batch 50 per piwe Cost each Cost to produce on hrrret lathe Total cast of batch Eng~neenng 2 h i at $201hr Taohg Selup 4 hr at $20lhr Run cost per batch $8 per piece Cost each

Cost to produce on engine lathe

ToEal cost of batch Engmeenng 1 h r at $20/hr Toohng (no coat) Setup 2 hr at$12 001hr Run cost per batch $12 Cost each
Coy1 to produce on NC lathe

Total cost of batch Eng~necnng and prograrnrnlng

Tooling Seiup 1 hr at $20/hr Run Loat per batch $2 psi plece

Cost each

I
lgure CS 26a shows a part design for a small flange to F b e made out of 1020 steel. Prepare a sequence of operations or process plan to make this part. Note that you need to machine the top, bottom, and all the holes. Specify the machines and the toois YOU would Use assuming YOU select bar stock as your raw material.The term Co bore in the drawing means counter bore. Check with Your instructor for an example of a process plan. Assume Me W size here is 1200 perwyear made in lots of 120 every month. Now assume that the designer provides you a design like CS 26b and calls for the flange to be made from cast iron. The casting will probably have a large hole in the center cored during the casting process. What casting process will you use? Prepare another process plan for the cast part. Will the sequence of operations be the same? What about the cost per unit? Which will be low 1.The part drawings failed to provide a critical specifics. tion in the design,what isit? 2,~he drawing failed to show the final geometry (in the I view) correctly. Redraw the part to show these corrections. 3, AS the process engineer, how would you advise the design engineer he screwed up twice?

Dimensions in mrn lin.)

Figure CS 26b

Figure CS 26 Two different designs of a flange. The design on the right suggested by Doyle et ai. in 3rd edition of
Manufacturing Pmcess and Materials for Engineers, Prentice-Hall.

Br&Design (The GuttingTool) .1 INTRODUC~ON Broaclling Speeds,A~c~lraey~ TO SHAPING '.2 INTRODUCTION AND PLANL~G Fbkh Maciune Tools for Shaping Broaching Materials and Planing Machines Sharpening Broaches Construction Workholding and Setup on MACHINES Planers 27.5 BROACHING 27.6 INTRODUCTIQN TO SAWING 1.3INTR~DUCT~ON T~ BROACHING v . 4 FUNDAMENTALS OF BROACHING Saw Blades Types of Sawmg Machines The Advantages and Limitations of Broaching

Power Hacksaws Bandsawing Machines Cuttlng Fluids FeedsandSpeeds Cecular-Blade Sawing Machines TO FILING 27.7 INTRODUCTION F~ling Machines ESTIMATMGCase Study: COST PLANING vs MILLING

2 7.1 INTRODUCTION
While mill~ng, drilhng, and turning make up the bulk of the nrachin~ng processes, there are many other ctupmaking (metal removal) processes This chapter will cover shaping, planing, broaching, sawing, and filing.

27.2

INTRODUCTION TO

SHAPING AND PLANING

The processes of shaping and planmng are among the oldest smgle-point machming processes. Shaping has largely been replaced by milhng and broaching as a production process, while planing still has applications in producing long flat cuts, l i e those in the ways of machine tools From a consideration of the relat~ve motions between the tool and the workpiece, shaping and planing both use a stra~ght-line cutting motion w ~ t h a single-point cuthng tool to generate a flat surface In shaping, the workpiece is fed at right angles to the cutting motlon between successive stroke of the tool, as shown m Figure 27-1, wheref, is the feed per stroke, V a the cutting speed, and d is the depth of cut (DOC). (In planing, d~scussed next, the workpiece 1s reciprocated and the toolls fed at r~ght angles to the cutling mot~on.) For either shapmg or planing, the tool is held in a clapper box, which prevents the cuttlng edge from bemg damaged on the return stroke of the tool in addition to pla~n flat surfaces, the shapes most commonly produced on the shaper and planer are those illustrated in Figure 27-2. Reiatlvely skrlled workers are requrred to operate shapers and planers,and most of the shapes that can he produced on them can also be made by much more product~ve processes, such as mlllmg, broaching, or gnnd~ng. Consequently, except for certam specla1 types, planers that w~ll do only planlng have become obsolete.Today,shapers are used mainly m tool and d ~ work, e m very low volu~lle production, or in the manufacture of gear teeth In shaping, the cutting tool is held in the tool post located in the ram, which reciprocates over the work with a forward stroke, cutting at velocity V and a quick return stroke at velocity VR.The rpm of d r ~ v e crank (N,) drives the ram and determines the velocity of the operation (see F~gure 27-ld). The stroke ratlo is

R, =

cutting stroke angle 360"

200" 360"

5 9

732

CHAITER 27

Other M a c h ~ n ~ n Processes g Vert~cal slide

cunln~ too' Unfinrshed surface

Y A i n g l e - p o ~ ntool t process
(a) Basic geometry for shapjng and plantng
(c)

Column

T O O post Tool

head#E

Speed

+--+ Feed
Motion
1
$ ',

(dl

,' ' ' /

\

! '

Shaper (quickreturn) mechar for drlvlng tool past work

of shaplng and plan~ng(a) The cuttlng speed, C: and feed per stroke f,. (b) Block diagram of the FIGURE 27-1 Bas~cs rnachlne tool (c) The cuttlng tool is held in a clapper box so the tool does not damage the workpiece on the return stroke. (d) The ram of the shaper carrles the cutting tool at cutting velocity Vand reciprocates at velocity Vn by the rotatlon a bull wheel turning at rpm n,.

Grooves

T-slot

Dovetails

Flats and angles

The tool is advancing 5 S 0 ' of the time.The number of strokes per mlnute is determined by the rpm of the drive crank. Feed IS in inches per strokc and is at rlght angles to the cuttlng direction As in other machrning processes. speed and Ceed are selected by the operator The length of strokc 1 must be greater than the length of the workplece (or length of cut) L, since velocity is posltlon variant Let 1 = twice the length of the block being cut,or2L The cutting velocity Vis assumed to be twlce the average forward velocity V or the ram.The genela1 relationship between cuttmg speed and rpm 1s

where D ir the dlameter (of the rotatmg bull wheel) in mches. For shaping, the cutting speed ib

Once a cutting speed 1s selected, the rpm of the machine can be calculated.Tables tor suggesled feed values,f'.are in inches per stroke (or cycle),and recommended depths of cut are also avalabla. The maximum depth of cut is based on the horsepower available to form the chtps.This calculat~on requires that the metal removal rate (MRR) he known. The MRR 1 s the volume of metal removed per unit tune

MRR

= -in.'/min

LWd
Tm

(27-4)

where W is the width of block being cut and L is the length of block bemg cut, so volume of cut = W L d ,where d IS the depth of cut and T," 1s the tlme in minutes to cut that volume. In general, T,, is the total length of the cul divlded by the feed rate. For shaping, T,,, is the width of the block dlv~ded by the teed rate f, o i the tool moving across the w~dth.Thus, for shap~ng,

Also,

where the number of strokes for the job is for a surface of width W.

MACHINE TOOLS FOR SHAPING

Shapers, as machme tools, are usually dasslfied according to their general design features as follows:

1. Horizontal a. Push-cut
b. Pull-cut or draw-cut shaper

FIGURE 27-3 The most widely used shaper 1 s the hor~zontal v.!,t,-cu<rr,,jcrline loo, ctlown nerc w t h no roo. it1 uie too post.
2. Vertical a. Regular or slotters

b. Keyseaters 3. Specla1 They are also class~fied as to the type of drive employed. mechanical drrve or h) drive. Most shaperp are of the hurrzonmlpush-cut type (F1gure27-3),where cut1 curs as the rampu~hes the tool across the work On horizontal push-cut shapers, the work is usually held in a heavy vlse m on the top s ~ d of e the table Shaper vise7 have a very heavy movable jaw, because must often be turned so that the cutting forces are directed against this jaw. In clamping the workpiece in a shaper vise, care must he exercised to ma that it rest5 sol~dly agalnst the bottom of the v~se (on parallel bars) so that it will deflected by the cutting force and so that it is held securely yet not distorted clamping pressure Most shaping is done with s~mple high-speed-steel or carbide-tipped cutting 1 held in a heavy, forged tool holder. Although shapers are versatile tools, the prec the work done on them 1s greatly dependent on the operator. Feed d~als on shag nearly always graduated in 0.001-in d~visions, and work 1 s seldom done to greate slou than this. A tolerance of 0.002 to 0.003 i n i s desirable on parts that are to hem, on a shaper, because t h ~ s glves some provision for variations due to clarnping,~ looseness or defle~tion of the table, and deflection of the tool and ram during cu

PLANING MACHINES Planing can be used to produce hor~zontal, vertical, or ~nclined flat surfaces a workp~eces (too large for shapers). However, planing n much less efficient tha h a w machning processes, such as milling,that will produce such surfaces. Consel planing and planers have largely been replaced by planer m~lling machines or m that can do both m~llmg and planing. Figure 27-4 shows the basic components and motions of planers. In most I the action is opposite to that of shaping.The work 1s moved past one or more ary smgle-point cutttng tools Because a large and heavy workpiece and tablet rec~procated at relatively low speeds, several tool heads are provided, often wit1 ple tools ~neach head. In addition, many planers are prov~ded with tool heads a] so that cuts occur on both directions o f the table movement. However, hecau smgle-point cutting tools are used and the cutting speeds are quite low,planers In productivity as compared with some other types of machlne tools

SECTION 27.2

Introduction to Shaping and Planing

735

Bed

--r

Feed

Mot~on
lb)

ial

lock diagram
shoolng the bas~c components of a double-housing planer

lock dlagram of a n open-slde planer

c-Feed

Mot~on

Schernat~c of ners. (a) Double-hous~ng ner with multiple tool heads and a large reciprocating le; (b)s~ngle-housing or $n-sldedplaner; ~nterchangeabie multiple tool der for use in planers. lotograph courtesy Gebr 3hrlnger GmbH.)
,URE 27-4

the most common double-hou~mg and smngle-housmgtypesThe Figure 27-4 dep~cls double-housing has a closed-housing structure, spanning the reciprocatmg worktable, with a cross rail su~norted at each end on a vert~cal column and carrymnE two tool heads An additional tool head is usually mounted on each column, so thai four tools (or four sets) can cut during each stroke of the table. Obviously, the closed-frame structure of this type of planer lmmits the sire of the work that u n be machmed. Open-side planers have the cross rail supported on a slngle column.Th~s design provide< unrestricted access to one side of the table to permit wlder workpieces to be accommodated. Some open-side planers are convert~ble,m that a second column can be attached to the bed when desued so as to provlde added suppoi t for the cross rail

WORKHOLDING AND SETUP ON PLANERS

Workpieces in planers are usually laige and heavy.They must be secul ely damped to re-: sist large cutting forces and the high-inertia forces that result from the rapid velocity? changes at the ends of the strokes. Specral stops are prov~ded a t each end of the work-: piece to prevent i t from sh&ing.

Considerable time 1s uvually required to set up the planer, thus reducing the the machine is available for producmg chlps Someumes special setup plates are use qulcksetup of the workpiece.Another procedure 1s to use two tables. Work isset one table wh~le another workpiece is being machined on the other The tables can b tened together for machining long workpieces. The large workpieces can usually support heavy cutting forces, so large dep cut arc recommended, which decrease the cutting time. Consequently, planer too ally are qwte massive and can sustain the large cutting forces Usually, the m u s h the tools is made of plain-carbon steel, wlth rips made from high-speed steel or c Chip breakers should b e used to avoid long and dangerous chips in ductile materi Theoretically, planers have about the same precmon as shapers.The feed and d~menslon-controlling dials are usually graduated 1n 0.001-in. divisions. However, cause larger and heavier workpleces are usually involved, and much longer beds tables, the work~ng tolerances for planer work are somewhat greater than for shap

INTRODUCTION TO BROACHING
es.The machime tool is called a broaching machine or a broach and the cutting tobl called the broach. Figure 27-5 shows the basic shape of a conventional pull hro this figure, P = pitch,ns is the number of semiroughing teeth, and nf is the numbe ishing teeth where the rise per tooth gets smaller from rough to finish.

a finished part in one pass.The heart of this process lies in the broaching tool, roughing, semifinishing, and finishing teeth art: combined into one tool, as s

'

See Figure 28-6 for

Section A-A'
(b)

Total broach length

P- pitch of teeth D - depth of teeth (0.4P) L- land behind cutting edge 10. R - radius of gullet (.25P)

hook angle or rake angle backoff angle or clearances R P T rise per tooth (chip load), t ,
w-

y-

(a) Photo of pull broach. (b) Basic shape and nomenclature for a conventional pull (hole) r roughing teeth; n, = numb broach. Section A-A' shows the cross section of a tooth. P = pitch; n ~ n u m b e of of semifinishing teeth; n,= number of finishing teeth.
FIGURE 27-5

Body of broach
f -

Body of broach traveling a t speed V

FIGURE 27-6 The feed II? broach~ng depends on the rlse per tooth t, (RPT). The sum of the R P T glves the depth o f cut, DOC P = pitch of teeth; D = depth of teeth (0.4P), L = land beh~ndcuttlng edge (0 25P), R = rad~us of gullet (0 25P), u = hook angle or rake angle; y = backoffangle or clearance angle.

it is produced with a singlelinear stroke of the tool across the workpiece (or the workpiece across the broach). Broaching competes economically with milling and boring and is capable of producing precision-machined surfacecesThe broach finishes an entire surface in a single pass. Broaches are used in production to finish holes, splines,and flat surfacesTypica1 workpieces include small to medium-sized castings, forgiogs,screw-machineparts, and stampings. This rise per tooth (RPT), also known as step o r the feed per tooth, determines the amount of materialremoved. No feeding of the broaching tool is required.Thefronta1 contour of the teeth determines the shape of the resultingmachined surface. As the result of these conditions being built into the tool, no complex motion of the tool relative to the workpiece is required and the need for highly skilled machine operators is minimized. Figure 27-7 shows apullbroach in a vertical pull-down broaching machine.The pull end of the broach is passed through the part, and a key mates to the slol.The broach is pulled through the part.The broach is retracted (pulled up) out of the part.The part is transferred from the left fixture to the right fixture. One finished part is completed in every manufacturing cycle.

I n broaching, the tool (or work) is translated past the work (or tool) with a single stroke of velocity V.The feed is provided by a gradual increase in height of successive teeth. The rise per tooth varies depending on whether thc tooth is lor roughing (tJ,semifinishing (t,),or final sizing or finishing (tf). In a typical broach there are three to five semifinishing and finishing teeth specified.The number of roughing teeth must be determined so that broach length, which is needed to estimate the cutting time, can be calculated. Other lengths needed for a typical pull broach are shown in Figure 27-6. The chip breakers in the first section of roughing teeth may be extended to more teeth if the cut is heavy or material difficult to machine.The distance between the teeth,called the pitch P,is important because it determines the tooth construction and strengths and the number of teeth actually cutting at a given instant. It is preferable that at least two teeth be in contact with the workpiece at any instant. The pitch or distance between teeth is where length of cut usually equals L,,as shown in hgure 27-5. The number of roughing teeth is

t, where DOC is the total amount of metal to be removed and is the nse per tooth

n, =

DOC - n,t,

n,tf

738

CHAITFR 27

Other Mach~ning Processes

Rotor t ~ e t h Flnishlng teeth

(a)Rotor- or lump-tooth broach des~gn

(b) Round, push-type broach w~th chlp-break~ng notches on alternate teeth except at theflnlshlng end

Rel~evedporf~on of cutting edge

End

vrew

TOP vlew

Side view

(c) Notched tooth, flat broach

(dl Progress~ve surfacebroach. (Courtesy of Detroit Broach & Machrne Company)

FIGURE 27-7 Methods to decrease force or break up chip rings in broaches (a) Rotor or jump tooth, (b) notc Increasing c h ~ p load); tooth, round, (c) notched tooth, flat design (overlapping teeth perrnlt large RPTs w~thout (d) progresswe tooth des~gn for flat broach.

The overall length of the broach for a pull broach 1s

LB = (n, + n,

+ n,) P + L, + L,,

The length of stroke 1s L = L, - L,, in inches, if the broach moves past 1 L , + L, if the work moves past the broacb.The cutting tvne is

where V is the cuttlng speed, in surface feet per minute. The metal removal rate depends upon the number of teeth (roughing)* ing the work. MRR (per tooth) = lZt,WV m?/m~nperroughing tooth where W is the w~dth of broach tooth.

SECTION 27.4

Fundamentals of Broaching

739

The number of roughing teeth in contact with part n than the part.

-- L,/P

for a broach longer

(27-12) MRR (for process) = 12t,WV in?/min where n is usually rounded off to the next-largest whole number. The pull broach must be strong enough so that it wdl not be pulled apart. The strength of a pull broach is determined by its minimum cross section, which occurs either at the root of the first tooth or at the pull end:

allowable pull

area of minimum section x Y.S. of broach material (27-13) factor of safety

where Y.S.is yield strength. The push broach must be strong enough so that it will not buckle. If the length-todiameter ratio, WD,, is greater than 25, the broach must be considered a long column that can buckle if overloaded. Let L = length from push end to first tooth, D,= root diameter at 0.5L, and S = factor of safety: allowable load 13.5
=

106 X D; SL2

For W D , less than 25, the normal broach loads are not critical. Calculation of the total push or pull load depends upon the number of teeth engaged, n estimated from LJP; the width of the cut, W,the RPT per tooth engaged, t,; and the shear strength of the metal being machined. The force necessary to operate a broach depends upon the material being broached, the conditions of the tool, and the nature of the process.Au empirical constant is required in the force calculation to account for the large amount of rubbing (friction) between the tool, the chips captured in the tooth gullet, and the workpiece. Let FCBbe the broach pull force in pounds: where 7$1sflow stress. T , found in Chapter 20, depends upon the hardness for the metal.Thisforce estimate can be used to estimate the horsepower needed for the broaching machine.

'

THE ADVANTAGES AND LIMITATIONS OF BROACHING Because of the features built into a broach, it is a simple and rapid method of machin- . ing.There is a close relationship among the contour of the surface to be produced, the , amount of material that must be removed, and the design of the broach.For example, the total depth of the material to be removed cannot exceed the total step provided in , the broach, and the step of each tooth must be sufficient to provide proper chip thick- ; ness for the type of material to be machined. Consequently,either a special broachmust be made for each job or the workpiece must be designed so that a standard broach can be used. Broaching is widely used and particularly well suited for mass production he- ;, cause the volume can easily justify the cost of the broaching tool, which can be easily I ' $15,000 to $30,000 per tool. It is also used for certain simple and standardized shape$ ! such as keyways, where inexpensive standard broaches can be used. Broaching was originally developed for machining internal keyways.However,its obvious advantages quickly led to its development for mass-production machining of var- ', ions surfaces, such as flat, interior or exterior, cylindrical or semicylindrical, and many ! irregular surfaces. Because there are few limitations as to the contour form that broach I teeth may have, there is almost no limitation in the shape of surfaces that can be produced by broaching.The only physical limitations are that there must be no obstruction ' to interfere with the passage of the entire tool over the surface to be machined and that the workpiece must be strong enough to withstand the forces involved. In internal ; broaching, a hole must exist in the workpiece into which the broach may enter. Such a hole can be made by drilling, boring, or coring. Broaching usually produces better accuracy and finish than can be obtained by drilling, boring, or reaming.Although the relative motion between the broaching tool and
8 8

Processes C H ~27R Other fvlach~ning the work usually is a single linear one, a rotational motion can be added to permit broaching of splral splines or gun-barrel rifling.

BROACH DESIGN (THE CUTTING TOOL)

Broaches commonly are classified by the following design features Purpose
S~ngle Comb~nation

Motton
Push Pull Stationary

Construct~on
Solid

Functton
Rough~ng Slvng Burnlsh~ng

Bullt-up

Figure27-5 shows the principal components of a pull broach and the shape and arr ment of the teeth. Each tooth is essentrally a single-edge cutting tool, arranged muc the teeth on a saw except for the step, whch determines the depth cut by each toot shown in Figure 27-6. The rise per tooth, which determines the chip load, vanes fr about 0.006 in. for roughing teeth III machmlng free-cuttmg steel to a minunum 0 001 for finishmg teeth.Typically the RF'I'is 0.003 to 0.006 in. in surIace broachlng and 0.0 L o 0.0025 in. on thc diameter for internal broaching.The exact amount depends on era1 factors Too-large cuts impose undue sfresses on the teeth and the work; too-s cuts result in rubbing rather than cutting action.The strength and ducthty of them being cut are the primary factors. Where it is desirable for each tooth to take a deep cut, as in broachlng casti or forgings that have a hard, abrasive surface layer, rotor-cut or lump-cut tooth desl may be used (Figure 27-7). In this design, two or three teeth in succession have the s diameter, or height,but each tooth of the group is notched or cut away so that it cuts a portlon of the circumference or width.This permits deeper but narrower cuts bye tooth without increasing the total load per tooth.This tooth design alro reduces the 0 and the power requirements. Ch~p-breaker notches are also used on round broache break up the chips (Figure 27-7b) A similar idea can be used for flat surfaces Tooth loads and cuttlng forces als be reduced by using the double-cut construction, ~hown in Frgure 27-7c Four con tive teeth get progress~vely wider.The teeth remove metal over only a porhon oft wldth until the fourth tooth completes the cut Another technique for reducing tooth loads utilrzes the principle illustrate F~gure 27-7d. Employed primar~ly for broaching wide, flat surfaces, the first few teet progressrue broaches completely machine the center, while succeeding teeth are o In two groups to complete the remainder of the surface. Rotor, double-cut, and gressive designs rcquire the broach to be made longer than if normal teeth were us and they therefore can be used only on a machlne having adequate stroke length. The cutting edges of the teeth on surface broaches may be either normal to the rectlon of motron or at an angle of from 5" to 20QThe latter, chear-cut broaches,pro smoother cutting action with less tendency to vibrate. Other shapes that can be broa are shown in Figure 27-8 along with push- or pull-type broaches used for the job. The pitch of the teeth and the gullet betwccn them must be sufficient to pro ample room for the chips. All chips produced by a glven tooth durlng its passage the full length of the workp~ece must be contained in the space between successive t At the same hme, it 1s desirable to have the pitch sufficiently small so that at le two or three teeth are cutnng at all times The hook determines the primary rake angle and is a functron of the mate be~ng cut. It is 15" to 20" for steel and 6' to 8" for cast iron. Back-off or end clear angles are from 1" to 3"to prevent rubbing Most of the metal removal is done by the roughcng teeth. Semifinzsh~ng teeth p vide surface smoothness, whereasfinzshing teeth produceexact size. On a new broac the finishing teeth are usually the same slze. As the first finishing teeth become those behind continue the sizing function. On some round broaches, burnzshzng are provided for finishing.These teeth have no cuttlng edges but are rounded di

FIGURE 27-8 Examples of push- or pull-type broaches. (Courtesy of RuMont Corporotfon.)

hard steel or carblde that arc from 0.001 lo 0.003 in. larger than the size of the hola.The resulting rubblng action smooths and sizes the hole They are used primarily on cast iron and nonferrous mctals. The pull end of a broach provldes a means of qu~ckly attaching the broach to the pulling mechanism The frontpilot allgns the broachin the hole before it begins to cut, and the rearpllor keeps the tool square with the finiqhed hole as it leaves the workpiece. Shank lcnngth must be ~ufficient to permit the broach to pays through the workpiece and be attachcd lo the puller before the first roughing tooth engage3 the work If a broach is is necessary. to be used on a vcrlicdl machine that has a tool-handhng mechanism, a1~1~1 A broach should not be used to remove a greater depth of metal than that for which it is designed-the sum of the steps of all the teeth. In designing workpieces, a mmimum of 0 020 in. should be provided on surfaces that are to be broached, and about 0 025 in. is the practical maximum.

BROACHING SPEEDS,ACCURACY, FINISH Depending on the metal being cut, cutting speeds for hroachingrange from low (25 to 20 sfpm) to high while complet~ng the surface in a s~ngle stroke, so the productivity is high. A complete cycle usually requires only from 5 to 30 seconds, with most of that time being taken up by the return stroke, broach handling, and workpiece loading and unloading. Such cuttmg conditions facilitate cooling and lubrication and result in very low tool wear rates, whlch reduce the necessity for frequent resharpening and prolong the life of the expensive broachlng tool. For a glven cutting speed and matenal, the force required to pull or p u ~ h a broach is a funchon of the tooth width, the step, and the number of teeth cuttlng. Consequently, it is necessary to design or spccify a b~oach within the sdoke length and power iim~tations of the machlne on which it 1s to he used. The average machining preclslon is typically iO.OO1-in.(20.02-mm) tolerance w ~ t h ~url'ace finish 120 to 60 RMS or better. Burrs are mlnlmal on the exit slde of cuts.

742

CHAP^ 27

Other Wlach~n~ng Ptoc~ses

FIGURE 27-9

Shell

construction for a pull broach.

BROACHING MATERIALS AND CONSTRUCTION Because of the low mtting speeds employed, most broaches are made of allc speed tool steel. Carbide-tipped broaches are seldom used for machining sR forgings, as the cutting edges tend to chip on the first smke, probably due ti rigidity in the combinahon of machine tool andcvtting tool.TNcoating of hi sted [IXSS) broaches islxcoming marecommon,greatly prolonging the l i e d When they are are& in cont~nuous mass-production machineses particularly~ broaching of cast iron, tungsten carb~de teeth may be used,permitting the hrc used for langp.eriods of time without resharpening. Internal bmachesare asually solid bur may be made af~hellsmbnnted 01 (Figure 27-9).When the broach (or a section of it) is subject to rapld wear, a s can be replaced.This will be much cheaper than replacing an entire solid brc construction, however,is initially more expensive than a solid broach of compa Small-surface broaches may be of solid construction, but larger ones u modular construction (Figure 27-10). Building in sections makes the broach cheaper tp constmet and sharpen. It also often provides sohe degree of int ability QE the sections for differentparts, bringing down the tool cost sjgnifici SHARPENING BROACHES Most broaches are resharpened by @ding; the hook faces of the teeth.%@1; ternal broaches must not be reground because this would change the size of tl Lands of flat-surface broaches are sometrmes ground, m which caseall ofthe ground to maintain their proper relationship.

l the factors that dcfermine the shape of the machined surface ar Because d tennine a l l cutting conditions except speed are built Into the broaching tool machines are relatively simple.Their primary functions are to impart plain rec motion to the broach and to provide a means for handl~ng the broach arrtom+

WGUREW-10 A

constructed broach to build and can be In sectloris.

Vertical
Push-broaching
Pull-down

Pull-up
Smiace

Arbor press with guidad ram 5- to SO-ton capanty Internal broaching Double ram d&gn most cornmod Long changmvet bmes Ram above table pulhg broach up Madines witb mult~ple rams common No handlmgof broach

Muinple shddes

Horizontal
Pull

Short CycleTunes
Longer atsok~e. and brnacbes vertical machmes laid on side Bas~cally Broachesstat~onary, work muves on mveyor Work held in fimtes Conveyor c h a ~ n bdds M u m Xotary broach stationary, work translates besleathrool Work held m fixtures

Surface
Conrlouous Rotaty

Most broaching machines are driven hydraulically, although mechanical drive is used in a few special typt?s.The major classificationrelates to whether the motion of the broach is vertical or horizontal, as givea in Table 27-1. The choice between vertical and horizontal machines is dete~mined primarily by the length of the stroke required and the available floor space Vertical machines seldom have strokes greater than 60 in. because of height limitations.Horizontalmachinescan have almost any length of stroke, but they require greater floor space The most common machine is the vertical pull-down machine shown m Figure 27-11. The worktable,usually having a~~herical-seated workhalder,sit$ below the broach elevator,withapullingmechanism below the table. When the elevator raises the broach above the table, the wmk can be placed into positiou.Ihe elevator then lowers the pilot end of the broach through the hole in the workpiece, where it is mgngirged by the puller.The -elevator then releases the upper end of the broach, and it is pulled through the workpiece. The workpieces are removed from the table, and the brbach is raised upward to be engaged by the elevatar mechanism In some machines with two ram$ one broach is bang pulled dgwn while the work is being unloaded and the broach raised at the other station. In Figure 27-11, the part is being broached in two passes, first on the leftsthenon the right.

Sawtng is a basic machining process in whi~h chips are produced by a succession of

small cutting edges, or teeth, arranged in a narrow line on a saw "blade." Gs shown in Figure 27-12, each tooth forms a chip progressively as it passes through the warkpiece. The chips are contamed within the spaces between successive teeth until the teeth pass from the work. Because sectmns of considerable size can be severed $.om the workpiece with the removal of only a small amount of the material in the farm of chihips,sawing is probably the most ecmomlcd of the b s i c machining processes with respwt to the waste of material and power coasl~mption, and in many cases with respect to labor. In recent years vast lmprovements have been made m saw blades (design and maienals) and sawing machine$ resulting in improved accuracy and precision of the process. Most sawingis done to sever bar stock and shapes into desrred lengthsfor use in other operations. These are many cases in which sawing is used to produce desired ior producing only a fc\v p;rrts. contour s.r\\ing may shape>.Fr~yucntl!..i~iid cs~)~'~iaIly bc r n c ~ r cconcrnii~.al ~. thrin nny olher rnxhining pruccss.

SAW BLADES
Saw blades are made m three baste configurations, The first type, commonly calked a hacksuw blade. is straight,relatively rigid, and of limited length, with teeth on one edge.

744

CHAPTER 27

Other Machining Processes

FIGURE 27-11

Vertical pulldown broaching mach~ne shown w~th parts in position ready for the two broaches to be inserted. A n extra part is shown lying at the front of the machine.

Workpiece

FIGURE 27-12 Formation of

chips in sawing.

flex~ble that a long length can he f i The second type, called a bandsuw, is sufiic~ently into a continuous band with teeth on one edge. The thrrd form 1 s a ngld d ~ s k 1 teeth on the penphery; these are called c~rcular saws or cold saws. F~gure 27-13 g11 standard nomenclature for a saw blade.

T h e best all-purpose 'welded-edge blade for 8awing varying sires, shapes, and cross sections. Cobalt-tough forcuaing wide range

For high-production cutting of solids, superalloys, tool steels, high-temperature alloys. Welded to length and coil stock.

The original and widely used welded-edge band blade for general-purpose sawing. Welded to length and coil stock.

Hardened back provides greater beam strength for more accurate sawing. Welded to length and coil stock.

Recommendedfor contour saws running over 3000 SFPM. Welded to length and coil stock.

Standard design has zero rake

Skip-tooth blade cieatschips, cuts nonferrous

sections

P
Raker set has a straight tooth between one left and one right

Variable pitch can change by section or individually

Wavy set for thin sections has progressive set, both directions

Straight set-left, finish

Cluster set has only a few straight teeth

then right-is for better

P
Variable pitch with 5" rake is more aggressive

P
Variable pitch with 10" rake sheds chips better

All saw blades have certain common and basic features: (1) material, (2) tooth form, (3) tooth spacing, (4) tooth set, and ( 5 ) blade thickness or gage. Small hacksaw blades are usually made entirely of tungsten or molybdenum high-speed steel. Blades for power-operated hacksaws are often made with teeth cut from a strip of high-speed steel that has been electron-beam-welded to the heavy main portion of the blade, which is made from a tougher and cheaper alloy steel (see Figure 27-13). Bandsaw blades are frequently made with this same type of construction but with the main portion of the blade made of relatively thin, high-tensile-strength alloy steel to provide the required flexibility. Bandsaw blades are also available with tungsten carbide teeth and TiN coatings.The three most common tooth forms are regular, skip tooth, and hook. Tooth spacing is very important in all sawing because it determines three factors. First, it controls

746

CHA~F 27 R Other Mach~ning Processes

the size of the teeth. From the viewpoint of strength,large teeth are desirable.Se toothspacing determines thespace (gullet) available to contain the chip that is fo The c h ~ p cannot drop from t h ~ space s unt~l it emerges from thc slot cut in the piece, called the kerf. Toosh set refers to the manner in which the teeth are offset the centerline in order to make the kerf wider than the gage (the thickness of the of the blade.This allows the saw to move more freely in the kerf, reducing rubbing, hon, and heating The kerf-gullet space must be such that there is no crowding of the Ch~ps should not become wedged between the teeth and not drop out of the gullet the saw emerges from the cut. Third, tooth spacing determines how many teeth will bear against the wo is very important in cutting thm matenal, such as tubing. At least two teeth sho in contact with the work at all times If the teeth are too coarse, only one tooth re the work at a given t~me, permitting the saw to rock, and the tccth may be st< from the saw Hand hacksaw blades have 14 to 32 teeth per inch In order to make it easier to a cut, some hand hacksaw blades are made with a short sectlon at the forward end ing teeth of a special form w ~ t h negative rake angles.Tooth spacing for power ha blades ranges from 4 to 18 teeth per inch. Raker-toothsaws are used in cutting most steel and iron. Sfrazght-set teeth ar for sawing brass, copper, and plastics. Saws w ~ t h ware-set teeth are used pi-mar cutting thin sheets and thm-walled tubing. The gage or blade thzckness of nearly all hand hacksaw blades is 0 0 blades for power hacksaws vary in thickness from 0.050 to 0.100 m.Hand hacksaw U are 'I, in. wide. Blades for powe come m two standard lengths, 10 and 12 m. A saws vary in length from 12 to 24 in. andin width from 1to 2 in. Wider and thicker are deqirable for heavy-duty work. As a general rule, in hacksawing the blade sho at least twice as long as the maximum length of cut that is to be made. Bandsaw blades are available in straight, raker,wave, or comb~nation sets. In to reduce the noise from high-speed bandsawing, it is becomlng increasingly to use blades that have more than one pitch,slze of teeth, and type of set. Bla is very rmportant in bandsawing because it determines the minimum radius tha cut.The mo5t common widths are from in. to '1, m ,although w~der blades ca tained. Because wider blades are stronger, select the wldest blade poss~ble. Howeve tmg small radii requires a narrower and weaker blade. Bandsaw blades come In spacings from 2 to 32 teeth per mch. Circular saws for cuttlngmetal are often called cold saws to distinguish th friction-type disk saws Friction saws do not make chips but rather heat the me melt~ng temperature at the point of metal removal. Cold saws cut rapldly and chips like a milling cutter while producing surfaces that are comparable in smo and accuracy with surfaces made by shttmg saws in a milling machine or by a cu In a lathe. Dzsk or circular saws necessar~ly d i i r somewhat from straight-blade for with teeth cut sues up to about 18 in. in diameter have an integral-tooth des~gn ly into the disk (F~gure 27-14). Larger saws use either segmented or inserted tee teeth are made of h~gh-speedsteel or tungsten carb~de. The rema~nder of the made of ordmary, less expensive, and tougher steel. Segmental blades are co segments mounted around the periphery of the d~sk, usually fitted with a t groove and fastened by means of screws or rivets Each segment contams sever If a single tooth is broken, only one segment needs be replaced to restore the sa operating condihon. As shown in Figure 27-14, circular saw teeth are usually beve1ed.A common form has every other tooth beveled on both sides; that is, the first tooth is beveled left side, the second tooth on both sides, the thud tooth on the right side, the tooth on the left side, and so forth Another method is to bevel the opposlte s successive teeth. Beveling is done to ~ r o d u c e a smoother cut. Precis~on circul made from carbide, which are becoming available, are very thin (0.03 m.) and hav cutting-off accuracy, around i0.00008 in., with negligble burrs.

SECTION 27.6 lntroduct~on to Sawing

Clrcular

Workpiece

saw blade
FIGURE 27-14

Circular sawlng a structural shape, using (left to nghi) an insert tooth, a segmental tooth, and an integral-tooth circular saw blade.

TYPES OF SAWING MACHINES Metal-sawmg machines may be classified as follows~

1. Reciprocating saw a. Manual hacksaw b. Power hacksaw (Figure 27-15) c. Abraslve disc 2. Bandsaw a. Vertical cutoff (Figure 27-16) b. Hor~zontal cutoff (Figure 27-17) c. Combination cutoff and contour (F~gure 27-18) d. Friction 3. Circular saw (Figure 27.14) a. Cold saw b. Steel frlction disk

rr2 ,-Tough

allav steel back

FIGURE 27-15 Power hacksaw blade (above) and hacksaw w~th automatic bar feeding (right) cutting two pieces of round stock.

748

CHA~E 27 K Other Machin~ng Processes

Blade changing
Easy blade loading from the top on all
modelsand qulck removal of guards

saw FIGURE 27-16 Front view and rear vlew of a horizontal bandsawing mach~ne Inset shows blade-changingoperation.

FIGURE 27-17 Vertical

bandsaw cuthng a plece of pipe, 45". show~ng head t~lted

URE 27-18 Contour Idsawing on vertical ~dsaw~ng rnach~ne, wn in inset.

POWER HACKSAWS As the n a ~ n e imphes. power hacksaws are machines that mcchanlcally reciprocate 27-15). These machines conslst of a bed, a workholding a large hacksaw blade (F~gure frame, a power mechan~sm for re~zprocating the saw flame. and some type of feedlng mechanism. Because of the inherent ineff~c~ency oicutt~ng In only one stroke d~rect~on, they have often been replaced by more eIfic~ent, horizontal bandsawing machines. BANDSAWING MACHINES The earliest metalcutting bandsaw~ng machines were dlrect adaptat~ons from woodcutting bandsaws Modern machlnes of t h ~ type s are much more sophisticated and versatlle and have been developed speafically for metal cuttlng To a large degree they were made possible by the development of vastly better and more flex~ble bandsaw blades and slmple flash-weld~ng equipment, which can weld the two ends of a strlp of bandsaw blade together to form a hand of any des~red 1cngth.Three basic types of bandsawlng mach~nes are in common use. Horlzontul metalcutting bandsancng muchlnr~ were developed to combine theflex~btlity of reclprocatlng power hacksaws and the continuous cutting actlon of vertlcal by a hydraulic bandsaws.These heavy-duty automatic bandsaws feed the saw vert~cally mechanism and have automatic stock feed that can he set to feed the stock laterally any desired dlstance after a cut IS completed and automat~callyclamp it for the next cut. slmulSuch machlnes can be arranged to hold, clamp, and cut several bars of mater~al taneously. Computer numerical control (CNC) bandsaws are ava~lable with automatic storage and retr~eval systems for thb bar stock. Smaller and less expensive types have swing-frame construction, with the bandsaw head mounted ~na pivot on the rear of the machlne. Feed is accomplished by gravity through r ~ t a t l o n of the head about the pivot point. Because of the11 continuous cuttlng actton, hor~zontal bandsawlng machines are very efficient (Figure 27-16).

Upright, cutoff, bandsawlng machines (Figure 27-17) are designed primarily cutoff work on single stationary workp~eces that can be held on a table. On ma chines the blade mechanism can be tilted to about 45", as shown, to permit cut angle They usually have automatic power feed of the blade Into the work, a stops, and provision for supplying coolant. Combination cutoff and contour bandsawirag machines (Figure 27-18) can be not only for cutoff work but also for eontour sawing.They are widely used for cu irregular shapes in connection with making dies and the production of small num of parts and are often equipped with rotary tables Addrtiorial features on the8 chines include a table that pivots so that it can be tllted to any angle up to 4S0.U these machines have a small flash butt welder on the vertrcal column, so that a stra length of bandsaw blade can be welded quickly into a contmuous band. A small ing wheel is located beneath the welder so that the flash can be ground from th to provide a smooth joint that will pass through the saw guides. This welding an ing umt makes it possible to cut internal openings in a part by first dr~lling a serting one end of the saw blade through the hole, and then butt welding the t together. When the cut is finished, the band is agarn cut apart and removed from opening. The cutbng speed of the saw blade can be varied continuously over a range to provide correct operating condltrons for any materia1.A method of power ing the work is provided, sometimes gravity-actuated. Contour-sawmg machines are made in a wide range of sizes, the principal si mension being the throat depth. S~zes from 12 to 72 In. are avallable.The speeds able on most machines range from about 50 to 2000 ftlmm. Modern horrzoutal band are accurate to k0.002 in. per vertical inch of cut but have feeding accuracy of t0.005 in., subject to the size of the stock and the feed rate. Repeatabdity from feed to the next may he 2 0 010 to t0.020 in. CNC-controlled sawlng centers with microprocessor controls have opened up automation aspects for sawing. Such control systems can improve accuracy to w c0.005 in. over entire cuts by controlling saw speed, blade feed pressure, and feed Speclal bandsawing machines are available with very hlgh speed ranges, up 14,000 ftimin. These are known as friction bandsawing machtnes. Material is not cut chip formation. Instead, the friction between the rapidly mowug saw blade and the w IS sufficient to raise the temperature of the mater~al at the end of the kerf t below the melting point, where its strength IS very low. The saw blade then p molten, or weakened, material out of the kerf. Consequently, the blades do not n be sharp; they frequently have no teeth-only occasional notches in the blade to removing the metal. Almost any material, including ceramics, can be cut by frlction sawing. Beca only a small portion of the blade is in contact with the work for an instant and th cooled by its passage through the air, it remains cool. Usually, the major portion o work, away from contact with the saw blade, also remains quite cool.The metal adja to the kerf is heat affected, recast, and sometimes harder than the bulk metal. It IS a very rapid method for trimrmng the flash from sheet metal parts, castings, and for

CUTTING FLUIDS
Cutting fluids should be used for all bandsawing, with the exception that cast ironi ways cut dry. Commercially available oils or hght cuttlng oils will give good results in ting kflous materials. Beeswar:or paraffinare common lubricants for cutting alum and aluminum alloys.

FEEDS AND SPEEDS

Because of the many different types of feed involved in bandsawing, it 1s not prac to provide tabular feed or pressure data. Under general conditions, however, an pressure, without forcing the work, gives best results. A nicely curled chip usually cates an ideal feed pressure. Burned or discolored ch~ps indicate excesswe press which can cause tooth breakage and premature wear. Most handsaws provide recommended cutting speed infarmation right on the chine, depending upon the material being sawed. In general, HSS blades are run at

to 300 ftimin when cutting 1-in.-thick, low- and medium-carbon steels. For high-carbon steels, alloy steels, and tool and die steels, the range is from 150 to 225 ftlmin, and most stainless steels are cut at 100 to 125 ftlmin.

CIRCULAR-BLADE SAWING MACHINES EvIachines employing rotating circular or cold saw blades are used exclusively for cutoff work. These range from small, simple types, in which the saw is fed manually, to very large saws having power feed and built-in coolant systems, commonly used for cutting off hot-rolled shapes as they come from a rolling mill. In some cases friction saws are used for this purpose, havingdisks up t o 6 ft in diameter and operating at surface speeds up to 25,000 ftlmin. Steel sections up to 24 in. can be cut in less than 1 minute by this technique. Although technically not a sawing operation, cutoff work up to about 6 in. is often done utilizing thin abrusive disks.The equipment used is the same as for sawing. It has the advantage thatvery hard materials that would be very difficult to saw can be cut readily. A thin rubber- or resinoid-bonded abrasive wheel is used. Usually a somewhat smoother surface is produced.

Basically, the metal-removing action in Wing is the same as in sawlng, in that chips are removed by cutting teeth that are arranged in succession along the same plane on the surface of a tool, called a file. There are two differences. (1) the chps are very small, and therefore the cuttlng action is slow and easily controlled, and (2) the cutting teeth are much wlder. Consequently, frne and accurate work can be done. F~les are classified according to the following: 1. The type, or cut, of the teeth 2. The degree o E coarseness of the teeth 3. Construction a. Single solid unlts for hand use or lo dle-filing machine? b. Band segments, for use in band-filrpg machines c. Disks, for use in drsk-fihng machines Four types of cuts are available. Stngle-cut j d e ~ have rows of parallel teeth that extend across the entire w ~ d t h of the frle at the angle of from 65" to 85". Double-cut files have two series of parallel teeth that extend across the width of the file. One serles m cut at an angle of 40" to 45" The other serles is coarser and is cut at an opposite angle that varies from about 10" to 80" A vixen-cut file has a series of parallel curved teeth, each extending across the fileface. On a rasp-cut file,each tooth is short and is rarsed out of the snrface by means of a punch. These four types of cuts are shown in Figure 27-19.

FIGURE 27-19 Four types of teeth (cuts) used in files. Left to r~ght:Single, double, rasp, and (Courtesyof curved (v~xen). N~cholsonFile Company.)

The coarseness of files is designated by the following terms, arranged 1n or& increasing coarseness: dead smooth, smooth, second cut, bastard, coarse, and r There is also a series of finer Swiss pattern files, designated by numbers from 00 Flies are available in a number of cross-sectional shapes:fZat, round, square angular, and half-round. Flat files can be obtained with no teeth on one or both na edges, known as safe edges. Safe edges prevent material from being removed from face that is normal to the one being filed. Most files for hand filing are from 10 to In length and have a pointed tang at one end on which a wood or metal handle c fltted tor easy grasping.

i l

FILING MACHINES
An experienced operator can do very accurate work by hand filing, but it can be ficult task. Therefore, three types of fihng machines have been developed that p quite accurate results to be obtained rapidly and with much less effort Die-filin chines hoId and reciprocate a file that extends upward through the worktable.% rides against a roller guide at its upper end, and cutting occurs on the downward therefore, the cutting force tends to hold the work agatnst the table.The table can ed to any desired angle. Such machines operate at Lrom 300 to 500 strokes per and the resulting surface tends to be at a uniform angle with respect to the table. accurate work can be done. Because of the reciprocatmg action, approximately 5 the operating time is nonproductive. Band-filing machine3 provide continuous cutting action. hlosl band fdtng on contour bandsawing machines by meansof a special band file that is substitu the usual bandsaw blade. The principle of a band file is shown in Figure 27-2 stra~ght file segments. about 3 in. long, are riveted to a flexible steel band n leading ends. One end of the steel band contains a slot that can be hooked over the other end to form a continuous band. As the band passes over the drive a wheels of the machine, it flexes so that the ends of adlacent file segments mo When the band becomes straight, the ends of adjacent seglnents move together terlock to form a continuous straight file. Where the tile passes through the wor it is gnlded and supported by a grooved gulde, whlch provldes the necessary sup

Band fle segment5 (d) are jo'nea.' together to forrn d conr,nbobc h r r d (I?) nhtch r ~ n r barid-liling niacnrne (c). (Co,rrcsy of OoALl Co.
FIGURE 27-20

(a)

Review Question5

753

Disk-type filing h~ne and some of the ilable tvwes of d~sk files. ~ourtesy Pi/ersey Manufacturing
URE 27-21

resist the pressure of the work against the file. Band files are available in most of the standard cuts and in several widths and shapes. Operating speeds range from about 50 to 250 filmin. Although band filing is considerably more rapid than can be done on a die-filing machine, it usually is not quite as accurate. Frequently, band filing may be followed by some finish filing on a die-filing machine. Some disk-filing machines have files in the form of disks (Figure 27-21).These are even simpler than die-filing machines and provide continuous cutting action. However, it is difficult to obtain accurate results by their use.

I I Key Words
clrcular saw cold saw firing hacksaw kerf planning pull broach push broach reciprocating saw rise per tooth samng shaping surface broach tooth set

Review Questions
What is unlque about the broaching process compared to the other basic machinlag processes? Can a thlck saw blade be used as a broach? Why or why not? Broaching machines are simpler in a bale design than most other machme tools. Why is th~s? . Why is broach~ng partiwlarly well suited for mass produc,

. In des~gning a broach, what would be thefirst thing you have

Why is it necessary to relate the design of a broach to the specific workplece that is to he machmed? What two methods can he utlltzed to reduce the force and power requlrements for a particular broaching out? For a given job. how would a broach havlng rotor-tooth deslgn compare in length with one havlng regular, full-wtdth

9. Why are the p~tch and radlus of the gullet between teeth on a broach of importance? 10 Why are broachrngspeeds usually relatively low, as compared w~th other machining operatlous? 11. What are the advantages oI shell-type broach construction? 12. Why are most broaches made from alloy or high-speedsteel rather than from tungsten carbide? 13 What are the advantages of TIN-coated broachlng tools? 14. For mass-pmduction operations, whrch process is preferred, pull-up broachlng or pull-down broachmg? 15. What is the difference between theroughlng teeth and the finishlng teeth in a typical pull broach? 16. Thesides of a square, bhnd hole must be machlned all the way to the bottom (who desrgned thls pdrtQ).The hole is first dnlled to full depth, then the bottom is milled flat Is it possible to machine the hole square by broaching7 Why or why not?

17. The interior,flat surfaces of socket wrenches, which havc one "closed" end, often are finished to size by broaching. By examining one of these, determine what design modification was incorporated to make broaching possible. 18. Why is sawing one of the most efficient of the chip-forming processes? 19. Explain why toolh spacing (pitch) is important in sawiilg, 20. What is ihe tooth gullet used for on a saw blade? 21. Explain what is meant by the "set" of the teeth on a saw blade. 22. How is tooth set related to saw kerf? 23. Why can a handsaw blade not be hardened throughout the entire width of the hand? 24. What are ihe advantages of using circular saws? 25. Why have bandsawing machines largely replaced reciprocatingsaws? 26. Explain how the hole in Figure 27-18 is made on a contour hand~aw~n machtne g

T, for a horvontal b 27 How would you calculate or est~mate saw cuttmg a 3 1 1 1 . round of 1040 steel7 28 What is the disadvantage ot using gravity to feed a sawln ting round bar stock? 29 To what extent is fihng dtfferent from sawlne? 30 What 1s a safe edgc on a file7 31 Why is an end filing - machine morc efficient than a diemachine? 32. How does a rasp-cut lile dtffer from other types of files? 33. How does the process of shaping differ from plamngq 34. How n teed per stroke mshaprng related to feed per too milling' 35 What are some ways to tmprovc the eliicrency of a p Do any of these apply to the shaper?

II Problems
1. A surface 12 in. long is to be machined with a flat,solid broach that has a rise per tooth of 0.0047 in. What is the minimum noss-sectional area that must be provided in the chip gullet between adjacent teelh? 2. The pitch of the teeth on a simple surface broach can be determined by equation 27-1. If a broach is to remove 0.25 in. of material from a gray iron casting that is3 in. wide and 17.75 in. long, and if each tooth has a rise per tooth of ,004 in., what will be the length of the roughing section of the broach? 3. Estimate the (approximate maximum) horsepower needed to accomplish the operalion described in Problem2 at a cutting First find the H P used per tooth and speed of 10 mlmin. (Hint: determine the maximum number of teethengaged at any time. What are those units?) 4. Estimate the approximate forcc acting in the forward direction during cutting for the conditions stated in Problems 2 and 3. 5. In cutting a 6-in.-longslot in a piece of AISI 1020 cold-rolled steel that is 1 in. thick, thc material is fed to a bandsaw blade with teeth having a pitch of 1.27 mm (20 pitch) a t the rate of 0.0001 in. per tooth. Estimate the cutting time for the cut. 6 ; The strength of a pull broach is determined by its minimum cross section, which usually occurs either at the root of the first tooth or at the pull end. Suppose the minimum root diameter is D,, the pull end diameter is D, and the width of the pull slot is W. Write an equation for the allowable pull, in psi, using 200,000 as the yield strength for the broach material. 7. Suppose you want to shape a block o l metal 7 in. wide and 4 in. long (L = 4 in.) using a shaper as set up in Figure 27.1. You have determined for this metal that the cutting speed should be 25 sfpm, the depth of cut needed here for roughing is 0.25 in., and the feed will be 0.1 in. per stroke. Determine the approximate crank rpm and then estimate the cutting and the MRR. 8. Could you have saved any time in Problem 7 by cutti block in the 7-in. direction? Redo with L = 7 and W 9. Derive lhe equation for shaping cuttingspeed,equati 10. How many strokes per minute would be required too cutting speed of 36.6 m (120 ft) per minute on a typi chanical drive shaper if a 254-mm (10-in.) stroke is us 11. How much time would he required to shape a 254 mm (I in.) wide and 203 mm (8 in.) long on drive shaper, using a cutting speed of 45.7 m ( minute, a Ieed of 0.51 mm (0.020 in.) per stroke, s overrun of 12.7 mm 'Iz in.) at each end of the cut? 12. What is the metal removal rate in Problem 11 if the de cut is 6.35 mm ('id in.)? 13. Suppose you decide to mill the flat surface describedin. lems 11 and 12. The work will h e done on a vertical machine using a 1.25-in.-diameter end mill (four tee cutting at 150 sfpm with a feed per tooth of 0.005 in cutting at d = 0.25 in. Compare the milling time an that of shaping. 14. A planer has a 10-hp motor, and 75% of the mot available at the cutting tool.The specific power lor iron metal is 0.03 Wlmm', or 0.67 hplin.31 min.What 1 s irnunl depth of cut that can be takenin shaping asu material if the surface is 305 X 305 mm (12 X 121, 0.25 in. per stroke, and the cutting speed is 54.9 & I (18 minute? Calculate the T, for planing the block olcaBt iron in 14 and then estimate T,"for milling the same surface. have to determine which milling process to use an speeds and feeds for an HSS cutter.

Cost Estimating--Planing vs. Milling

about 6,000 hr/yr.The charge for the use of the machine IS $2.50/hr and labor and overhead in addrtion are charged at $20 OO/hr.The other machlne 1 s a large vertical sp~ndle CNC rnlil~ng machine that costs $165,000 new, deprec~ated over a 20-year tlrne perlod also. The charge for the use

$31,00/piece for themlll. Purchasing has not ~ssued an order (quatltlty not yet decided) so KC needs to determine the BEQ SQ she knows which rnachlne to use when the order IS placed.

2 . 1 INTRODUCTION 28.2 ABRASIVES Abrasive Grain Size and Geometry WHEEL STRUCTURE 28.3 GRINDING AND GRADE G Ratio Bonding Materials for Grinding Wheels Abrasive Machining Versus Conventional Grinding Versus Low-Stress Grinding

28.4 GRINDING WHEEL IDENTIFICATION Grinding Wheel Geometry Balancing Grinding Wheels Safety in Grinding Use of Cutting Fluids in Grinding 28.5 GRINDLNG MACHINES Cylindrical Grindi'ng Centerless Grinding Surface Grinding Machines Disk-Grinding Machines Tool and Cutter Grinders

Mounted Wheels and Coaled Abrasives 28.6 HONING 28.7 SUPFRFINISHWO Lapping 28.8 FEE ABRASIVES Ultrasonic Machining Waterjet Cutting and Waterjet Machi Abrasive Jet Ma Design Considerations case Study: OVERHEA INSTALLATION

111 28.1 INTRODUCTION

Abrcrsive machining is a material removal process that involves the inte abrasive grits with Lhe workpiece at high eutting speeds and shallow penetrat' The chips that are formed resemble those formed by other machining Unqueslionably, abrasive machining is the oldest of the basic machining Muscums abound with examples of utensils, tools, and wcapons that ancie produced by rubbing hard stoncs against softer materials to abrade away portions, leaving desired shapes. For centuries, only natural abrasives were ava grinding, while other more modern basic machining processes were develope superior cutting materials. However, the developmenl of manufactured ahrasi a better fundamental understanding of the abrasive machining prooess have res placing ahrasive machining and its variations among the most important of all t machining processes. The results that can be obtained by abrasivemachining range from the fin smoothest surfaces produced by any machining process, in which very little removed, to rough, coarse surfaces that accompany high material remova ahrasive particles may be (1)free; (2) mounted in resin on a bell (called cuutcd or, most commonly ( 3 ) close packed into wheels or stones, with abrasive together by bonding material (called bondedpruduct or a grinding wheel). F shows a surface grinding process using a grinding wheel. The depth of cut rd i mined by the infeed and is usually very small, 0.002 to 0.005 in., so the arc of (and the chips) is small. The table reciprocates back and forth beneath the r wheel. The work feeds into the wheel in the cross-feed direction. After the wo of the wheel, the wheel is lowered and another pass is made, again remov' of thousandths of inches of metal.Tbe metal removal process is basically th abrasive machining processes bur with important differences due to spacin grains (grains in contact with work) and the rigidity and degree of fixation of Table 28-1 summarizes the primary abrasive processes. The term abrasive applied t o one particular form of the grinding process is unfortunate, because a process are machining with abrasives. Compared to machining, abrasive machining processes have three uniq acteristics. First, each cutting edge is very small, and many of these edges can cu taneously. When suitable machine tools are employed, very fine cuts are pos ' fine surfaces and close dimensional control can be obtained. Second, because e

SIXXIOX 28.2 Abrasives

757

(Down grinding) c Table motion

+ (Up grinding)
Oblique view

Side view

FIGURE 28-1 Schematic of rurface grinding, showing infeed and cross feed motions along with cutting speeds V5,

and workpiece velocity V#. hard abrasive grits, including diamonds, are employed as cutting tool materials, very hard materials, such as hardened steel, glass, carbides, and ceramics, can readily be machined. As aresult, the abrasive machining processes are not only important as manufacturing processes, they are indeed essential. Many of our modern products, such as modern machine tools, automobiles, space vehicles, and aircraft, wuld not be manufactured without these processes.Thkd,in grindin&you have no control over the actual tool geometry (rake angles,cutting edgeradins) or all the cutting parameters (depth of cut). As a result of these parameters and variables, grinding is a complex process. To get a handle on the wmplexity,Table 28-2 presents the primary grinding parameters, grouped by their independence or dependence. Independent variables are those that are controllable (by the machinc operator) while the dependent variables are the resultant effects of those inputs. Not listed in the table is workpiece hardness, which has a significant e k c t on allthe resulting effects Workpiece hardness will be an input factor but it is not usually controllable.

28.2 ABRASIVES
An abrasive is a hard material that can cut or abrade other substances Natural abrasives have existed from the earliest times. For examplc, sandstone was used by ancient peoples to sharpen tools and weapons. Early grinding wheels were cut from slabs of sandstone, but because they were not uniform in structure throughout, they wore unevenly and did not produce consistenl results. Emery, a mixture of alumina (A120,) and magnetite (Fe,04), is another natural abrasive still in use today and is used on coated paper and cloth (emery paper). Corundum (natural A1,0,) and diarnotids are other naturally occurring abrasive materials. Today, the only natural abrasives that have commercial

asive machiniirg

Bonded Bonded bcltcd Bonded

sive waterjct

Free in jet Frce in liquid

High MRR, to obtain dzsired shapes and approximate sizes High MRR, rough rapid lechnique to clean up and dchurr castings, irlrgings "Stones" containing fine sbrasivcs;primarily a hole-finishing proccrs Fine particles emheddcd in snft mela1 or cloth: primarily a surface-finishing process Water jets with velacilies up to 30UO sfpm cany abrasive particles (silica and 8arntt). Vibraling tool impacts abrasives at high velocity

CHAPIER 28 Abrasive Machin~ng Processes

Independent ParameterslControllable Dependent VariableslResultlngEffects

Gnnd~ng wbeel sclcct~on Abrasivc type

Frdm 61%
H a d n s s grade Openness of s t r u m r e Bond~ng medm Dressing of wheel Type of dressrng fool Feed and depth of cut Sharpness of dressing tool Machrne settlngs Wheel speed Infeed rate (depth of cut) Crass-feed rate Workplece speed R ~ g ~ d iof t y setup Type dnd quality of machrne Grmdmngfluid

Forces per u n ~wtdth t of wheel Normal Tangenhat Surface finish Mauenal removal rate (MRR) raho) Wheel wear (G, ar $rind~ng Thermal effects Wheel surface changes Chemtcal effects Horsepower

Qrx

Clwnhn&\ Method of appheation

importance are quartz, sand, garnets, and diamonds. For example, qu primarily in coated abrasives and in air blasting, hut artificial abrasives are inroads in these applications.The development of artificial abrasives having form properties has permitted abrasive processes lo become precision man processes. Hurdness, the ability to resist penetration, is the key property for an Table 28-3 lists the primary abrasives and their approximate Knoop hardness The particles must be able to decompose at elevated temperatures. Two 0th ties are significant in abrasive grits-attrition and friability.Atbition refers to sive wear action of the grits resulting in dulled edges,grit flattening, and whee Frirrhility refers to the fracture of the grits and is the opposite of toughness. In it is important that grits be able to fracture to expose uew,sharp edges. Artificial abrasives date from 1841,when E. G. Acheson, while attempti duce precious gems, discovered how to make silicon carbide (Sic). Silicon made by charging an electric furnace with silica sand, petroleum coke,salt, and By passing large amounts of current through the charge, a temperature is maintained for several hours, and a solid mass of silicon carbide crysta the furnace has cooled, the mass of crystals is removed, crushed, and graded (so

Abraclvc Mater~al
Quartz Alummum ox~de

Year of D~scovery
7

Hardne~s (Knoop)
320

1893

160WZ100

170&2400
1400-2000

Sand blasting Softer and tougher than six used on steel, iron, brass, s stainless and cast iron

Borazon [cublc boron nttnde stnlnless (CBN)] Dlarncmd (synlhellc)

1200-1400

700-BW)

FIGURE 28-2 Loose abras~ve grains at high magnification, showing the~r irregular, sharp cutting edges. (Courtesy of Norton Company)

various desired sizes. As can be seen in Figure 28-2, the resulting grit$ or grains, are irregular in shape, with cutting edges having every possible rake angle. Silicon carbide crystals are very hard (Knoop 2480), friable, and rather brittle.This limits their use. Silicon carbide is sold under the trade names Carborundum and Crystolon. Alunzinum oxide (A120,) is the most widely used artificial abrasive. Also produced in an arc furnace from bauxite, iron filings, and small amounts of coke, it contains aluminum hydroxide, ferric oxide, silica, and other impurities. The mass of aluminum oxide that is formed is crushed, and the particles are graded to size. Common trade names for aluminum oxide abrasives are Alundum and Aloxite. Although aluminum oxide is softer (Knoop 2100) than silicon carbide, it is considerably tougher. Consequently,it is a better general-purpose abrasive. Diamonds are the hardest of aU materials. Those that are used for abrasives are either natural, off-color stones (called garnets) that are not suitable for gems, or small, synthetic stones that are produced specifically for abrasive purposes. Manufactured stones appear to b e somewhat more friable and thus tend to cut faster and cooler.They do not perform as satisfactorily in metal-bonded wheels. Diamond abrasive wheels are used extensively for sharpening carbide and ceramic cutting tools. Diamonds also are used for truing and dressing other types of abrasive wheels. Diamonds are usually used only when cheaper abrasives will not produce the desired results. Garnets are used primarily in the form of very finely crushed and graded powders for f i e polishing. Cubic boron nitride (CBN) is not found in nature. It is produced by a combination of intensive heat and pressure in the presence of a catalyst. CBN is extremely hard, registering at 4700 on the Knoop scale. It is the second-hardest substance created by nature or manufactured and is often referred to, along with diamonds, as a superabrasive. Hardness, however, is not everything. CBN far surpasses diamond in the important characteristic of thermal resistance. At temperatures of 650C, at which diamond may begin to revert to plain carbon dioxide, CBN continues to maintain its hardness and chemical integrity. When the temperature of 1400C is reached, CBN changes from its cubic form to a hexagonal form and loses hardness. CBN can be used successfully in grinding iron, steel, alloys of iron. Ni-based alloys, and other materials. CBN works very effectively (long wheel life, high G ratio, good surface quality, no burn o r chatter, low scrap rate, and overall increase in partsishift) on hardened materials (R, 50 or higher). It can also be used for soft steel in selected situations. CBN does well at conventional grinding speeds (6000 to 12,000Mmin), resulting in lower total grinding in conventional equipment. CBN can also perform well at high grinding speeds (12,000 ftlmin and higher) and will enhance the benefits from future machine tools. CBN can solve difficult-to-grind jobs, but it also generates cost benefits in many production grinding operations despite its higher cost. CBN is manufactured by the General Electric Company under the trade name of Borazon.

760

CHAFER 28 Abrasive Machining Processes

FIGURE 28-3 Typical screens for sifting abrasives into sizes. The larger the screen number (of opening per linear inch), the smaller the grain size. (Courtesy of Corborundum Cornpony.)

ABRASIVE GRAIN SIZE AND GEOMETRY

mechanical sievingmachinesThe number of openings per linear inch in a sieve (or me through which most of the particles of a particular size can pass determines the grain (Figure 28-31, A no.24 grit would pass through a standard screen having24 openings per inch would not pass through one having 30 openings per inch. These numbers have si

from 120 grit for CBN to 400 grit for diamond. Sizes from 240 to 600 are.designat flour sizes. These are used primarily for lapping, or in fine-honing stones for fine ishing tasks. The grain size is closely related to the surface finish and metal removal rat grinding wheels and belts, coarse grains cut faster while fine grains provide better

to the screen number by g

-- 0.7iS.

I
12 24

36

48

60
w

80

FIGURE 28-4

M R R and surface

fin~sh versus grit size.

G r i t sire decreasing

SECTTON 28.2

Abrasives

761

FIGURE 28-5 The rake angle of abrasive particles can be positive, zero, or negative.

b t
nee

as).
rsa; 220
2 to

the grain diameter decreases, the number of active grains per unit area chips are sma1l.A~ increases and the cuts become finer because grain size is the controlling factor for surface finish (roughness). Of course, the MRR also decreases. The grain shape is also important, because it deternines the tool geometry-that is,the back rake angle and the clearance angle at the cutting edge of the grit (Figure 28-5). In the figure, y is the clearance angle, 8 is the wedge angle, and a is the rake angle. The cavities between the grits provide space for the chips, as shown inFigure 28-6.The volume of the cavities must be greater than the volume of the chips generated during the cut. Obviously, there is no specific rake angle but rather a distribution of angles.Thus a grinding wheel can present to the surface rake angles in the range of +45' to -60' or greater. Grits with large negative rake angles or rounded cutting edges do not form chips but will rub or plow a groove in the surface (Figure 28-7).Thus abrasive machining is a Open cav~ty B~nderforrn~ng (vo~d) bond post
i

Abraslve part~cle
i

l as

all-

--

--

caviky
full O f

~a)r~y

~a&ty

filled

chips
Cutting Grin

voids between the gram must be large enough to hold all the ch~ps during the cut.

Side view of ara~n

chip Workpiece

~nritious-1 wear of grit

MLoading

FIGURE 28-7 The grits interact with the surface in three ways: cutting, plowing, and rubbing.

762

CHAWER 28

Abrasive

FIGURE 28-8 SEM micrograph of a ground steel surface showing a plowed track (T) in the middle and a machined track (M)above. The grit fractured, leaving a portion of the grit in the surface (X), a prow formation (P), and a groove (C) where the fractured portion was pushed farther across the surface. The area marked (0)is an oil deposit.

FIGURE 28-9 SEM micrograph of stainless steel chips from a grinding process. The tops (T) of the chips have the typical shearfront-lamella structure while the bottoms ( 6 ) are smooth where they slide over the grit 4800x.

SsCrroN 28.3 Grinding Wheel Structure and Grade

763

used to keep the grains sharp. As the grains become dull, cutting forces increase,and there is an increased tendency for the grains to fracture or break free from the hondmg material.

HEEL STRUCTURE AND

GRADE

Grindmng,wherein the abrasives are bonded togbther into a wheel, is the most common abrasive machinmg process The performance of grindmg wheels is greatly affected by the bonding material and the spahal arrangement of the particles' grits. The spacing of the abrasive particles with respect to each other is called slruciui-e. Close-packed grains have dense structure; open structure means widely spaced grains. Open-structure wheels have larger chip cavitles but fewer cuthng edges per unit area (Figure 28-l0a). The fracturing of the grits is controlled by the bond strength, which 1s k n ~ w n as the gmde. Thus, grade is a measure. of how strongly the grams me held in the wheel It is really dependent on two factors: thestrength of the bonding materials and the amount of the bonding agent connecting the grains. The latter factor is illustrated in Figure 28-lob. Abrasive wheel* are really porous. The grains are held together with "posts" of bonding material. Tf these posts are large in cross section, the fqrcz required to break a grain free from the wheel is greater than when the posts are small, If a high d~slodgiugfmceis required, the bondis s a d to be hard,Ifonly asnlall torce is required, the bond is said to be so$. Wheels are commonly referred to as hard or soft, referring to the net strength of the bond, resulting from both the strength of the bonding mat& ial and its disposition between the grains.

G RATIO
The loss o f grams from Lhe wheel means that the wheel 1s changing size.The grtnding ratio, or G raho, is defined as the cubic inches oi stock removed bvided by t h e cubic inches of wheel lost. In conventional grinding, the Gratiois in the range 20:1 to 80:l.The G ratio is a measure of grinding production and reflects the amount of work awheel can do during its usefull1fe.A~ the wheel losesmaterial,it must be reset or repositioned to maintam workpiwe size. A typical v~trifiedgrinding wheel will consist of 50 uol% abrasive particles, 10 vol% bond, and 40 vol% cavities; that is, the wheek have porosity.The manner in which the wheel performs is influenced by the following factors:

1 . The mean force required to dislodge a gram from the surface (the grade of the wheel) 2 . The cavity size and distribution of the poroslty (the structure)

Open spaclng

Med~urn spacing

Dense spacing

Weak "posts" Open spanng

Medlum srren$th "posts" Open spacing

Strong "posts" Open spaclng

Abras~ve Mach~n~ng Processes

3. The mean spacing of active grains in the wheel surface (grain size and structure) 4. The properhes of the grain (hardness, attrition, and friability) 5. The geometry of the cutting edges of the grains (rake angles and cutt~ng-edge ra compared to depth of cut) 6. The process parameters (speeds, feeds, cutting flu~ds) and type of grindlng (surf or cylindrical) It is easy to see why grinding is a complex process, difficult t o control.

BONDING MATERIALS FOR GRINDING WHEELS

Bonding material 1s a very importanl factor to be conqidered in selccting a grin wheel It determines the strength of the wheel, thus establishng the maximum opera speed It determines the elastic behav~or or deflection of the grits in the wheel du grinding.The wheel can be hard or rlgld, or it can be flexible. F~nally, the bond deter the force requ~red to dislodge an abrasive particle from the wheel and thus plays am role in the cutting action. Bond materials are formulated so that the ratio of bond matches the rate of wear of the abrasive grits. Bonding material$ in common use the following:

1. Vitrified bonds are composed of clays and other ceramic substances The abr
particles are mlxed w ~ t h the wet clays so that each grain is coated.Wheels are for from the mix, usually by pressmg, and then dned.They are then fired in a k~ln,w results in the bonding material's becommg hard and btrong, having properkes lar to glass V~tnfied wheels are porous, strong, r~gid, and unaffected by o h , wat temperature over the ranges usually encountered in metal cutting. The ope speed range In mast cases is 5500 to 6500 ftiInm, but some wheels now opera surface speeds up to 16,000 fdnun. 2. Restno~d, or phenolic resins, can be used. Because plastics can be compound have a wide range of properties, such wheels can be oblalned t o cover a va work conditions. They have, to a considerable extent, replaced shellac and wheels. Composite materials are being used in rubber-bonded or resinoidwheels that are to have some degree of flexibility or are to recelve considerable and s ~ d loading. e Various natural and bynthetlc fabr~cs and fibers, glass fiber nonferrous wlre mesh are used for this purpose. 3. Sil~cate wheels use silicalc of soda (waterglass) as the bond material.The whee formed and then baked at about 500'F for a day or more Because they are brittle and not so strong as v~tnfied wheels, the abranve grams are released readily. Consequently, they mach~ne at lower surface temperatures than wheels and are usetul in grlnding tools when heat must be kept to a mimmum. 4. Shellac-bonded wheels are made by mixing the abraslve grains wlth shellac in mixture, pressing or rolhng Into the desired shapes, dnd bak~ng for several about 300F.This type of bond is used pnmarlly for strong, thin wheels havi elasbcity. They tend to produce a high polish and thus have been used in such parts as camshafts and mill rolls. 5. Rubber bondmg is used to produce wheels that can operate at h ~ g h speeds have a considerable degree of flexibil~ty so as toreslst slde thrust. Rubber, other vulcanizmg agents are mixed with the abrasive grainsThe mlxture is out Into sheets of the desired thickness, and the wheels are cut from these vulcanized. Rubber-bonded wheels can be operated at speeds up to 1 They are commonly used for snagging work m foundries and tor lhm cutoff 6. Superahraslve wheels are either electroplated (single layer of superabrasl outside diameter of a steel blank) or a thrn segmented drum of v~trified rounds a steel core.Thestee1 core provides dimensional accuracy, and the segments provlde durabihty, homogeneity, and repeatability while incre l~fe The latter type of wheels can use resin, metal, or vitrlfted bonding. bond grade and structure (also called abraslve concentration) is cntlcal.

SECTION 28.3 Grtnd~ng Wheel Structure and Grade

765

For the electroplated wheels, nickel is used to attach a singlc layer of CBN (or diamond) t o the O D of an accurately ground or turned steel blank. For the vitrified wheel. superabrasives arc m x e d with bond~ng media and inolded (or preformed and sintcred) into segments or a ring.The rmg is mounted on a split stcel body. Poroslty is vaned (to alter structure) by varying preform pressure or by uslng "porc-forming" addlt~ves to the bond materlal that are vaporrzed during the sintenng cycle. The steel-cored segmented design can rotate a t 40,000 sfpm (200 mls) whereas a plain wtrified wheel may burst at 20,000 fpm.

ABRASIVE MACHINING VERSUS CONVENTIONAL GRINDING VERSUS LOW-STRESS GRINDING The condition wherein very rapid metal removal can be achieved by grinding is the one to which some have applied the term nbrasive mqchining.The metal removal rates are compared with, or exceed, those obtainable by milling or turning or broaching, and the size tolerances are comparable. It is obviously just a special type of grinding, using abrasive grains as cutting tools, as do all other types of abrasive machining. Abrasive grinding done in an aggressive way can produce sufficient localized plastic deformation and heat in the surface so as to develop tensile residual stresses, layers of overtempered martensile (in steels), and even microcracks, because this process is quite ribusive. See Figure 28-11 for a discussion of residual stresses produced by varius surface grinding processes. Conventional grinding can be replaced by procedures that develop lower surface stresses when service failures duc to fatigue or stress corrosion are possible.This is accomplished by employing softer grades of grinding whccls,reducing thc grinding speeds and infecd rates, using chemically active cutting fluids (e.g., highly sulfurized oil or KNO, in water), as outlincd in the table of grinding conditions in Figure28-1l.These proccdures may require the addition of a variahlc-speed drive to the grinding machine. Generally, only about 0.005 to 0.010 in. o l surface stock needs to be finish ground in this way, as the depth of the surface damage due lo conventional grinding or abusive grinding is 0.005 to 0.007 in. High-strength steels, high-temperature nickel, and cobalt-based alloys and titanium alloys are particularly sensitive to surface deformation and cracking problems from grinding. Other postprocessing processes, such as polishing. honing, and chemical milling plus peening, can be used to remove the delhrmed layers in critically stressed parts. lt is strongly recommended,however. that testing programs be used along with service experience on critical parts before these procedures are employed in production.
Grtndlng cond~t~ons Abuslve Conventional
AG
CG
A46KV 4,5006.500 ,001003 ,040-060 40100

Low-stress LSG
A46HV or A601V 2500-3000 ,0002005 ,040-,060 40-100

Wheel Wheel speed

ftirnln

A46MV 6,00018,000

0 .ur

-20 -40

Q
0

grinding ILSG)
0 ,002 ,004 ,006 .008 ,010

Down feed in./pass Cross feed in.ipass Table speed ftimin

Fluid

002004
.04& ,060 40100
Dry

-60

Sol oil
(1: 20)

Sulfurized oil

Depth below the surface{in.)

FIGURE 28-11 Typical residual stress distributions produced by surface grinding with different

grinding conditions for abusive, conventional, and low-stress grinding. Material is 4 3 4 0 steel. (From M. Field and W. P. Kosher, "Surfacelntegrity in Grinding," in New Developments in Grinding, Carnegie-Mellon University Press, Pittsburgh, 1972, p. 666.)

766

CMER 28

Abrasive Machining Processes

Rotary disk

Rotary cup

Grinding wheels lose geometry during use and need truing

fzd

dressing tool

Diamond roll

Diamond block

FIGURE 28-12 Truing methods for restoring grinding geometry include nibs, rolls, disks, cups, and blocks.

In the casting and forging industries, the term often used for abrasive mac snagging. Srragging is a type of rough ma risers, and rough spots from castings or chining.Theprimary objective is to remove substan much regard for accuracy, so this is a form of abrasiv or swinggrinders ordinarily are used. Portable ele for this purpose and for miscellaneous g Grinding wheels lose their geomet A single-point diamond tool can be used to true t to expose new grains and new cutting edges on ing can also be accomplished by grindi powered rotary device using conventional abr a trued wheel surface by these methods is poo As the wheelis used, there is a tendency become lodged in the cavities between the wear, flatten, and polish). Unless wheel will not cut as well and will t arrangement for stick dressing a grind on the grains to increase,ideally resulti the bond, thus providing a continuous ous action ordinarily will not occur for ligh grinding wheels do become somewhat self-d heated and turn a blujsh temper color (thi

Dressing stick pushed into the wheel

FIGURE 28-13

Schematic arrangement of stick dressing versus truing.

SECTION 28.4 Grlndlng Wheel ldent~flcat~on

767

Crushtruea roll

UOUS

uing -truil the een Centertype form grinding fully dressed condition A burned surface, the consequence ol an ox~de layer fo~mat~on, results in the scrappmg of several workpleces before parts of good quality are ground. Resin-bonded wheels can be trued by grinding with hard ceramics such as tungsten carbide.The procedure ior truing and dressing a CBN wheel in a surface grinder mght be as tollows: Use O OOM-in. downfeed perpass and cross feed slightly more than half the wheel thickness at moderate table speeds The wheel speed is the same as the gr~nding speed. The gnnding power will gradually increase, as the wheel is getting dull, while being trued. When the power exceeds normal powcr drawn during workpiece grindmg, stop the trumg operation. Dress the wheel Lace open usmg a J-grade stick, with abrasive one gnt size smallef than CBN Contlnue the truing. Repeat this cycle until the wheel is completely trued. Modern grinding machines are equipped so that the wheel can be dressed andlor trued continuously or intermittently whlle grindlng cont1nues.A common way to do this is by crush drenmg (F~gure 28-14). Crush dressmg consists of forcing a hard roll (tungsten carbide or high speed steel) having the same contour as the part to be ground against the grinding wheel while t t LS revolving-usually quite slowly. A water-based coolant is used to flood the dresslng zone at 5 to 10 gal1mm.Tne crushing action fractures and dislodges some of the abrasive grains, exposing fresh sharp edges, allowing free cuttlng for faster infeed rates. This procedure is usually employed to produce and maintain a special contour to the abrasive wheel.This is also called wheel profiling. Crush dressin is a very rapid method of dressing grinding wheels, and because it frac4. Lures abrasive grams, it results in tree cuttmg and somewhat coolergrinding.The resultmg surfaces may be slightly rougher than when dlamond dresslug is used.

Most grinding wheels are identified by a standard mdrk~ngsystem that has been estabhahed by the American National Standards Inst~tute.This system is illustrated and explained in Figure 28-15.The first and last symbols in the marklng are left to the discretion of the manufacturer.

GRINDING WHEEL GEOMETRY The shape and size of the wheel are critical selection factors. Obviously, the shapemust permit proper contact between the wheel and all of the su~face that must be ground Grinding wheel shapes have been standardized, and eight of the most commonly used types are shown in Figure 28-16.Types 1,2,and 5 are used prmanly for grind~ng external or Internal cylmdrical surfaces and for plain surface grmd~ng.Type 2 can be mounted for grinding on e~ther the periphery or the s ~ d of e the wheel.Type 4 ts used with tapered safety flanges so that if the wheel breaks during rough gnnding, such as snagging, these

Sequence
Pref~x

1
Abrasive
Type

2
Abraslve (Grain1 Size

3
Grade

4
Structure

5
Bond Type Manufacturer's Record

MANUFACTURER'S SYMBOL INDICATING EXACT KIND OF ABRASIVE (USE OPTIONAL]

A TFA

Regular Alummum 0nde--j Treated Alumnum Oxlde

Coarse 8 10

3A

B BF E 0 R RF S V

Resinoid Resinoid Rein Shellac Oxychloride Rubber Rubber Reinforced Silicate Vitrified

soft

GC Rc
DA

G v n SOlcon Carbids M I X ~ U ,s ~l ~ r o n cschlde

Medium

Hard

ABCDEFGHIJKLJMNOPQRSTUV WXYZ
Grade Scale

Standard bonded abrawve wheel-marching rystem /ANSI Standard 874 13 19771

Sequence
Prefix

1
Abraeve Type

2
Abras~ve /Gram1 Sue

3
Grade

4
Concentrat~on

5
Bond Type

6
Bond Mod~f~cation

7
Depth of Abras~ve

8
Manufa Rec

MANUFACTURER'S SYMBOL INDICATING EXACT KIND OF ABRASIVE. /USE OPTiONALi 1

1
Coarse to 8 30 Fine 280

Manufacturer's designation: may be number or symbol

B V M

Resin Vitrified Metal

Manufacturer's Notation of Special Bond Type or Modification

Working depth of abrasive section in inohes or millimeters. inches illustrated. Manufa, identifi

s ~ m (use OF

ABCDEFGHIJKLMNJOPQRSTUVWXYZ Soft ................................................................................ Hard

Whssl-marehmgsyrtarnfor d~srnond sndcublc baronnrtrlde wheelstANSI Sfandsrd874 13 19771

FIGURE 28-15

S t a n d a r d m a r l t i n y s y s t e m s for yr~ndiny wheels (ANSI s t a n d a r d 374. 13-1977).

SECTION 28.4

Grlnding Wheel ldentificat~on

769

1.

Straight

5. Cylinder

2.

Recessed one side

6. Straight cup

.,

P -

L H ( 'G~
7. Flaring cup

3. Recessed two sides

IGURE 28-16 Standard

rlnding wheel shapes xnmonly used. (Courtesy of .arborundum Compony.)

4.

Tapered

8.

Dish

flanges will prevent the pieces of the wheel from flying and causing damage.Type 6, the straight cup, is used primarily for surface grinding but can also be used for certain types of offhand grinding. The flaring-cup type of wheel is used for tool grinding. Dish-type wheels are used for grinding tools and saws. 'Qpe 1, the straight grinding wheels, can be obtained with a variety of standard faces. Some of these are shown in Figure 28-17. The size of the wheel to be used is determined primarily by the spindle rpm values available on the grinding machine and the proper cutting speed for the wheel, as dictated by the type of bond. For most grinding operations the cutting speed is about 2500 to 6500 ftimin. Different types and grades of bond often justify considerable deviation from these speeds. For certain types of work using special wheels and machines, as in thread grinding and "abrasive machining," much higher speeds are used.

90'

..,. . .

, ,

, ,

R = 3T R
10

R=

T R=;i

R=

IGURE 28-17 Standard face

ontours for straight grinding

vheels. (Courtesy of

. .

:arborundum Compony.)

770

Crwnlm 28 Abrasive Machining Processes The operation for whrch the abrasrve wheel is intended wrll also influenc wheel shape and size.The major use categories are the followtng

1 . Cuttzng off for shcrng and slothng parts; use thm wheel, organic bond 2. Cylwtdn'cal between centers' grinding outslde diameters of cylindncal workp~ec 3. Cylindrical,centerle~s. grindingoutside diameters with work rotated by regulating 4. Internal cyl~ndrzcal: gnnding bores and large holes 5. Snaggmng:removlnglarge amounts of metal without regard to surfacefinish or taler 6. Surface gnnding. grinding flat workpieces 7. Tool gr~ndmg: for grinding cutting edges on tools such as drills, millmg cutters,
reamers, and slngle-point high-speed-steel tools 8. Offhand grmdmg: work or the grinding tool is handheld In many cases, the classification of processes coincides wtth the classificati machines that do the process. Other factors that wrll influence the choice of wheel selected include the workpiece material, the amount of stock to be removed, the s of the workpiece, and the accuracy and surface finish desired. Workpiece maten a great impact on choice of the wheel. Hard, high-strength metals (tool stee steels) are generally ground wrth aluminum oxide wheels or cub~c boron nitnde Silicon carbide and CBN are employed in grindmg brittle materials (cast iron rarnics) as well as softer, low-strength metals such as alummum, brass, copper, and br Diamonds have taken over the cutting of tungsten carbides, and CBN is used for slon grrnding of tool and die steel, alloy steels, stainless steel, and other very matenalsThere are so many factors that affect the cutting actlon that there are no and-fast rules with regard to abrastve selection. Selection of grain size a determined by whether coarse or fine cutting an are desired. Coarse grams take larger depths of cut and cut more rap~dly. Hard with fine grains leave smaller tracks and the~efore are usually selected lor finishing If there 1s a tendency for the work material to load thc wheel, larger grains with a open structure may be used for fin~shmg.

BALANCING GRINDING WHEELS Because of the high rotation speeds involved, g r ~ n d ~ n wheels g must never less they are in good balance.As1ight tmbalance will produce vibrations th waviness in the work surface. It may cause a whecl to break, with the probahil~ty nous damage and injury.The wheel should be mounted with proper bushings so fils snugly on the spindle of the machine. Rings of blotttng paper bhould be plae tween the wheel and the flanges to ensure that the clamping pressure is evenly d uted. Most grinding wheels will run in good balance if they are mounted prope trued Most machines have provision for compensating for a small amount of wh balance by attaching weights to one mounting flange Some have provtsion for se tomatic balancing with weights that are permanently attached to the machine sp SAFETY I N GRINDING Because the rotational ~peeds are quite high, and the strength of grinding wheels 1 ly much less than that of the materials being ground, senous accidents occur much quently in connection with the use of grmding wheels.Virtually all such accidents c avoided and aTe due to one or a combination of four causes first, grindmg whe casionally operated at unsafe and improper speeds All gnnding wheels are clea with the maximum rpm value at which they should be ro(ated.They are a 1 1t siderably above the designatedrpm and are safe at the specified speed unless abused. should never, under any condmt~on, be operated above the rated speed. Second, a very man form of abuse,frequently accidenfal,~~ dropping the wheel or striking it agarnst a object This can cause a crack (which may not be readtly vtsible), resultrng m suhse fa~lure of the wheel while rotatmg at high speed under load. If a wheel s dropped or agamst a hard object,it should be drscarded and never used unless tested at above the speed in a properly designed test stand.A third common cause of grinding wheel fail

SECTION 28.5
Air s c r r a r

Grinding Machines

771

$,"s\

scrubber

CBN-

wheel

M, "
eck,
'91.)
Lower nozzle

1I'

improper use, such as grinding against the side of a wheel that was designed for grinding only on its periphery.The fourth and most common cause of injury from grinding is the absence of a proper safety guard over the wheel andlor over the eyes or face of the operator. The frequency with which operators will remove safety guards from grinding equipment or fail to use safety goggles or face shields is amazing and inexcusable.

USE OF CUTTING FLUIDS IN GRINDING Because grinding involves cutting, the selection and use of a cutting fluid is governed by the basic principles discussed in Chapter 21. If a fluid is used, it should be applied in sufficient quantities and in a manner that will ensure that the chips are washed away, not trapped between the wheel and the work.This is of particular importance in grinding horizontal surfaces. In hardened steel, the use of a fluid can help to prevent fine microcracks that result from highly localized heating.The air scraper shown in Figure 28-18 permits the cutting fluid (lubricant) to get onto the face of the wheel. Metal air scrapers disrupt the airflow. Upper and lower nozzles cool the grinding zone, while a high-pressure scrubber helps deter loading of the wheel. Much snagging and off-hand grinding is done dry. On some types of material, dry grinding produces a better finish than can be obtained by wet grinding. Grinding fluids strongly influence the performance of CBN wheels. Straight, sulfurized, or sulfochlorinated oils can enhance performance considerably when used with straight oils. ACHINES
Grinding machines commonly are classified according to the type of surface they produce. Table 28-4 presents such a classification,with further subdivision to indicate characteristic features of different types of machines within each classification.Grinding on all machines is done in three ways. In the first, the depth of cut (dJ is obtained by infeed-moving the

chininq Processes

Cylindr~Cal ~nternal

Internal hameters of holes

Flat surfaces

Surface conventional
Creep feed

Chucking Planetary (work sthoonary) Centerless Reciprocating table or

I
I

Tool grinders
Other

Tool angles and geomeines

Spectal or any of the above

ulllversal Speoal Disk, contour, thread, fler~ble sbdft,swing frame,snag, pedestal, bench

wheel down Into the work or the work up into the wheel (Figure 28-19), The ( surface 1 s then produced by traversing the wheel across (cross feed) the workpi vice versa (Fwure 28-19). In the second method, known as plunge-cur grinding, the3 , movement is of the wheel being fed radially into the work wkle the latter revolv centers. It is similar to form cutting on a lathe; usually a formed grinding wheel is (Figure 28.14). In the third method, the work is fed very slowly past the wheel an

downfeed

lnfeed controlled by handwheel (manual)

, . " . ,r ,
Insets wheelhead.

urface grinder, with

...,
. ;

-lvements of
88bL

,",,L",

,Lv,-

lnfeed infinitely variable in

0.0001 t o 0.0025" increments

controlled automatically

Diamond-coated roll in process dressing

Grinding wheel continually

Compensates downward to
maintain size

1
&

High-volume
cutting fluid

Approx. 20 mrn; approx. 4.4 m m Conventional grinding Creep feed grlnd~ng

FIGURE 28-20 Conventional gr~nding contrasted to creep feed grind~ng.Note that crush roll dress~ng 1 s used here; see Ftgure 28-14.

grmding for CBN applicahons.) The CFG method, often done in the surface grindlng mode, is markedly different from conventional surface grmding.The depth of cut IS increased 1000 to 10,000 times, and the work feed ratio is decreased in the same proportion: hence the name creep feed ,qrmdtng The long arc of contact between the wheel and the work increases the cutting forces and the power required. Therefore, the machine tools to perform t h ~ s type of grinding must be specially designed with high static and dynamic stability,stick-shp-free ways, adequate damping, increased horsepower, infinitely variable spindle speed, varable but extremely consistent table feed (especially in the low ranges), high-pressure cooling systems, integrated devices for dressing the grinding wheels, and specially designed (soft with open structure) grinding wheels.The process is mainly being applied to grind~ng deep slots with straight parallel sides or to grinding complex profiles ID diff~cult-to-grind materials. The process 1s capable of produc~ng extreme precision at relatively high metal removalrates. Because the process can operate at relatively low surface temperatures, the surface integrity of the metals being ground is good. However, In CFG, the grinding wheels mubt malntaln their initial profile much longer,so continuous dressing is used that is form-tru~ng and dressing the grindmg wheel throughout the process rather than between cycles Continuous crush dressing results in higher MRRs, Improved dimensional accuracy and form tolerance, reduced grinding

total downfeed or depth (d) 1s accompl~rhed in a single pass (Hgure 28-20).Thisis called creep feed grindtng (CFG). (Table 28-5 compares CFG lo convenhonal and high-speed -

Cil~ndmg Vanable
Wheel speed (fprn)

Convenrronal Gr~nding
5500-9500 versus 4500-6500

Creep Feed Gmnd~ng Highpeed Gnnd~ng

500&9000 versus 3000-5000 0.5-5 O.lN-0.250 12000-25000 5-20 250-.500

v~lrl[~rd

Table speed (fpm) Feed (j;)in./pass Grinding fluids

80-150 0.0005-0.0015 10% heavy-duty soluble oil or >5% light-duty soluble for light feeds

Sulhiizsd or sulfochlorinated straight grinding oil applied at 80 to 100 gallrnin at 100 psi or

more

forces (and power), and reduced thermal effects while sacrificing wheel wear. Cr grindmg e l i i a t e s preparatory operations such as milling or broachmg, since pr ground into the solid workpiece.This can result in significant savings in unit pa Gnnding machines that are used for precision work have certain importa acterlstlcs that permit them to produce parts having close dimensional toleranc are constructed very accurately, with heavy, rigid frames to ensure permanency o ment Rotating parts are accurately balanced to avoid vibration. Sp~ndles are m In very accurate bearings, usually of the preloaded ball-bear~ng type. Contra provided so that all movements that determine dimensions of the workpiece c made with accuracy-usually to 0 001 or 0.00001 in. The abrasive dust that results from grrndrng must be prevented from en between moving parts All ways and bearings must be fully covered or protect seals. If thls is not done, the abrasive dust between movmg parts becomes embe the softer of the two, causing it to act as lap and abrade the harder of the twos resulting in permanent loss of accuracy. These speclal charactenstics add consrderably t o the cost of these mac require that they be operated by trained personnel. Production-type gnnders fully automated and have higher metal removal rates and excellent dimensional a Fine surface finish can be obtalned very economically.

CYLINDRICAL GRINDING
Center-type cylzndrical grindzng is commonly used for producing external surfaces. Figures 28-14 and Figure 28-21 show the basic prrnclples and mot1 process The grrndlng wheel revolves at an ordinary cutting speed, and thew rotates on centers at a much slower speed, usually from 75 to 125 ftimm.llI wheel and the workpiece movem opposite directions at their polnt of contact. of cut is determmed by infeed of the wheel or workp~ece. Because this motion termines the finished diameter of the workpiece, accurate control of this mo required. Provision is made to traverse the workpiece wrth the wheel, or the be reciprocated past the wheel. In very large grinders, the wheel 1s reciprocate the detail of the of the massiveness or the work. For form or plunge grind~ng, maintained by periodic crush roll dressing. A plazn center-type cylmdrical grrnder is shown in Figure 28-21. On this work is mounted between headstock and tailstock centers. Solid dead cente ways used in the tallstock, and provision is usually made so that the headstock c be operated either dead or alive. High-precision work is usually ground wit head~tock center, because this eliminates any possibility that the woikpiece will of round due t o any eccentricrty in the headstock The table assembly can be reciprocated, in most caTes, by using a hydra The speed can be varied, and the length of the movement can be controlled b of adjustable tnp dogs. Infeed is provided by movement of the wheelhead at rrght angles to th dmal axis of the table.The spindle is driven by an electric motor that is also m the wheelhead. If the Infeed movement is controlled manually by some type drive to provlde control to 0.001 m. or less, the machine 1s usually equrpped wrt readout equipment to show the exact stze heingprodnced. Most production-type have automatic rnfeed with retraction when the desired size has been 0btained.S chines are usually equipped with an automatic diamond wheel-truing device that the wheel and resets the measuring element before grinding is started on each The longitudinal traverse should be about one-fourth to three-fourths of th width for each revolution of the work. For light mach~nes and f i e finishes, it sha held to the smaller end of this range.The depth of cut (infeed) vanes with th of the grinding operation and the finish desired When grinding is done to ob rate size, infeeds of 0.002 to 0.004 m are commonly used for roughing cuts For the infeed is reduced to 0.00025 to 0.0005 in.The design allowance for grind be from 0.005 to 0.010 in. on short parts and on parts that are not to be h a long or large parts and on work that is to be hardened, a grinding allowance

Wheelhead sl~deway
/

Machine base

Table

Workpiece (Dl

Table

Tailstock

Movements 1. Wheel 2 Work (rotates) 3. Traverse 4. lnfeed

0.015 to 0.030 in. is desirable. When grinding is used primarily for metal removal (called abrasive machining), infeeds are much higher, 0.020 to 0.040 in. being common. Continuous downfeed is often used, with rates up to 0.100 in./min being common. Grinding machines are available in whicll the workpiece is held in a chuck for grinding both external and internal cylindrical surfaces. Chucking-type external grinders are production-type machines for use in rapid grinding of relatively short parts, such as ball-bearing races. Both chucks and collets are used for holding the work, the means dictated by the shape of the workpiece and rapid loading and removal. In chucking-type internal grinding machines, the chuck-held workpiece revolves, and a relatively small, high-speed grinding wheel is rotated on a spindle arranged so that it can be reciprocated in and out of the workpiece. Infeed movement of the wheelhead is normal to the axis of rotation of the work (Figure 28-21). CENTERLESS GRINDING Centerless grinding makes it possible to grind both external and internal cylindrical surfaces without requiring the workpiece to be mounted between centers or in a chuck. This eliminates the requirement of center holes in some workpieces and the necessity for mounting the workpiece, thereby reducing the cycle time. The principle of centerless externul grinding is illustrated in Figure 28-22. Two wheels are used.The larger one operates at regular grinding speeds and does the actual grinding. The smaller wheel is the regulating wheel. It is mounted at an angle to the plane of the grinding wheel. Revolving at a much slower surface speed-usually 50 to 200 ftlmin-the regulating wheel controls the rotation and longitudinal motion of the workpiece and is a usually a plastic- or rubber-bonded wheel with a fairly wide face.

776

CHA~E 28 R Abrasive Machining Processes

A.

B.
C.

D.
E.

Grinding wheel Grinding face Regularing wheel Workpiece Work rest blade

FIGURE 28-22 Centerless

grinding showing the relationship among the grinding wheel, the regulating wheel, and the workpiece in centerless method. (Courtesy of Carborundum Company.)

fl=Angleoftilt of regulating wheel

Movements wheel 2. Work 3. Regulating wheel 4. Infeed 5. Traverse

1. Grinding

the grinding wheel and regulating wheel.This axial feed is calculated approximately by the equation

F
where

ND sin 4

F = feed (mmlmin or in./min)

= diameter

of the regulating wheel (mm or in.)

N = revolutions per minute of the regulating wheel

4 = angle of inclination of the regulating wheel

Centerless grinding has several important advantages: 1 . It is very rapid; infeed ccnterless grinding is almost continuous. 2. Very little skill is required of the operator. 3. It can often be made automatic (single-cycle automatic). 4. Where the cutting occurs, the work is fully supported by the work rest and the lating wheel.'Il~ispermits heavy cuts to be made. 5. Because there is n o d 6. Large grinding wheels can be used, thereby minimizing wheel wear. Thus centerless grinding is ideally suited to certain types of mass-production opera The major disadvantages are as follows:

1. Special machines are required that can do no other type of work.

2. The work must be round-no flats, such as keyways, can be present. 3. Its use on work having more than one diameter or on curved parts is limited. 4. In grinding tubes, there is no guarantee that the OD and Internal Diameter are concentric. Special centerless grinding machines are available for grinding balls and tapered pieces.The centerless grinding pri~~ciple can also be applied to internal grinding, b external surface of the

IS started. However, it assures that the internal and external surfaces will be concentric. The operation is easily mechanized for many apphcations.

SURFACE GRINDING MACHINES Surface grinding machines are used primarily to grind flat surfaces.However formed, irregular surfaces can be produced on some types of surface grinders by use of a formed wheel.There are four basic types of surface grinding machines,differing in the movement of their tables and the orientation of the grinding wheelspindles (Figure 28-23):
1 . Horuontal qpindle and reciprocat~ng table 2. Vertical spindle and reciprocating table 3. Horizontal spindle and rotary table 4. Vertical spindle and rotary table The most common type of surface grinding machine has a reciprocating table and horizontal spindle (Figures 28-19). The table can be reciprocated longitudinally either by handwheel or by hydraulic power. The wheelhead is given transverse (cross-feed) motion at the end of each table motion, again either by handwheel or by hydraulic power feed. Both Lhe longitudinal and transverse motions can be controlled by limit switches. Infeed or downfeed on such grinders is controlled by handwheels or automaticalIy,The size of such machincs is determined by the size of the surface that can be ground. In using such machines, the wheel should overtravel the work at both ends of the table reciprocalion, so as to prevent the wheel from grinding in one spot while the table is being reveraed.The transverse or cross-feed motion should be one-fourth to threefourths of the wheel width between each stroke.

Movements 1. Wheel 2 lnfeed 3 Work table traverse

A
B

Grind~ng wheel
Grrnd~ng face Shaft Workplece

C.

D .
E

Magnetic chuck on table

[c)

Movements 1. Wheel 2. Work tabfe totailon 3 lnfeed 4. Cross feed

[d)

Movements
1. Wheel 2. Inieed 3. Work teble rotation

CHAPTER 28 Abrasive Machining Processes

Vertkal-spindle reciprocanng-table su@dce grinders differ basically from tho6 horizontal spindles only in that their spindles are vertical and that the wheel dm must exceed the width of the surface to be ground. Usually, no transverse mo of e~ther the table or the wheeIhead is provided. Such machines can produce very surfaces. Rotary-ruble rurffacegrzndem can have either vertical or horizontal spmdleq those with horizontal spindles are limited in the type of work they wiU accommodate therefore are not used to a great extent. Vertccal-spmdlerotary-tdble ~urface grin are primarily production-type machines. They frequently have two or more grin heads, and therefore both rough grinding and finish grindmg are accomplished in rotation of the workpiwe. The work can b e held either on a magnebc chuck or in cia1 fixtures attached to the table. By using special rotary feeding mechanisms,machines of t h ~type s often are automatic.Parts are dumped on the rcttary feeding table and Led automatically onto holding devices and moved past the grinding wheels. After they pass the last y"" head, they are automatically unloaded.

DISK-GRINDING MACHINES
Disk grinders have relatively large side-mounted abrasive disksThe work is heldaga one side of the disk for grinding. Bothsingle- and double-disk grinders are used:in latter type the work is passed between the two disks and is ground on both sides taneously. a n these machmes, the work is always held and fed automatically. single-disk grmders the work can be held and fed by hand while resting on a s table.Although manual disk grindmgis not very precise,flat surfacescan beobtaine rapidly with little or no tooling cost. On specialized, production-type machines, exc accuracycan be obtained very economically.

TOOL AND CUlTER GRINDERS Simple, single-point tools ate often sharpened by hand on bench or pedestal grin (off-hand grinding).More complex tools, such as milling cutters, reamers, hobs, single-pomt tools for praductlon-type operations require more sophikticated grin machines, commonly called unrversal$ool and cutter grinders. These machines are ilar to small universal cyhdrical center-type grinders, but they d~ffer in four imp0 re8pecrs:

1 . The headstock is not motorized. 2. The headstock can be swiveled about a horizontal as well as a vertical axis. 3. The wheelhead can be raised and lowered and can be swiveled through at rotation about a vertical axia. 4. All table motions are manual. No power feeds being provided
Specificrake and clearanceanglesmust be created, often repeatedly,on a or on duplicate tools.Too1 and cutter grinders have a high degree of flexibility them so that the requiredrelationships between the tool and the grinding whee established for almost any type of tool. Although setting up such a grinder ' complicated and requires a highly skiled worker, after the setup is made for a p job, the actual grinding 18 accampli&ed rather easily Figure 28-24 shows several setups on a tool and cutter grinder. Hand-ground cutting tools are not accurate enough for automated mac processes. Many numerically controlled (NC) machine tools have been sold on the that they can position wosk to very close tolerances-within Q.M)01 to 0.0002 to have the initial workpieces produced by those machines out of tolerance by a 0.015 to 0.020 in. In most instances, the culprit was a poorly ground tool. For e twist driU with a point ground 0.005 in off-center can "walk" as much as 0.015 in.,& ingpaor hole location. Many wmpanies are turning to computer numeric contro grinders to handle the regrinding of their cutting tools. A six-axis CNC grinder is c of restoring the, proper tool an& (rake and clearance),concentricityY cutting edg dimensional size

.- - .

is held in a device that perm~ts all possible angles to be ground. (b) Edges of a large hand reamer are being ground.

Three typ~cal setups for grind~ng single- and mult~ple-edge tools on a universal tool and cutter grinder. (a, J ~ , F ~ , C -

1 cutter is sharpened w~th a cupped grind~ng wheel.

MOUNTED WHEELS AND POINTS Mounted wheels and points are small grinding wheels of varlous shapes that are permanently attached to metal shanks that can be inserted in the chucks of portable, high-speed electrlc or air motors. They are operated at speeds up to 100,000 rpm, depending on their diameters, and are used primarily for deburrmg and finishing in mold and dte work. Several types are shown in Figure 28-25. COATED ABRASIVES Coated abra~zves are being used increasingly in finishlug both metal andnonmetal products. These are made by gluing abras~ve grains onto a cloth or paper backmg (Figure 2826) Synthebc abrasives--aluminum oxide, silicon carbide, aluminum, zirconia, CBN, and diamond-are used mast commonly, but some natural abrasives-sand, flint, garnet, and emery-also are employed.Various types of glues are utllized to attach the abrasive grains to the backing, usually compounded to allow the finished product to have some flexibility. Coated abrasives are available in sheets, rolls, endless belts, and disks of various slzes. Some of the available forms are shown iuFigure 28-26. Although the cutting action of coated abrasives basically is the same as with grindmg wheels, there is one major difference: they have little tendency to be self-sharpenedwhen dull grains are pulled from the backing. Consequently, when the abrasive particles become dull or the belt loaded, the belt must be replaced. Fmer grades result in finer first cuts but slower material removal rates (MRR).This versatile process is now widely used for rapid stock removal as well as fine surface fishing.

28-25 Examples of d abms~ve wheels and . Courtesy of Norton

Belt composition

Grit Size-grade
vs
Approx. Fln~sh irms) Name

Bonds
Makecoat Sire coat

i %

Resin over glue Resin over resin

Glue Resin WP = waterproof

Platen grinder

Honing is a stock-removal process that uses fine abrasive stones to remove very amounts of metal. Cutting speed is much lower than that of grinding.The processi to size and finish bored holes,remove common errors left by boring (taper, wavines tool marks), or remove the tool marks left by grinding.The amount of metal rein is typically about 0.005 in. or less. Although honing is occasionally done by hand, finishing the face of a cutting tool, it usually is done with special equipment. Most ing is done on internal cylindrical surfaces, such as automobile cylinder walls.T%e ing stones are usually heldina honing head, with the stones being held against the with controlled light pressure. The honing head is not guided externally but, ins floats in the hole, being guided by the work surface (Figure 28-27).

II
Schematic c
>wingthe
I

Ve= osc~llat~ng speed Vr = rotatlng speed

V, A
= resulting cutting

speed
=

inclined angle

the stones a and oscillate e to produce pattern. Typ

P, are glven
Honing Parameters Conventional Abrasives Diamonds

For:

CBN
35-90
2-4

Hlgh MRR Best quallty servlce

V , (mlmm)
P, (NJmm?

20-30
1-2
5-30

40-70
2-8 40-70

V,(m/mn)

20-60

The stones are given a complex motlon so as to prevent a slngle grit from repeatmg ~ t path s over the work surface. Rotation is combined with an oscillatory axla1 motlon. For external and flat surfaces, varying oscillatory motions are used. The length of the motions should be such that the stones extend beyond the work surface at the end. A cutting fluid 1s used in virtually all hontng operations The cr~t~cal process parameters are rotai~onal speed, V,, oscillation speed, V,, the length and posltlon of stroke, and the homng stlck pressure. V, and the inchation angle are both products of V,) and V, Virtually all honing is done with stones made by bondlng together varlous hne artific~al abrasives Honlns stones differ from gr~nding wheels in that add~tional materials, such as sulfur. resm, or wax, are often added to thc bonding agent to mod~fy the cuttlng action.The abrasive grains range In slzefram80 to 600 gnt.The stones are equally spaced about the periphery of the tool. Reference values for Vc and honing stick pressure, P,, for various ahraslves are shown in Figure 28-27. Single- and multiple-spindle honing machines are ava~lahle in both horizontal and vertical types Some are equlpped with specla1 sensitive measuring devices that collapse the homng head when the desired slze has been reached. For honlng smgle, small, Internal cylindrical surfaces, a procedure is often used wherein the workpiece rs manually held and rec~procatedover a rotat~ng hone. If the volume of work is sufficient, honing is a fairly inexpenswe process. A complete honing cycle, including loading and nnloadtng the work, is often less than one minute. Size control within 0.0003 in is achieved routinely.

JPERFINI!

G
Superfinirh~ngis a vanatlon of honing that is typically u ~ e on d flat surfaces.The process is.

1, Very light, controlled pressure, 3 0 to 40 p s ~ 2. Rapid (over 400 cycles per mlnute), short strokes-less than 'i41n. 3. Stroke paths controlled so that a single grit never traverses the same path twtce 4. Coplous amounts of low-viscosity lubricant-coolant flooded over the work surface

782

C H ~28 R Abrasive Machining Processes

Peaks penetrating the

a

FIGURE 28-28 In superfinishing and hon~ng, a fllm of lubricant is establ~shed between the work and the abrasive stone as the work becomes smoother.

abrasive action

Film of lubricant rna~nta~ned between smoother surface and hon~ng stlck No further

This procedure, illustrated in Figure 28-28, results in surfaces of very umfom,repea smoothness. Superfinishing is based on the phenomenon that a lubricant of a given viswsi establish and maintain a separating,lubricatingfilm between two mating surfaces i roughness does not exceed a certain value and if a certain critical pressure. holdin apart, is not exceeded. Consequently, as the minute peaks on a surface are cut the honing stone, applied with a controlled pressure, a certain degree of smoo achieved.The lubricant establishes a continuous film between the stone and the piece and separates them so that no further cutting action occurs. Thus, \yl pressure, lubricant, and honing stone, each workpiece is honed to the same smoothness. Superfinishing is applied to both cylindrical and plane surfaces The amo metal removed usually is less than 0.002 in., most of it being the peaks of the su roughness. Copious amounts of lubricant-coolant maintain the work at a uniorm l l abraded metal particles to prevent scratching perature and wash away a

LAPPING
Lapping is an abrasive surface finishing process wherein fine abrasive particle5 charged (caused to become embedded) ~ n t a o soft material, called a lap.The ma the lap may range from cloth to cast iron or copper, but it is always softer t materid to be finished, being only a holder for the hard abrasive particles. Lap applied to both metals and nonmetals. As the charged lap is rubbed against a surface,the abrasive particles in the s of the lap remove small amounts of material from the surfaceto be mach~ned.Thi abrasive does the cutting, and the soft lap is not worn away because the abras' cles become embedded in its surface instead of moving across it.This action a1 curs when two materials rub together in the presence of a fine abrasive: the soft forms a lap, and the harder one is abraded away In lapping,the abrasive is usually carried between the lap and the works some sort of a vehicle, such as grease, oil, or water. The abrasive particles are fro grit up to the finest powder sizes. As a result, only very small amounts of removed, usually considerably less than 0.001 in. Because it is such a slow metal process, lapping is nsed only to remove scratch marks left by grinding or honin obtain very fkal or smooth surfaces, such as are required oil gage blocks or fo tight seals where high pressures are involved. Materials of almost. any hardness can be lapped. However, it is difficult to materials because the abrasive tends to become embedded.The most commo terial is fine-grained cast iron. Copper is used quite often and is the common for lapping diamonds. For lapping hardened metals for metallographic exa cloth laps are used. Lapping can be done either by hand or by special machines. In hand lap lap is flat, simdar to a surface plate. Grooves are usually cut across the surface to collect the excess abrasive and chips. The work is moved across the surfac lap, using an irregular, rotary motion, and is turned frequently to obtain a unifo ting action.

In lapping machines for obtaining flat surfaces,workpieces are placed loosely in holders and are held against the rotating lap by means of floating heads. The holders, rotating slowly,move the workpieces in an irregular path. When two parallel surfaces are to be produced, two lapsmay be employed,one rotating below and theother above the workpieces. Various typesof lapping machines are available for lapping round su1faces.A special type of centerless lapping machine is used for lapping small cylindrical parts, such as piston pins and ball-bearing races. Because the demand for surfaces having only a few micrometers of roughness on hardened materials has become quite common, the use of lapping has increased greatly. However, it is a very slow method of removing meta1,obviously costly compared with other methods, and should not he specified unless such a surface is absolutely necessary.

1 28.8 FREEABRASIVES
ULTRASONIC MACHINING
Ultrasonic machining (USM), sometimes called ultrasonic impact grinding,employs an ultrasonically vibrating tool to impel the abrasives in a slurry at high velocity against the workpiece.The tool is fed into the part as it vibrates along an axis parallel to the tool feed a t an amplitudeon the order of several thousandths of an inch and a frequency of 20 kHz. As the tool is fed into the workpiece, a negative of the tool is machined into the workpiece. The cutting action is performed by the abrasives in the slurry, which is continuously flooded under the tool.The slurry is loaded up to 60% by weight with abrasive particles Lighter abrasive loadings are used to facilitate the flow of the slurry for deep drilling (up to 2 in. deep). Boron carbide, aluminum oxide, and silicon carbide are the most commonly used abrasives in grit sizes ranging from 400 to 2000. The amplitude of the vibration should be set approximately to the size of the &.The process can use shaped tools to cut virtually any material but is most effective on materials with hardnesses greater than R,40, including brittle and nouconductive materials such as glass. Figure 28-29 shows a simple schematic of this process. USM uses piezoelectric or magnetostrictive transducers to impart high-frequency vibrations to the tool holder and tool.Abrasive particles in the slurry are accelerated to great speed by the vibrating tool. The tool materials are usually brass, carbide, mild

RE 28-29 Sinking a hole in a workpiece with an ultrasonically vibrating tool driving an abrasive slurry.

steel, or tool steel and will vary in tool wear depending on thetr hardness Wear r (workpiece material removed versus tool materiallost) from 1:l (for tool steel) to (for glass) are possible. Because of the high number of cyclic loads, the tool mus strong enough to resist fatigue fdure. The cut will be oversize by about twice the slze of the abrasive particles b used, and holes wdl be tapered, usually hmiting the hole depth-to-diameter rati about 31.Surface roughness is controlled by the size of the abras~ve parttcles (finer ish with smaller particles). Holes, slots, or shaped cavities can be readily eroded in hard material-conductive or nonconductive, metallic, ceramlc, or compo Advantages of the process include that it is one of the few m a c h n g methods cap of machining glass. Also, it is the safest machining method. Skin is impervious t process because of ds ductility. High-pitched noise can be a problem due to seco vibrations. In addition to machining, ultrasonic energy has also been employed fo ing, lapping, deburring, and broaching. Plastics can be welded using ultrasonic en

WATERJET CUTTING A N D ABRASIVE WATERJET M A C H I N I N G Waterlet cutting (WJC), also known as waterjet machirung or hydrodynamic much uses a htgh-velocity fluid jet impmgtng on the workpiece to perform a slitting ope (Figure 28-30).Water is ejected from a nozzle orifice at high pressure (up to 60,OO The jet is typically 0.003 to 0.020 in.in dtameter and exrts the orifice at velocities> 3000 ftlsec Key process parameters include water pressure,orifice diameter, wate rate, and working distance (distance between the workpiece and the nozzle). Nozzle materials Include synthetic sapphire, due to its machinability and tance to wear.Too1 life on the order of several hundred hours is typical Mechanis tool failure rnclude chipping from contaminants or constriction due to mmeral d This emphasizes the need for high levels of filtration prior to pressure ~ntensifi In the past, long-cham polymers were added to the water to make the jet more ent (~.e.,not come out of the jet dispersed). However, with proper nozzle design,a coherent water jet may be produced without additives. The advantages of WJC include the ability to cut materials without bur crushing the material being cut.Ftgure 28-30 shows a comparison (end new) of corrugated boxboard wlth a mechanical knife and with WJC.The mechanism fn rial removal is simply the impingmg pressure of the water exceeding the comp f themater1al.This limits the materials that can be cut by the process to1 strength o plasttcs, and other soft nonmetals, which is the major d~sadvantage of the p Alternative fluids (alcohal, glycerine, cooking oils) have been used in processing baked good$ and frozen foods. Other disadvantages include that the process and requues operators to have hearmg protection. The majority of the metalworktng applications for waterjet cuttmg re addition of abrasives.Thts process is known as abrasive %aterjetcuttmg (AW range of materials, including metals, plastics, rubber, glass, ceramics, and comp be machined by AWC. Cuttmg feed rates vary from 20 m.imm for acoustic in.imin for epoxies and 500 in./min for paper products. Abrasives are added to the watejet in a mkmg chamber on the downstr the waterjet onfice. A single, central waterjet wth slde feeding of abrasives in chamber is shown m Figure 28-30. In the mixing chamber, the momentum of the transferred to the abrasive particles, and the water and particles are forced out thr AWC nozzle orifice, also called the muring tube.This design can be made quite however,it also experiencesrapid wear in the mixing tube. An alternate configur feed the abrasives from the center of the nozzle \nth a converging set of angled Imparting momentum to the abrasives.This nozzle design produces better water and abrasives as well as Increased nozzle Me.The inside dtameter of t IS normally from 0.04 to 0.125 m. in diameter, These tubes are nonnally made Generally, the kerf of the cut is about 0.001 in. greater than the no AWC requires control of additional process parameters over waterjet mac cluding abrasive mafenal (density, hardness, shape), abrasive size or grit, ab rate (poundsper minute), abrasive feed mechamsm (pressurized or suction),

Abrasive hopper

High-pressulewater to no~zle
X-Y Gantry robat

Abrasive cutting head

Pneumatic control irne High-pressure water in

Mount~ng -------.

extenston tube

URE 23-30 Schernatlc diagram of hydrodynamic jet machini rntensfier elevates the fluid to the desired nozzle pressure while accumulator smoothes out the pulses in the fluid jet. Schematic n abraslve waterjet machining n o d e 1s shown on the right.

ABRASIVE JET MACHINING One of the least expensive of the nontraditional processes iy abrasivejet machining (AJM).
AJM removes material by a focused jet of abrasives and is similar in many respects to AWC, with the exception that momentum is transferred to the abrasive partrcles by a jet of inert gas.Abrasive velowt~es on the order of 1000ftlsec are possible withAJM.The small mass of the abrasive particles produces a m~croscale chlpping action on the workp~ece material.Th~s makes AJM ideal for processing hard, brittle materials,includingglass, silicon, tungsten, and ceramics. It is not compatible with soft, elastic materials. Key process parameters rnclude working d~stance, abrasive flow rate, gas pressure, and abrasive type. Working drstance and feed rate ale controlled by hand. If necessary, a hard mask can be placed on the wurkplece to control dimensions. Abrasives are typically smaller than those used in AWC. Abrasives are typically not recycled, since the abrasives are cheap and are used only on the order of several hundred grams per hour. To minimi~e part~culate contaminat~on of the work environment, a dust-collection hood should be used in concert with the A I M system.

DESIGN CONSIDERATIONS IN GRINDING

Almost any shape and ssze of work can be finished on modern gnnding equipment, including flat surfaces,straight or tapered cylinders,irregular external and internal surfaces, cams, antifriction bearing races, threads, and gears. For example, the most accurate threads are formed from solrd cylindrical blanks on spec~al thread ginding machines. Gears that must operate without play are hardened and then finish ground to close tolerances.llvo important design recommendations are to reduce the area to be ground and to keep all surfaces that are lo be ground in the same or parallel planes (F~gure 28-31). 'his 18 an example of d o ~ g for a manufacfur~ng (DFM). Abrasive machining can remove scale as well as parent metal Large allowances of material, needed to permit conventronal metalcutting tools to cut below hard or abrasive mclusions, are not necessary for abrasive machining. An allowance of 0.015 in. is adequate, assuming, of course, that the part IS not warped or out of round. This small allowance requirement results in saving* in machining trme, m material (often 60% less metal is removed), and m shipping of untinished parts.

Orlg~nal des~gn of base plate

Onglnal deslgn of crankshaft bearlng bracket

Redes~gned to reduce w e ~ g h t and grlndlnq tlme

Redes~gn ellmlnated shoulders and made part su~table for Qrlndlng i n single setup

788

CRAIIFR28

Abrasive Machining Processes

Ilr Key Words

abrasive machining abrasive waterjet aluminum oxide attrition ccnterless grinding coated abrasive corundum creep f e d grinding crush dressing cabic boron nitride (CBN) cylindrical grinding diamond dressing emery friability G ratio garnet grade grinding honing lapping quartz resinoid bond rubber bond shellac bond silicate bond silicon carbide

ultrasonic machiniq vitrified bond

Ell Review Questions

1. What are machining processes that use abrasive particles for 23. The number of grains per square inch which actively con cutting tools called? 2. What is attrition in an abrasive grit? 24. Why are centerless gfndem so popular in industry corn 3. Why is friability an important grit property'? to center-type grindersl 4. Explain therelationship between grit size and surface finish. 5 . Why is aluminum oxide used more frequently than silicon 25. Explain how a SEM micrograph is made. Check the Inte or the library to find the answer. carbide as an zbrasive? 6. Why is CBN superior to silicon carbide as an abrasive in 26. Why are vacuum chucks and magnetic chucks widely use surface grinding but not in milling'! some applications'! 7. What malerials conlmonly are used as bonding agents in 27. How does creep feed grinding differ from conventi grinding wheels? surface grinding? 28. Why does a lap not wear, since it is softer than th 8. Why is the grade of a bond in a grinding wkcel important? being lapped? 9. How does grade dilfer from strutlture in a grinding wheel? 29. How do honing stones differ from grinding wheels? 10. What is crush dressing? 30. What is mcant by "charging"a lap7 11. How does loading differ from glazing? 12. What is meant by the statement that grinding is a mixture of 31. Why is a honing head permitted to float in a ho pmccsses? been bored? 13. What is accomplished in dressing n grinding wheel? 14. How does abrasive machining differ from ordinary grinding? 15. Whal is n grinding ratio or G ratio? How thick are these chips? 16. How is the feed of the workpiece controlled in centerless 34. What is the inclined angle in honing and what deter grinding? 35. Whal are the common causes of grinding accidents. 17. Why is grain spacing important in grinding wheels? 18. Why should a cutting fluid be used in copious quantities 36. What other machine tool docs a surface grinder resem 37. F~gure 28-11 showed residual stress distributions pro when doing wet grinding? by surface grinding. What is a residual stress? 19. How does plunge-cut grinding compare to cylindrical grinding? 20. If grinding machines are placed among other machine tools, 38. In grinding, what is infeed versus crosu Seed? what precautions must be taken? is vmy nois~r.Why? 21. What is the purpose of low-stress grinding? 22. How is low-stress grinding done compared to conventional 40. In AWC, what kceps the abrasive jet Srom machini grinding? orifice?

CI Problems
1. Perhaps you have observed the following wear phenomena:

A set 01marble or wooden steps shows wear on the treads in the rcgions where people step when they climb (or descend) thc steps. The highel- u p the steps, the less the wear on the tread. Given that soles of shoes (leather, rubber) are far softer than marble or granite,explain: a. Why and how the steps wear. b. Why the lower steps are more worn than the upper steps.

2. Explain why it is that a small particle of a be used to ahrade a surface made of the same rn (i.e., why does the small particle act harder or strong@ the bulk material)? 3. In grinding, both the wheel and workpiece are movi rotating). Using the data in Figure 28-11 and assu you are doing surface grinding (see Figure 2&1), some typical MRR values? How do these compar values for other machining processes, such as milli is the significance US this?

Ol(erkers$Crane Inst~liaZion
~ckolas has oeen g.ven tnc tasn of tnslal ng an orerhead crane (Figure CS 28) In one bay 01 iho vonTurkovidl ~anufaciurin~ Campany and assembly plant. Brackets for the rails of the crme are to be mounted on eight columns, four on each s~de of the bay area facing each other The rails for the crme w~ll span four columns. Each bradtet on each column w~ll need six holes in a circular paEerh.The holes must be accurately spaced wlthln 5 mlnutes of a c of each other.The axls of the holss must be parallel and normal to the face of $hecolumns.The center of the bolt hole clrcle musf be at a height of at least 20 i t from the floor, hut fhscenters of all elght bolt noIe a T C I & S must be on the same parallel plane, so that the rails for the craneare level and parallel with each other. Four o l the cdumns along the w a U have theft'facss flush with the wall

surface so thal mechancal clamp ng or attachments canno wemng of not be used The ou,ia.ng code WIU perm11 anything to thwe Columns 1. How would Nnkolas proceed to get the bolt holes located ~nthe ~rlght posttlon on the beams? 2 How would he get the hole patterns located properly with resped to each other on all e~ght beams? 9 List the equipment he will need 4. Make a skslch of any specral to01 you reaornmend. (H~nt: Check Chapter 25 for dr~llllg deslgns and ask your favorlte civ~l engtneer for suggestions regafd~ng surveying.)

I11

Abotd

30ft

NMstandard

-I

About 20 f t

Hor~zontal w~th re8pect to and rn line and in same plane as fanng columns

Goneref

Floor Inof level)

Figure CS 28 Overhead crane rnstallaklonat the vonTurkovtch Manufaclurlng Company.

THREAD AND GEAR MANUFACTURING

29.1 INTRODUCT~~N Screw-Thread Standardization and Nomenclature Types oI Screw Threads 29.2 THREAD MAKING Cutung Threads on a Lathe Cutting Threads on a CNCLathe Cutting Threads with Dies Selt-OpeningDie Heads 29.3 INTERNAL THREAD CUTTINCTAPPINC Collapsing Taps Hole Preparation Tapping in Machme Tools Tapping CnttingTune Special Threading andTapping Machines Common Tapping Problem Tapping H~gh-Strength Materials 29.4 THREAD MILLI 29.9 GE

11 29.1 INTRODUCTION
Screw threads and gears are important machine clements. Threadzng, thread cutt~ thread rolling refers to the manufacture of threads an external diameters Tapping to machining threads in (drilled) holes. Without these processes, our present tec loglcal society would come to a grinding halt. More screw threads are made each than any other machined element.They range in size from those used in small wa to threaded shafts 10 m. in diameter.They are made m quanuties ranging from several million duplicate threads. Their precision varies from that of inexpensiv ware screws to that of lead screws for the most precise machine tools. Consequ is not surprising that many very different procedures have been developed for $crew threads and that the produchon cost by the vanous methods varies great1 tunately, some of the most economical methods can provide very accurate results. ever, as in the des~gn of most products, the designer can greatly affect the ease an of producing specified screw threadsThus, understanding thread-making processes mits the designer to specify and incorporate screw threads into designs while avo needless and excessive cost. Gears transmit power or motion mechanically between parallel, intersec ' nonintersectmg shafts.Although usuallyhldden from sight,gears are one of them portant mechanical elements in our civilizat~on, possibly even surpassing the w slnce most wheels would not be turnmg were power not being applied to them gearsThey operate at almost unlimited speeds under a wide vanety of condit l~ons are produced each year in n7es from a few m~llirneters up tomore than 3 ameter. Often the requirements that must be, and arc routtnely,met in their manufa are amdzingly precise. Consequently, many special machines and processes have developed for produang gears Let us begin by discussing threads.

SCREW-THREAD STANDARDIZATION AND NOMENCLATURE A screw thread is a ridge of uniform section in the form of a helix on the external o ternal surface of a cylinder, or in the form of a conical spiral on the external or in surface of a frustrum of a cone These are called stralght or tapered threads, respe 1y.Tapercd threads are used on pipe jomts or other applications where hquid-tight are required. Straight threads, on the other hand, are used in a wide vanety of ap tions, most commonly on fastening devices, such as bolts, screws,and nuts, and as m

Helix angle

De~th of thread

S~ngle thread = 1 x pitch Double thread = 2 x patch

Triple thread = 3 x pltch

'--sod
Thread angte

-C
3 0 ' 3 0 "

elements on parts that are to be fastened together. But, as mentioned previously, they find very important applications in transmitting controlled motion, as in lead screws and precision measuring equipment. The standard nomenclature for screw-thread components is illustrated in Figure 29-1. ' h e symbol P, the pitch, refers to the distance from a point on one screw thread to the corresponding point on the next thread, measured parallel to the length axis of the part. In 1948,representatives of the United States, Canada, and Great Britain adopted the Unified and American Screw Thread Standards, based on the form shown in Figure 29-2. In 1968 the International Organization for Standards (ISO) recommended the adoption of a set of metric standards based on the basic thread profile. It appears likely that both types of threads will continue to be used for some time to come. In both the Unified and I S 0 systems, the crests of external threads may he flat or rounded. The root is usually made rounded to minimize stress concentrations at this critical area.The internal thread has a flat crest in order to mate with either a rounded or V-root of the external thread.A small round is used at the root to provide clearance for the flat crest of the external thread. In the metric system, thepitch is always expressed in millimeters, whereas in the American (Unified) system, it is a fraction having as the numerator 1 and as the denominator the number of threads per inch. Thus, 16 threads per inch,'Il,, is a 16 pitch. Consequently,in the Unified system, threads are more commonly described in terms of threads per inch rather than by the pitch. While all elements of the thread form are based on thepitch diameter, screw-thread sizes are expressed in terms of the outside, or major diameter and the pitch or number of threadsper inch. In threaded elements, lead refers to the axial advance of the element during one revolution; therefore, lead equals pitch on a single-thread screw.

TYPES OF SCREW THREADS

Eleven types, or series, of threads are of commercial importance, several having equiva29-3. As has been indicated, lent series in the metric system and Unified systems. See the Unified threads are available in a coarse (UNC and NC), fine (UNF and NF), extrafine (UNEF and NEF). and three-"pitch" (8,12,and 16) series, the number of threads per inch in accordance wiih an arbitrary determination based on the major diameter. Many nations have now adopted I S 0 threads into their national standards. Besides metric IS0 threads, there are also inch-based I S 0 threads, namely the UN series with which people in the united State&Canada, and Great Britain are familiar. I S 0 offers a wide range of metric sizes. Individual countries have the choice of accepting all or a selection of the I S 0 offerings. The size listings of metric threads start with " M and continue with the outside diameter in millimeters. Most I S 0 metric thread sizes come in coarse,medium, and fine

la1 External thread Rounded or fiat crest

of root contour and flank of thread

dimensional tables used for calculation of stress areas Rounded root optional or resulting from tool wear

Axis of external thread

lb) Internal thread

D - Major diameter of thread of nut

d = Major diameter of screw

Dl = Minor diameter of thread of nut dl = Minor diameter of screw

9=

Pitch diameter of thread of nut

(effective) diameter

I
Minor diameter

U/

I
FIGURE 29-2

11- 1'
90q

'

'

W
4

d2 = Pitch diameter of screw

14 t

H = Height of the complete theoretical

thread profile

H1 = Engagement
P = Pitch

AXISof screw thread --

Basic profiles of Unified and American screw threads: (a) external, (b) internal, and (c) metric.

SECTION 29.1

Introduction

793

Types o f Screw Threads

1 . Coarse-threadser!es/UNC and NC) For general use where not subjected to

vibratron.

2. Ffne-fhread serfes (UNF and NFI. For most automotive and aircraft work
3. Extra-fine-thread series (UNEF and NEF) For use w ~ t h th~n-walled matar~al or where a maximum number of threads are requ~red i n a gtven length.
4. E~ght-threadsenes (BUN and 8N). Elght threads per Inch for all diametersfrom 1to 6 in Used pr~marilyfor bolts on pipe flanges and cyl~nder head studs where

an inttlal tenslon must be set up to reslst steam or alr pressures. 5. Twelve-thread sene3 (12UN and 12N). Twelve threads per inch for d~ameters from 'through 6 In. Not used extensively
2

6. S~xteen-threadsenes L16UN and 16NJ. Sixteen threads per ~neh for d~arneters that requlre a fine from through 6 in. Usedfor a widevariety of appllcat~ons

thread

7. Amertcan Acme thread. This thread and the foliowlng three are used prlmartly in transmitt~ng power and motlon.
American Acme thread
I-p-

Mtn d e p t h D = O S x P Wldth F = 0.3707 X P Width C= F - 0.0052

8. Buttress thread. Buttress thread

9. Square thread.

Square thread Depth D = 0.5 X P Wfor screw = 0.5 X P Wfor thread groove In nut- 0.5 X P plus 0 001 to 0.002" clearance
1 1 . Amerrcan, standard pipe thread.Th~s thread is the

rpq C
10. 29" Worm thread.

Depth D = 0 750 X P W~dth of flat =0125xP Flank angle = 4 5 O In plane of axls

standard tapered thread used on pipe iornts in t h ~ s country. i n /it. The taper on all prpe threads is

29-Worm thread

Amerlcan standard plpe thread E = Effective thread length -A Taper = 314 in. per ft. 4 p +

Depth D = 0.6866 X P Depth F = 0.355 x P Depth C= 0,370 X P Angle = 2gQ In plane of axis

4 c C -

G-(0.05

1.1) P

FIGURE 29-3 Types of screw threads.

pitches. When a coarse thread is designated, it is not necessary to spell out the pitch example, a coarse 10-mm outside diameter (OD) thread is called out as "M 12." thread has a pitch of 1.75 mm, but the pitch may be omitted from the call-0ut.A 12-mm OD thread is available. It has a 1.25-mm pitch and must be designated "Iv 1.25" an extra-fine 12-mm OD thread having 0.75 mm-pitch would receive the d nation "M 12 x 0.75." The symbol "x" is not employed as a multiplication symbol in metric prac rather, it is used to relate these two attributes of the threads. The full description thread fastener obviously includes information beyond the thread specification.E type, length, length of thread, design of end, thread runout, heat treatment, appliec ishes, and other data may be needed to fully specify a bolt besides the designation o thread.The "x" should not be used to separate any of the other characteristics Here is another example of an I S 0 thread designation: metric thread designation M6 x 0.75-5g6g nominal size pitch tolerance class designation L t o l e r a n c e position tolerance grade tolerance posit~on L t o l e r a n c e grade crest-diameter tolerance symbol

i
I

pitch-diameter

j tolerance symbol

In the I S 0 system, tolerances are applied to "positlons" and "grades."Toleran sitions denote the h t s of pitch and crest diameters,using"e" (large),"g" (smaU),an (no allowance) for intemaI threads.The grade is expressed by numerals 3 through 9. Cfi 6 is roughly equivalent to U.S. grades 2A and medium-quality, general-purposet h & Below 6 is fine quality andlor short engagement. Above 6 is coarse q d t y andioa length of engagement. In the Unified system, screw threads are designated by symbols as follows:

I I

I I -thread I -number
-nominal

L c l a s s of thread series of threads per inch size

!

This type of designation applies to right-hand threads. For left-ha '" reads," ' LH are added after the class of thread symbol. In the Unified system, manufacturing tolerances are speciriea oy three Class 1is for ordnance and other special applications Class 2 threads are the nom duction grade, and Class 3 threads have minimum tolerances where tight I qulred. The letters A and B are added after the class numerals to ind~cate ex internal threads, respectively. The avail$pility of fasteners, particularly nuts, containing plastic them self-locking and thus able to resist loosening due to vibration, and the cia1 coatings that serve the same purpose, have resulted in Less use of finer-th~ fasteners in mass production. Coarser-tbread fastener! subject to cross-threading (binding)

Three baslc method4 are used to produce threads. cuttmg. rollzg. and ca~ting.Although both external and internal threads can be cast,relatively few are made in this manner, primarily m connection with die casting, investment casting, or the moldmg of plastics. Today, by far the largest number of threads are made by rolling. Both external and internal threads can be made by rolling, but the material must be ductile. Because rollrng is a less flexlble process than thread cutting.1~ is restricted essenhally to standa~dized and simple parts. Consequently,large numbers of external and ~nternal threads still are made by cutting processes, including giinding and tapping. ExternalThread Cutting Methods
Threading on an engine lathe Threading on an NC lathe With a die held in a stock (manual) With an automatic die (turret lathe or screw machine) or NC lathe By milling By grinding

Internal Thread Cutting Methods

Threading (on an engine liltbe or NC lathc) With a tap and holder (manual NC. machine, semiautamatic, or auturniilic) With a cc~llapsibletap (twret lathe, screw machine, or special threading machinc) By milling

C U T I N G THREADS ON A LATHE Lathes provided the first method for cutting threads by machine. Although most threads are now produced by other methods, lathes still provide the most versatile and fundamentally simple method. Consequently, they often are used for cutting threads on special workpieces where the configuration or nonstandard slze does not pennit them to be made by less costly methods. There are two basic requirements for thread cuthng on a lathe. F~rst, an accurately shaped and properly mounted tool is needed because thread cutting is a form-cutting operatlon. The resulting thread profile is determined by the shape of the tool and its position relative to the workpiece. Second, the tool must move longitudinally in a specific relationship to the rotation of the workpiece, because this determines the lead of the thread.This requirement is met through the use of the lead xrew and the split nut, which provide positive motion of the carnage relative to the rotation of the spindle. To cut a thread,it ts also eswntial that a constant positional relationshrp be mamtained among the workpiece, the cutting tool, and the lead screw. If this is not done, the tool will not be positroned correctly in the thread space on successive cuts. Correct relationship is obtained by means ot a tlzreadmng d ~ a(Figure l 29-4), which is driven directly by the lead screw through a worm gear. Because the workpiece and the lead screw are directly connected. the threadmg dial provides a means for estabhshing the desired pos~tional relat~onship between the workpiece and the cutting tool. The threading dial IS graduated into an even number ot major and half divlrions. If the feed mechanism is engaged in accordance with the following rulescorrect positioning of the tool will result:
1. For even-number threads: at any line on the dial 2. For odd-number rhreads: at any numbered h e on the dial 3. For threads ~nvolvrngi numbers: at any odd-numbered line on the dial starting lme on the d ~ a l 4. For or threads. return to the or~ginal To start cutting a thread, the tool usually is fed inward until it just scratcheb the work,and the cross-slide dial readmg is then noted or set at zero.The split nut is engaged and the tool perm~tted to run over the desired thread length. When the tool reaches the end of the thread, ~tn quickly withdrawn by means of the cross-sllde control.The split nut 1s then disengaged and the carriage returned to the starting position, where the tool is clear of the workpiece At this point the future thread w~ll be indrcated by a fine scratch 1ine.Thispermits the operator to check the thread lead by means of a scale or thread gage to assure that all settings have been made correctly.

Depth of cut

of cut

Compound straight

Compound swiveled

la)

FIGURE 29-4 Cutting a screw thread on a lathe, showing the method of supporting the work and the relationship o f the tool to the work with the compound swiveled. Inset shows face of thread~ng dial. Next. the tool is returned to its init~al zero depth position by returning th slide to the zero setting. By using the compound rest, the tool can be moved in proper depth for the first cut. A depth of 0.010 to 0.025 in. is usually used for cut and smaller amounts on each successive cut, until the final cut is made with of only 0.001 to 0.003 in. to produce a good finish. When the thrcad has been ly to its fuU depth, it is checked for size by means of a mating nut or thread g ting is continued until a proper fit is obtained. Figure 29-4 illustrates two methods of feeding the tool into the work. I fed radially, cutting takes place simultaneously on both sides of the tool. form-cutting procedure, no rake should be ground on the tool, and the top of must be horizontal and be set exadly in hne with the axrs of rotation of the wot erwise, the resulting thread profile will not be correct. An obvious disadvantag method is that the absence of side and back rake results in poor cutting (exc iron or brass).The surface finish on steel will usually be poor. Consequently, method commonly is used,with the compound swivelled 20". The cukting then f the tool, and some side rake can be provided. primarily on the left-hand edge o Proper speed ratio between the sprndle and the lead screw is set by means gear-change box. Modem industrial lathes have ranges of ratios available so tha ly all standard threads can be cut merely by setting the proper levers on the change gear box.

Cutting tool

Point on tool

Start point
(X, 2,)

RE 29-5 Canned

"tines called C codes are on CNC lathes to produce dr. See Chapter 26 for CNC

Specifiesthe absolute Xcoordinate of the tool after axial infeed. Initiates the single-passthreading cycle. Specifiesthe absoiute Zcoordinate of the tool after the threading pass Za Fn n Specifiesthe feed rate X, Z , Specifies the absolute Xand Z coordinates of the start point.
X,

G32

Cutting screw threads on a lathe is a slow, repetitious process that requlres considerable operator skiU.The cutting speeds usually employed are from one-third to onehalf of regular speeds to enable the operator to have time to manipulate the controls and to ensure better cutting.The cost per part can be high. which explains why other methods are used whenever possible.

CUTTING THREADS ON A CNC LATHE Computer numerical control (CNC) lathes and turning centers can be programmed to machine straight, tapered,or scroll threads.Threads machined using the same type of special tool have the thread shape shown in Figure 29-4.The tool is positioned at a specific starting distance from the end of the work (Figure 29-5). This distance will vary from machine to machine. Its value can he found in the machine's programming manual.The CNC software will have a set of preprogrammed machining routines (called G codes) specifically for threading. Beginning at thc start poinl, the tool accelerates to the feed rate required to cut the threads. The tool creates the thread shape by repeatedly following the same path as axial infeed is applied. For standard V-threads, the infeed can be applied along a 0" or 29' angle.The depth of cut for the first pass is the largest.The cutting depth is then decreased for each successive pass until thc required thread depth is achieved.A final finishing pass is then made with the tool set at the thread depth. CUlTING THREADS WITH DIES Straight and tapered external threads up to about 1; in. in diametcr can be manually cut quickly by means of threading dies (Figure 29-ha). Basically, these dies are similar to hardened, threaded nuts with multiple cutting edges.The cutting edges at the starting end

RE 29-6 (a) Solid ding die; (b) solidustable threading die; threading-die stock for round (die removed). (Courtesy of W-Greenfield Tap & Die.)

Ic)

798

C H A ~ Thre R ad ~ and ~ Gear Manufactur~ng

are beveled to aid in starting the dles on the w0rkpiece.A~a consequence, a few thr a1 the inner end of the workpiece are not cut to full depth. Such threading dies are of carbon or high-speed tool steel. Solid-type dies are seldom used m manufacturing because they have no provi for compensating fol wear.The solid-adjustable type (Flgure 29-6b) IS split and c adjusted over a small range by means of a screw to compensate for wear or to p a variation in the fit of the resulting screw thread.These types of threading dies a ally held in a stock for hand rotation. A suitable lubricant 1s desirable to p smoother thread and to prolong the life of the dre, since there is extensive frlction d the cutting process

SELF-OPENING DIE HEADS A major disadvantage of solid-type threading dres is that they must be unscrewed the workpiece to remove them.They are therefore not suitable for use on high-s production-type machines, and self-opening dce heads are used instead on turret 1 screw machines, numerically controlled (NC) lathes, and special threadmg mach cuttlng external threads. There are three types of self-opening die heads, all having four sets of adju multiple-point cutters that can be removed for sharpening or for interchang different thread sizes. This permits one head to be used for a range of thread siz Figure 29-7). The cutters can be positioned radially or tangentially,resulting in les flank contact and friction rubbmg. In some selt-opening &e heads, the cutters are c with an Interruption ln the circular form to provide an easily sharpened cutting The cutters are mounted on the holder at an angle equal to the helix angle of the As the name implies, the cutters m self-opening die heads are arranged to automatically when the thread has been cut to the desired length, thereby pe the die head to be quickly wrthdrawn from the workpiece. On die heads used on lathes, the operator must usually reset the cutters ln the closed posit~on beforem the next thread.The die heads used on screw machines and autornatlc threadin ch~nes are provided wlth a mechanism that automatically closes the cutters aft heads are withdrawn. Cutting threads by means of self-opening die heads is frequently called chasing. However, some people apply this term to other methods of thread cutting to cutting a thread ln a lathe.

'HREAD

CUTTING-TAPPING

The cuttmg of an internal thread by means of a multiple-point tool is called thre p z g , and the tool is called a tap. A hole of diameter slightly larger than the mi ameter of the thread must already exist,made by dnlliigireaming, boring, or die For small holes,solid hand taps (Figure 29-8) are usually used.The flutes create edges on the thread profile and provide space for the chips and the passage of fluid. Such taps are made of either carbon or high-speed steel and are now ro coated withTiN..The flutes can be either straight, helical, or spual. 29-8) have square shanks and are usually made in sets Hand taps (F~gure The taper tup has a tapered end that will enter the hole a sufficient distance to h the tap. In addit~on, the threads increase gradually to full depth, and therefore of tap requires less torque to use. However, only a through-hole can be thre pletely wlth a taper tap because it cuts to full depth only behind the tap A b h d hole can be threaded to the bottom using three types of taps insuccesslo the taper tap has the thread started in proper alignment, aplug tap, which has on tapered threads to provide gradual cutting of the threads to depth, is used threads as deep into the hole as its shape will p e n t . A bottoming tap,having n threads, is used to finish the few remaining threads at the bottom of the h depth. Obviously, producing threads to the full depth of a blind hole is tim and it also frequently results in broken taps and defectlve workpieces. Suc tions usually can be avoided if designers will give reasonable thought to the

rChamfer angle Chamfer depth Chamfer rellef

Taps operate under very severe conditions because of the heavy frictron (high torque) mnvolved and the difficulty of c h ~ p removal. Also, taps are relatively fragllle. Spira1;fluted taps (Figure 29-9) prov~de better removal of chips from a hole, part~cularly In tapplng materials that produce long, curling chips. They are also helpful in tapping holes where the cutting action is interrupted by slots or keyways.The splrul point cuts the thread w ~ t h a shearing action that pushes the ch~ps ahead of the tap so that they do not mterfere wlth the cutting action and the flow of cultlug fluid ~ n t o the hole.

COLLAPSING TAPS Collapsing taps are s~milar to self-opening d ~ heads e m that the cutting elements collapse inward automatically when the thread is completed. Thls permits w~thdrawing the tap from the workpiece without the necessity of unscrewing it Gom the thread.They can either be self-settmg, for use on automatic machines, or require manual setting for each cycle. Figure 29-10 shows some of the types available.

HOLE PREPARATION Dr~Uing is the most common method of preparing holes for tapping, and when close control over hole size is requ~red, reaming may also be necessary.The dr~ll slze determrnesthe final thread contour and the drrlling torque.Unless otherwsespecified, the tap drill size for most materials should produce approximately 75% thread, that is,75% of full threaddepth.

Plug tap

nuarn11 C " M h

- . - - "-...
1
Height of thread2

for singlet)lead 2 for dou bk3thrt Pitch hrea

/

Length of shank

Thread length amfe ointil

Center hole
Back taper A n bunt pdch diam at "BUis smaller rha p~tch d~am st "A"

Half anglaofthrea

of thread

, plugI (pj, and bottoming taper (t3' (b) taps, which are used serially i n threadrr~gholces. (Courtesy of TRWGreenfieId Tap & Die.)

FIGURE 29-8 Terminology for a plug taF,with photography of

Taper

FIGURE 29-9 (Left to right)

Spiral-fluted tap; spiral-point tap; spiral-point tap cutting chips; fluteless bottoming tap and fluteless plug tap for coldforming internal threads; cross section of fluteless forming tap. (Courtesy of JRW-Greenfield Tap & Die.)

Cross section

Lobe

SECTION 29.3 Internal Thread Cutting-Tapping

801

Solid adjustable

ALT collapsible

2LLS receding collapsible

FIGURE 29-10 Solid-

adjustable and collaps~ble taps. (Courtesy of Teledyne Londis

Machme.)
LL receding

collapsible

ELT collaps~ble

LBLM

receding collaps~ble

TAPPING I N MACHINE TOOLS Solid taps also are used in tapping operations on machine tools, such as lathes, drill presses, and special tappingmachines. In tapping on a drill press, a tapping attachment is often used. These devices rotate the tap slowly when the drill press spindle is fed downward against the work. When the tapping is completed and the spindle raised, the tap is automatically driven in the reverse direction at a higher speed to reduce the time required to back the tap out of the hole. Some modern machine tools provide for extremely fast spindle reversal for backing taps out of holes. When solid taps are used on a screw machine or turret, the tap is preventedfrom turning while it is being fed into the work. As the tap reaches the end of the hole, the tap is free to rotate with the work. The work is then reversed and the tap, again prevented from rotating, is backed out of the hole. The machine should have adequate power, rigidity, speed and feed ranges, cutting fluid supply, and positive drive action. Chucks, tap holders, and collets should be checked regularly for signs of wear or damage. Accurate alignment of the tap holder, machine spindle, and workpiece is vital to avoid broken taps or bell-mouthed, tapered, or oversized holes. TAPPING CUTTING TIME The equation to calculate the cutting time for tapping 1s (approximately)

where

N = spindlerpm T, = cutting time (min) D = tap dlameter (in.) L = depth of tapped hole or length of cut (in ) N = number of threads per inch (tpi)(feed rate) V = cutting speed (sfpm) A, = allowance to start the tap (min)
AR = allowance to wlthdraw the tap (min)

SPECIAL THREADING AND TAPPING MACHINES Spewal machimes are available for production threading and tapping. Threading chlnes urually have one or more splndles on which a self-openmg dle head is mount with suitable means for clamping and feeding the workpiece Special tapping mach using selfcollapsing taps substituted for the threading dies are also available. More c monly, tapping machines resemble drill presses, modifled to provlde spindle feedsb upward and downward, with the speed and feed more rapid on the upward motion. COMMON TAPPING PROBLEMS Tap overloadmg 1s often caused by poor lubrication,lands that are too wlde, chipspac in the flutes, or tap wear. Surface roughness in the threads has many causes. A nega grind on the heel will prevent the tap from tearing the threads when baclung out. When a tap loses speed or needs more power,it generally indicates that the dull (or Improperly ground) or the chips arc packed in flutes (loaded).The flute be too shallow or the lands too deep. When tapping soft ducttle metals, loadmg can ally be overcome by polishing the tap before usage. Improper hole size due to drill wear increases the percentage of threads bemg Dull tools can aka produce a rough finish or workharden the hole surface and caus tap to dull morc quickly. Check to see that the axls of the hole and tap are aligned tap cuts when backing out, check to see if the hole is oversix TAPPING HIGH-STRENGTH MATERIALS H~gh-strength, thermal-resistant materials, sometimes called "exotics," cause sp problems in tapping. A varlety of materials are classified as exotics stainless steel, c~pltatlon-hardenedstainless steel, high-alloy steels, iron-based auperalloys, tit Inconcel. Hastelloy, Monel, and Waspalloy. Their most important attribute is their strength-to-welght ratio. Each material pre~ents different problems to effic~ent tapplng, but they all certain similaritles.Toughness and general abrasiveness lop the list It IS also dlffic Impart a good surface finish L o exotics: heat tends to localize in the shear zone,an otics tend lo workharden and grab the tool. A tap's chamfer and flrst tull thread do virtually all the cutting. m e rema ground threads serve merely as chasers. Because ol this, taps to thread exotics ar creaslngly manufactured with short threads and reduced necksThey diinmlsh pr caused by materlal closure and provide more space for coolant and chip ejectio When tapping exotles,the largest tap core dlameter possible should be apphe ting 75 or 65% threads places less stress on the taps, lengthening tool llfe and red breakage To cut threads in exotic alloys successfully, taps must comhne geom speclfically tailored for those ~naterlals and be made of premium tool steels sub] to precisely controlled heat-treatment processes Stainlesb steel is known to workharden and to have slow heat-dlssipatlon tenstics;stainless steel requires a tap geometry with a positive 6' to 9" workhardening and redunng torque. Gnndlng an appropriate eccentrtc thread and taper rehef onto the tap will reduce friction. A surface treatment That 1s. the tool should be made from high-vanadium, high-cobalt tool surface treatment so that coolant adheres to it. Stainless steel generates long ch requires a tap with a 38' helix angle and adequate flute depth to promote chip e tton. Proper hook and rad~al relief guarantee accurate thread-hole slze and long to When tapping a titanium alloy, the matenal's tendency to concentrate heat in contact area must be considered. Concentrated heat often leads to excessive cuttin wear Tltanium generates average-to-short ch~ps, is abrasive, and 1s prone to chip w and high frictlon.These character~st~cs degade tap performance and shorten tool Taps for threading titanium are constructed of premium tool steel, nitr~ded ness.Titanium nitride (TIN) coatings cannot be used because they react chemica the workpiece materia1,causing rapid tap failure. For tapplng through-holes, a 3' to and short thread design with h ~ g h eccentrrc relief will promote efficient chip evac Nickel-based Inconel, Monel, Waspalloy, and Hastelloy present severe t problems Among their machining characteristics are toughness, workhardenin

retention, and built-up-edge (BUE).Taps designed to thread these materials need tremendous stab~hty and a strong cross-sectional construction. The most popular tap materials for mckel alloys are high-vanadium, h~gh-cobalt tool steel or powdered metal (PM) tool steel. Tappmg blind holes ~nthese alloys requrres a 3" to 5" rake to shear and deflect the cuttmg forces downward toward the tap's root,its strongest area.A 26" helut angle promotes chip evacuation, and a nitride or TiN coating reduces friction and too1 wear. Because of nickel alloys' toughness, taps to cut them Should have the longest taper possible. Thrs allows the cutting edges to progressively gain thread hetght before the Grst full thread beglns its cut, distributing the load over a wider area. Spiral-pointed, straight-flute taps have a four- to five-thread taper. For tapping blind holes, the first two or three t h r e a d e m o r e if possible-should be tapered.

CUTTING FLUID FOR TAPPING

Cutting flu~ds should be kept as clean as possible and should be supplied in coplous quantities to reduce heat and frict~on and to aid in chip removal. Lang tap life has been reported to result from routing high-pressure coolants through the tap to flush out the chips and cool the cutting edges. Recommended cutting fluids are listed in Table 29-1.

1 29.4 THREAD MILLING

H~ghly accurate threads, particularly in larger sizes, are often form milled. Either a singleor a mult~ple-form cutter may be used A single-form cutter having a single annular row of teeth is Wed at an angle equal to the helix angle of the thread and is fed inward radially to full depth while the work u s stanonary The workpiece then is rotated slowly, and the cutter simultaneously is moved longitudinally,parallel with the axis of the work (or vice versa), by means of a lead screw, untll the thread is completed. The thread can be completed m a single cut, or roughing and finish cuts can be used. This process is used primarily for large-lead or multiple-lead threads. Some threads can be milled more quickly by using a multiple-form cutter having multiple rows of teeth set perpendicular to the cutter axis (the rows having no lead).Thc cutter must be slightly longer than the thread to be cut.It is set parallel with the axis of the workpiece and fed inward to full-thread depth while the work isstationary The work then 1s rotated slowly for a little over one revolution, and the rotating cutter is simultaneously moved longitudinally with respect to the workp~ece (or vice versa) acwrding to

Work Matenal
Aluminum Brass Naval brass Manganese bronze Phosphor bronzr C ~ P P ~ Iron, cast malleable Magnes~um Monel metal stceis Up to 0 25 carbon Freemachtn~ng 0 3 0 4 60 carbon,annealed 0 30-0 60 carbon,heal treated Tool, hgi-carbon, HSS Stainless Titan~um Zinc d ~ casttngs e

Cuttlng Flu~d
Kerosene add lard o ~ lkerosene , and hght-base 011 Soluble oil or light-base oil Mineral oil with lard or light-base oil Mineral oil with lard or light-base oil Mineral oil with lard or light-base oil Mineral oil with lard or light-base oil Dry or soluble oil Soluble oil or sulfur-base oil Light-base oil diluted with kerosene Sulk-base oil Sulfur-base or soluble oil Sulfur-base or soluble oil Chlorinated sulfur-base oil Sulfur-base oil Chlorinated sulfur-base oil Chlorinated sulfur-base oil Chlorinated sulfur-base oil Kerosene and Lard oil

804

C H A ~ 29 R Thread and Gear Manufacturing

the thread lead. When the work has revolved one revolution, the thread is complete process cannot be used on threads having a heltx angle greater than about 3", h a clearance between the sides ot the threads and the cutter depends on the cutter diamr belng substanbally less than ihat of the workprece,Thus, although the process is rap] use is restricted to threads of substantial diameter and not more than about2 in lor As shown m F~gure 29-11, advances in CNC computer controls have led to th mllllng on three-axis machines. Today's CNC can helically interpolate the axial controll~ng the thread p~tch with c~rcular feed controlling the circumference of the th External threading
n

[J Right hand (climbmrlltng) B Left hand iconvent~onal miil~ng)

12 Left-hand (climbm ~ l l m g ) P# Right hand (conventional millmg)

Internal threading

I 3Left-hand (convent~onal milling)

CI Right hand (conventional mtlling) rn Left-hand (climb milling)

rn R~ght-hand(climb millcng)

Thredu mil irlg art irpas.t~ 10:.19 celters i v t n multiroorn 'ndc~abfe-~,arui.!t.-<c~sert c.jrrnrs nas intrud..ceu atr.,~t laenly y e J s ago. Th,: r:..tlrr <:all pr<rn...-3 a f~nisnca tnrcao n one helical pass.

FIGURE 29-11 Thread milling on a three-axis NC machine can produce a complete thread in a single feed revolution (Fred Mason, American Mochinrst,
November, 7 988 1

SECTION 29.6

Thread Rolling

805

The cutter has teeth shaped I i e the desired thread form.The cutter rotates at high speeds while its axis slowly moves around the partin a planetary arc just over 360a.The cutter advances axiaIlg a distance equal to one p~fch to generate the helical path.Thread milling has advantages in d~arneters and coucenlriciover 1.5 i n , including better surface hn~sh ty and Ure ability to prtlduce nght- or left-hand threads w ~ t h the same tool. Thrilling (dnlling plus threading) produces threaded holes by cornbmmg short hole drilling w~th thread milling using a conlbrnatlon tool with a drill point and a thread mill body The details of the process are shown in figure 29-12 and can be done on any CNC machining center. Compared to tapping, the process elminates two tools, two tool holders, and two tool change cycles as the single tool combmes the drill ream and tap functions into one 1ool.Threaded-hole depths are limited to about three hole diameters.

DING
Grinding can produce very accurate threads, and it also permits threads to be produced in hardened materials. Three basic methods are used. Center-type grindz~g wrth wral feed is the most common method, be~ngsimilar to cutting a thread on a1athe.A shaped grind~ng wheel replaces the s~ngle-point tool. Usually,a single-ribbed grinding wheel is employed, but multiple-ribbed wl~eels are used occasionally. The gr~nding wheels are shaped by special d~amond dressers or by crush dressing and must be inclined to the helix angle of the thread. Wheel speeds are in the high range. Several passes are usually required to complete the thread. Center-type infeed thread gnrzdfng is srmilar to multiple-form milling in that a mulas the length of the desired thread, is used The wheel is fed hple-ribbed wheel, as w~de inward radially to full thread depth, and the thread blank is then turned through about l1iZ turns as the grinding wheel is fed axially a little more than the width of one thread. Centerlessthread grind~ng is used for making headless strtscrews.The blanks are hopper fed to the regulating wheel, which causes them to traverse the grinding wheel face,from which they emerge in completed form. Production rates of 60 to 70 screws of If2-in length per minute are possible.

Thread rolliirg is used to produce threads in subsrantial quantities. This is aeold-forrning process operation in whieh the threads are formed by rolling a thread blank between hardened dles that cause the metal to flow radially inlo the desired shape. Because no metal is removed in the form of chips, lessmaterial is required, resulting

S H A 29 ~

Thread and Gear Manufacturing in substantial savings In addition, because of cold working, the threads have great strength than cut threads, and a smoother, harder, and more wear-res~stant surface 0btained.h addition, the process i s fast, with production rates of one per second b common. The quality of cold-rolled products is cons~stently good. Ch~pless operatlo are cleaner and there is a savmgs in mater~al(lS% to 20% savings in blank stock welg is typical). Thread rolling is done by four basic methods. Ihe simplest o and onemovable flat rolhng die (Figure 29-13). After the blank is place stationary die, movement of the moving die causes the blank to be ro dies and the metal in the blank is displaced to form the threads.As th across the die parallel with its longitudinal axis. Prior to the end of the stroke of the mg die, the blank rolls off the end of the stationary che,its thread being completed. One obvious characteristic of a rolled thread is that its major diameter alway greater than the diameter of the blank. When an accurate class of fit is desired, the arneter of the blank is made about 0.002 in. larger than the thread-pach diameter. is desired to have the body of a bolt larger than the outside diameter of the rolled thre the blank for the thread is made smaller than the body. Thread rolling can be done wlth cylindrical dies. Figure 29-13 roll method commonly employed on turret lathes and screw machines.Two var~ati are used. In one, the rolls are retracted while the blank is placed in position. The r then move inward radially,while rotating, to form the thread. More commonly the th rolls are contained ln a self-opening die head simllar to the conventional type used cutting external threads. The die head is fed onto the blank longitudinally and fo the thread progressively as the blank rotates. With this procedure, as in the case of threads, the innermost 11i2to 2 threads are not formed to full depth because of the p gressive action of the rollers. The two-roll method is commonly employed for automatically producing 1 quantities of externally threaded parts up to 6 in. m diameter and 20 in. m length. planetary-type machine 1s for mass production of rolled threads on diameten up to Not only is thread rollmg very economical, the threads are exce strength. The cold working contributes to increased strength, particul root areas.There is less hkelihood of surface defects (produced by act as stress raisers. Large numbers of threads are rolled on thm, tubular pro nal and internal rolls are used. The threads on electric lamp bas amples of this type of thread.

CHIPLESS TAPPING Unfortunately, most internal threads cannot be made by rolling; there is insuffi space within the hole to permt the required rolls to be arranged and supported, the required force8 are too high. Howeveqrnany internal threads,up to about l'1,i diameter, are coldformed in holes in ductile metals by means tap and its special cross section are shown in Hgure 29-9.The the same as in rolling external threads. Because of the forming inv tion, the torque required is about double that for cutting taps. must be controlled carefully to obtain full thread depth without excessive However, fluteless taps produce somewhat better accuracy th lrfe is often greater than that of high-speed-steel (HSS) machine taps. A lubr. should be used, water-soluble oils being quite effective.Fluteless taps are esp able for forming threads m dead-end holes because no chips are produced, m both plug and bottoming types. MACHINING VERSUS ROLLING THREADS Threads aremachmed or cut when full thread depth 1s needed (more th essary) for short production runs, when the blanks are not very accura ity to the shoulder in end threadmg 1s needed,for tapered threads,or w material is not adaptable for rolling.

Part transfer D o i n 4 Full form starting

------------.----

--- - - - -<--CZi

Moving die

Stationary die Method of thread rolling, machine using flat die

(8)

Full form finish reliefs on both sides (b) Stationary die design (top view)

Reciprocating flat die

0.403
engaging
(C)

penetration Action of die forming thread

Fully formed thread

1041 Steel

id) Blank and finished part ion rt.) at 100 per min

cylindrical d ie

Blank

Moving cylindrical die Stationary-segment die (some machines use more than one segment) 1

Rotary die

:=---- . . - -- -. - Centerline of unskewed die

Centerline of . . skewed die

: Centerline of .. . DlanK

/ Blank

Skewed-axis dies (el Two-die cylindrical machine

(g) Planetary-type machine

Pressu Three cylindical dies in rolling position

\

cylindrical dies in retracted position

(f) Three-die cylindrical machine

1-13 Roll forming threads using flat die thread rolling process shown in (a) and (b). The threads forming actio :) and the product in (d). Three variations of cylindrical rolling are shown in (e), (0, and (g).

808

CHAPTER 29

Thread and Gear Manufacturing

circular pitche

rTboth

thickness

FIGURE 29-14

Gear-tooth

nomenclature.

Base circle Pitch circle

--f--

As with threads, we need to have an ~ntroductory understanding of the product we can understand the process.

GEAR THEORY AND TERMINOLOGY Basically, gears are modifications of wheels, with gear teeth added to prevents and to ensure that their relative motions are constant. However, it should bf that the relative surface velocities of the wheels (and shafts) are determined diameters of the wheels. Although wooden teeth or pegs were attached to c make gears in ancient times, the teeth of modern gears are produced by ma1 or forming teeth on the outer portion of the wheel.Thepitch circle (Figures 29, 29-15) corresponds to the diameter of the wheel.Thus the angular velocity of a determined by the diameter of this imaginary pitch circle.Al1 design calculations I to gear performance are based on the pitch-circle diameter or, more simply, tt diameter (PD).
k2teyf gear 1

Addendum circle

Clearance'

FIGURE 29-15 Tangent pitch circles between two gears produce a p~tchpoint.

I
I

/ /

SECTION 29.7

Gear Maktng

801)

string

CURE6 I ' * Method Of enefating afl involute curve by winding a str~ng from a

For two gears to operate properly, their pitch circles must be titngential to each other. The point at whiih the two pitch nrcles are tangent, at which they intersect the centerline connecting thex centers of rotation, is called the pitch pomL The common normal at the point of contact of mating teeth must pass through the pitch point. ?his wndition IS illustrated in Figure 29-15. To provide uniform pressure and mutian and to minimize fridon and wear, g e m are designed to have rolling motion belween mating teeth rather than Sliding motion To achieve th& condition, most gears utilize a tooth form that is based on an ~nuohte urrve.This is the curve that 1s generated by a pomt on a straight line when the line rolls around a base circle. A somewhat sunpler method of developing an involute curve is that shown in Figure 29-16. Lly unwinding a tautly held string from a point on the base circle, point A, an involute enme is generated. There are three other reasons for using the involute form for gear teeth.First, such a tooth form provides the desued pure rolling action. Second, even if apair of involute gears is operated with the distance between the centers slightly too lafge or too small, the common normal at the point of contact between mating teeth will always pass through the pitch point. Obviously, the theoretical pitch circles msuchcase6 will be increased or decreased slightly. Thad, the lzne of action or path 07 contact-that is, the locus of the points of contact of matmg teeth--$ a stra~ght line that passes through the pitch point and is tangent to the base circles of the two gears. Cutting an involute shape in gear b l a h can he done by simple form cutting (i.e., milling the shape into the workpiece) or by generating, Generating involves relative motion between the workpiece and the cutting tool. True involute tooth form can be produced by a cutting tool that has s ~ q h t - s i d e d teeth. This permits a very accurate involute tooth profile to beobtained through thense of a simple and easily made cutting tooL.The stra~ght-sided teeih are p e n a rolling motion relative tothe workpiece to ereafe the curved gear-tooth face, that is, the involute shape. The basic size of gear teeth may be expressed in two ways.Tbe common practice, especially in the Ontted States and England,is to express the dimensions as a functim of the dihmetml pitch (DP). DP i s the number of teeth (N) per unit of pitch diameter (PD): thus (PD) = NIPD. DimensianaIly,DP involves inches in the English system and millimerersin the S1 system, and it 1s a measure a f tooth 5ze.Metricgears use the module system (&'),defined as the pitch dzamster d~videdb.y the number of teeth, or M = PDIN. It thus is the reciprocal of diametraI pit& andis expressed in millimetels. Any two gears having the same diametralpitch or module willmesh properly if they are mounted so as to have the correct distances and relationship7lle important tooth elements can be specrfied in terms of the diamehal pitch or the module and are as follows:

1. Addendum: the r a b l distance fram the pitch circle to the outside &meter.
addendum = -znches DP 2. Dedwdum. the radial dista~lce fium ihe pitch circle to the root circle. IT I s equal to the addendum plus the clearance, wh~ch is provided to prevent the outer corner of a tooth from touching against the bottom of the tooth space. 3. C6rctl6arpzrck the dtstance betw~en corresponding points of adjacent teeth, measured along the pitch chde: ddiametral pitch 4. Too$% thzkness: the thickness of a tooth, measured along the pitch circle.When tooth thickness and the correspondingtooth space are equal, no backlash exists in a paif of mating gears. 5. Face width. the length of the gear teeth in an axial plane. 6, Tooth face the mating surface between the pitch circle and the addendum circle. 7 . Toothflunk; the mating surface betweenthe pitch nrcle and the root circle. & Presslire m g l e the angle between n tangent to the tooth profile and a line perpendicular to the pitch surface.

Four shapes of involute gear teeth are used in the United States:

1. 14;' pressure angle, full depth (used most frequently)

2. 14;' pressure angle, composite (seldom used) 3. 20" pressure angle, full depth (seldom used) 4. 20" pressure angle, stub tooth (second most common)

In the 1 4 g full-depth system, the tooth profile outside the base circle is an invo curve. Inward from the base circle, the profile is a straight radial line that is joined the bottom land by a small fillet.With this system,the teeth of the basic rack have str sides.The composite system and the 20" full-depth system provide somewhat teeth. However, with the 20" full-depth system considerable undercutting occur dedendum area, therefore, stub teeth often are used. The addendum is short 20%,thus permittrng the dedendum to be shortened a similar amount.This resultsin strong teeth without undercutting Table 29-2 gives the formulas for computing th mensions of gear teeth rn the 141120 full-depth and 20' stub-tooth systems,

PHYSICAL REQUIREMENTS OF GEARS A consideration of gear theory leads to five requirements that must be met inord gears to operate satisfactorily:

1. The actual tooth profile must be the same as the theoretical profile.
2. Tooth spacing must be unlform and correct. 3. The actual and theoretical pitch circles must be coincident and be concentric

the axis of rotation of the gear. 4. The face and flank surfaces must be smooth and sufficiently hard to resist wear prevent nolsy operation. 5. Adequate shafts and bearings must he provlded so that deslred center-to-cen tances are retained under operabond loads. The first four of these requirements are determmed by the material selecti manufacturing process.The various methods of manufacture that are used repres tempts to meet these reqlurements to varylng degrees wlth miulmum cost, and th fectiveness must be measured in terms of the extent to which the resulting gears em these requirements. Before looklng at the ways to manufacturegears,let's look at some examples of g

14'12" Full Depth

Pltch d~ameter (PD) Addendum Dedendum Outs~de dlarneter Clearance Tooth th~ckness
-

20,Stub Taoth
N DP 08 DP
-

N DP

1 DP 1157 DP N+2 DP 0 157 DP 1508 DP

1 DP N+16 DP 0 2 DP

1.508 DP

DP = Numher of teeth (N) per unli of pltch diamcter (PD)

SECTION 29.8 Gear Types

811

The more common types of gears are shown in Figure 29-17. Spur gears have straight teeth and are used t o connect parallel shafts. They are the most easily made and the cheapest of all types. The teeth on helical gears lie along a helix, the angle of the helix being the angle between the helix and a pitch cylinder element parallel with the gear shaft. Helical gears can connect either parallel or nonparallel nonintersecting shafts Such gears are stronger and quieter than spur gears because the contact between mating teeth increases more gradually and more teeth are in contact at a given time. Although they usually are slightly more expensive to make than spur gears, they can be manufactured in several ways and are produced in large numbers. Helical gears have one disadvantage. When they are in use, a side thrnst is created that must be absorbed in the bearings. Herringbone gears neutralize this side thrust by having,in effect,two helical-gear halves, one having a right-hand and the other a lefthand helix. The continuous herringbone type is rather difficult to machine but is very str0ng.A modified heningbone type is made by machining a groove, or gap, around the gear blank where the two sets of teeth would come together. This provides a runout space for the cutting tool in making each set of teeth.

Different types of gears

Spur gear

Helical gears

Herringbone gears

Straight bevel gears

Skew bevels

Splral bevels

Hypoid gears

Spiral gears

FIGURE 29-17 There are

many different types of gears for transmitting motion, resulting in a wide variety of gear-making processes.

Worm

gears

Internal gear

812

CHAPTER 29

Thread and Gear Manufacturing

angle, it cannot be driven by the mating worm gear This principle

IS

frequent1

so that more teeth are engaged at a yven trme and the engagement 1s progressi poid bevel gears alro have a curved-tooth shape but are des~gned to operate wi intersecting axes. Rear-dnve automobiles used hypoid gears ln the rear axle so
Zero1 bevel gears have teeth that are clrcular arcs, providing somewhat sh-onge than can be obtamedin a comparable btraight-tooth gear.They are not used exte ly. When a pair of bevel gears are the same size and have their shafts at r~ght a they are termed mzter gears. A crown gear is a special form of bevel gear havmg a 180" cone apex angle. fect, it is a disk with the teeth on the side of the disk. It may also be thought of as a that has been bent into a circle so that its teeth h e m a plane. The teeth may he str or curved. On straight-tooth crown gears the teeth are rad~al. Crown gears arese used, but they have the important quality that they willmesh properly with a bevel of any cone angle,provlded that the bevel gear has the same tooth form and diam pitch This Important principle is incorporated in the design and operation of two Important types of gear-generating machines that will be d~scussed later. Most gears are of the external type, the teeth forming the outer periphery

center of the gear.

1 29.9 GEAR MANUFACTURING

Whether producedmlarge or small quantit~es,mceU$ or injob shop batches,the sequ of processes for gear manufacturing requues four sets of operat~ons. See Flgure 29-

1. Blanking (turning) , 2. Gear cnttrng (hobbing and shaving) 3. Heat treatment 4. Grinding
Blanktngrefers to the ~nitial formmg or machining operations that prodnc finished part ready for gear cutting, starting from a piece of raw material. Tu

Turmm

Dear
cutrlng

CiUenrh~ng Grhndtng

Tooth proflle

ial Manlnacturlng procass

Tooth-face measurement

Tamh runout

The sctlon of a hoh as its teeth progress through and outthe tooth space i n

a gear.

Relatlonshlp afthe hob and the gear blank

171

machrntng a spur gear

Shav~no outterandgesr
fcl Shaving

{hi Hobb~ng

FIGURE 29-18 The typ~cal gear-making procers (for very accurate gears) ~nvolves both hobbing and shaving followed by grind~ng after heat treatcng.

The gear-cutting or -machining operations can be divided ~ n t o operations executed prior to heat treatment, when the materlal is still soft and easily machinable, and after heat treatment, performed on parts that have acqu~red high hardness and srrength. Heat treatment glves the material the strength and durability to withstand high loads and wear but resuIts in a reduction in dimensional and geometrical accuracy The metallurgical transformations that occur during hardening, quenehmng, and tempering cause a general qualify deteri~ration in the gears. Therefore, precision grinding operations are used on external and internal bearing diameters, critical length dimensions, anu fine surface finishes after heat treatment. Cyhndrlcal gr~nders~angle-head grinders, infernal grinders,and surfade grinders are commonly used. Gears are made in very large numbers by cold-roll forming; in addition, significant quantities are made by extrusion, by blanking, by wsting, and some by powder metallurgy and by a farglngprocess. However, it is only by machining that all types of gears can be made in all sms, and although roll-formed gears canhe made with accuracy sufficient for most appheations, men for automobile transmissions, machinrng still is unsurpassed for gears that must have very high accuracy,Also, roll fommg can be used only on dnctlle metals.

29.10 MACHINING OF GEARS

FORM MILLING
Form cuttmg or f o r mrnilli-rg an a horizontal milling machine is illuslrated in Figure 29-19. The multiple-tooth form cutter has the same form as the space between adjacent teeth-The tool is fed radially toward the center of the gear blank to the desired tooth depth, then across the tooth face to obtain the required tooth width. When one tooth space has been completed, the tool is withdrawn, the gear blank is indexed using a dividing head, and the next tooth space is cut. In machining gears by the form-cuttmgprocess, the form cutter 1s mounted on the mach~ne arbor, and the gear blank is mounted on a mandrel held between the centers of some type of ~ndexing device. Basically, form cutting is a simple and flexible methad of machining gears. The equipment and cutters required are relatively

les L o n ng, a*IS.

SECTION 29.10

Machining of Gears

815

IRE 29-20 Bl~nd gear

e broach~ng machine for uung internal gears The h~ne has automated p~ckplace arms for loadlunload. riesy of Apex Broach and h~ne Company)

BROACHING Brouchrng is another way to form cut teeth. All the tooth spaces are cut simultaneously, and the tooth is formed progress~vely.The circular table in Figure 29-20 holds 10 sets of progressive tooImg.The table rotates, moving one set of tooltng a1 a hme under two workpieces The arms load and unload a set of parts every 15 secondr,so the cycle time 1s very quick. Excellent gears can be made by broachzng. However, a separate broach must be provided for each size of gear. The tooling tends to be expensive, restr~cting this method to large volumes. GEAR GENERATING Most high-quality gears that are made by machming are made by the generufrngpr6cesr. This process is based oil the princ~ple that any two involute gears, or any gear and a rack, of the same diametral pitch w~ll mesh together propcrly. Utilizing this principle, one of the gears (or the rack) is made into a cutter by proper skarpening It can be used to cut into a mating gear blank and thus generate teeth on the blank. The two principal methods for gear generating are shaplng and hnhbing. SHAPING To carry out the shaping process, the cutter and the gear blank must be attached rigidly to thsir rrepectlve shafts, and the two shafts must be interconnected by suitable gearmgso that the cutter and the blank rotate pos~tively w~th respect to each other and have the same pitch-hne velocities. To start cuttlng the gear, the cutter is reciprocated vertically and is Ced radially into the blank between successive btrokus When the desired tooth depth has been obta~ned,the cutter and blank are then slightly indexed after each cutting slroke.The resulting generatmg actlon is indicated schematically in Figure 29-21a and shown in the cutting of an actual gear tooth in F~gugure 29-21b. Figure 29-22 shows a mach~ne called a gear shaper. Gear vhupers generare gears by a reciprocating tool motlon. The gear blank is mounted on the rotating table (or vertical spindle) and the cutter on the end of a vertical,reuprocating spindle.The splndle and the table are connected by means oi gears so that the curter and gear blank revolve with the same pitch-hne velocity. Cutting occurs on the downstroke (sometimes on the upstroke). At the end of each ~ u t t i n g stroke, the spindle carrying the blank retracts slightly to prov~de clearance between work and roo1 on the return stroke. The conventional cutter for gear shaping 1s made from h~gh-speed steel, which can be coated to unprove wear life w t h a superhard laytr ot tltan~um nitr~de (TIN)nwng the physical vapor deposition (PVD) process. Recently, new throwaway d~ak-shaped insert regrinding and recoating optools have been developed for gear shaping, el~m~nating erations on the convent~onal tools. Regrind~ng the conventional cutter requlres two adjustments (resetting opetations) on the machine tool. Because the throwaway blades are all slzed the same, machine resetting after cutting tool changeover is eliminated.

Thread and Gear Manufacturing Gear shaper cutter

Base circle Gear being generated

(a)

Shape of gear shaper cutter chip

(Top) Generating action of a Fellows gear shaper cutter. (Bottom) Series of photographs showing various stages in generating one tooth in a gear by means of a gear shaper, action taking place from right to left, corresponding to the diagram above. One tooth the cutter was painted white. (Courtesy of Fellows Gear Shaper Company.)
FIGURE 29-21

Either straight- or helical-tooth gears can he cut on gear shapers. To cut heli

gcar is removed and a new blank is put in place o n the tenth unit. Planetary gearsh holding six gear blanks move in planetary motion about a large, central gear cutte . cutter has n o teeth in one portion to provide a space where the gear can be removed a new blank placed on the empty spindle.

sometimes called dotters. Usually, the ram is pivoted near the top so that it can be s outward from the column through an arc of about 10".

SECTION 29.10

Machin~ng of Gears

81

Side clearance
+ i

on X-Y

FIGURE 29-22 This machlne tool 1s a gear shaper. The blank is rotatlng wh~le the cutter is reciprocating vertically, as shown in the inset. The tool is very complex and is shown in detail below. (Cdurtesy Fellows Gear Shoper Company.)

Thread arid Gear Manufactur~ng Because one circular and two straight-lme motions and feeds are available, cal shapers are very versatile tools and thus find considerable usein one-of-a-kind ufacturmg. Not only can vertical and lncbned flat surfaces be machined, but exte and internal cylindrical surfaces can be generated by circular feedmg of the table b strokes This may be cheaper than turning or bormg for very small lot sizes,A shaper can be used for generatmg gears or machinmg curved surfaces,interior surfa and arcs by using a stationary tool and rotating the workpiece. A keyseater is a s type of vertical shaper designed and used exclnsrvely for machining keyways on side of wheel and gear hubs. For machining contrnuous hernngbone gears, a Sykes generating machlne is used.

HOBBINC
Involute gear teeth could be generated by a cutter that has the form of a rack. Su cutter would be simple to make but has two major disadvantages. First, the cutter (o blank) would have to reciprocate, w ~ t h cutting occurring only dunng one stroke di tion. Second, because the rack would have to move longitudinally as the blank rota the rack would need to be very long (or the gear very small) or the two would not b mesh after a few teeth were cut A hob overcomes the preced~ng two d~fficulues. shown in Figure 29-23, a hob can be thought of, basically, as one long rack tooth that been wrapped around a cylinder in the form of a hehx and fluted at mtervals to pro a number of cutting edges. Relref is provlded behind each of the teeth.The cross sec of each tooth, normal to the helix, 1s the same as that of a rack tooth. (A hob can be thought of as a gashed worn gear.) The action of a bobbing machine cuttmg a spur gear is illustrated in hgure29-2 cut a spur gear, the axis of the hob must be set off from the normal to the rotati axis of the blank by the hehx angle of the hob. In cuttmg helical gears, the hob mu set over an additional amount equal to the helrx angle of the gear.The cuttmg of a by means of a hob is a continuous action.The hob and the blank are connected by pr gearing so that they rotate m mesh.To start cutting a gear, the rotating hob is fed in untll the proper setting for tooth depth is obtained.The hob is then fed in a drrec parallel with the axis of rotation of the blank. As the gear blank rotates, the teet generated and the feed of the hob across the face of the blank extends the teeth t desired tooth-face width. Hobbing is rapid and economical. More gears are cut by this process by any other. The process produces excellent gears and can also be used for sp and sprockets. Single-, double-, and triple-thread hobs are used. Multiple-thread increase the production rate but do not produce accuracy as high as single-thread Gear-hobbing machines are made in a wide range of slzes. Machines for cuttin curate large gears are frequently housed in temperature-controlled rooms, and the perature of the cuttlhg fluld is controlled to avold dimensional change due to variat In temperature.

COLD-ROLL FORMING The manufacture of gears by cald-roll form~nghas been highly developed and wi adopted in recent years. Currently, milhons of high-quality gears are produced an ly by this process, many of the gears in automobile transmssions are made thts wa indicated in figure 29-24, the process is basically the same as that by which screw th are roll formed, except that in most cases the teeth cannot he formed in a singler tion of the forming rolls; the rolls are fed inward gradually during several revolu Because of the metal flow that occurs, the top lands of roll-formed teeth a smooth and perfect in shape, a depressed line between two slight protrusions can o be seen, as shown encircled in Figure 29-24. However, because the top land play part in gear-tooth action, if there is sufficient clearance m the mating gear, this ca no dlfflculty Where desired, a llght turning cut 1s used to provide a smooth top land correct addendum diameter. The hardened forming rolls are very accurately made, and the roll-formed teeth usually have excellent accuracy. In addition, because the severe cold working

+Hob

Flute,

Keyway \

Pitch

-Cutting

lengthface width Hobs are rotary cutters w~th ened on therr faces. Nearly strarght tooth s~des generate ~nvolutes.
hei~cal rows of teeth sharp-

Tooth

cylinder

Tooth thickness Top rad~us

Lead angle* Pressure angle

face

Hub

Hob

Direct~on of hob feed

Pulley

LC)
FIGURE 29-23 (a) Three views of hob and gear blank, (small spur gear) (b) Hob. (c) Schematic of rnechan~sm of a hobb~ng machine, shown hobbing a large spur gear.

duces tooth faces that are much smoother and harder than those on the typical machined gear, they seldom require hardening or further finishing, and they have excellent wear characteristics. The process is rapld (up to 50 times faster than gear machinmg) and easily mechanized. No chips are made and thus less material is needed. Less skilled labor is required. Small gears are often made by rolling a length of shaft and then shcing off the individual gear blanks. Usually, soft steel is requlred and 4 to 5 in, in diameter IS about the limit, with fewer than six teeth, coarser than 12 diametral pitch, and no pressure angle less than 20".

CHAPTER 29

Thread and Gear Manufacturing

, ,

FIGURE 29-24 On the upper left, the method for cold form~ng gear teeth on a spline uslng three dies 1 s shown. (Upper ri Worm gear being roll formed by means of two rotating rolling tools with typical worm made by rolling. (Lower left) Gear IT by roll~ng. (Courtesy of Landla Machine Componp)

OTHER GEAR-MAKING PROCESSES Gears can be made by the various casting proceses. Sand-cast geurs have rough faces and are not accuratediiensionally.They are used only for services where th moves slowly and where noise and inaccuracy of motion can be tolerated. Gears by die casting aremore accurate and have fair surface fimsl~.They can be used to mit light loads at moderate speeds. Gears made by invesfmentmsrurg may be ac and have good surface characferisfics They Can be made of strong materials top their use in transmitting heavy loads. Inmany instances, gears that are to be finish machining are made from cast blanks, and in some larger gears the teeth can be c approximate shape to reduce the amount of maehmmg. Large quantities of gears are produced by Madking in a punch pressThe t of such gears usually does not exceed about 'Ila in. By shaving the gears after blanked, excellent accuracy can be achieved. Such gears are used in clocks, watche ters, and calculating machines. Fine blankmg is also used to produce thin,flat ge good quality. High-quality gears, both as to d~mensional accuracy and surface quality, made by the powdermetallurgy process. Usually, this process is employed only for sizes,ordinarily less than 1in. in diameter. However, larger and excellent gears are by Forging powder metallurgy preforms. This results in a product of much greate powder metallurgy me sity and strength than usually can be obtained by ord~nary

rn

and the resulting gears give excellent service at reduced cost Gears made by this process often require little or n o fimshing. Large quantities of plast~cgears are made by plasnc m o l d ~ n g . l quahty e of such gears is only fair, and they are suitable only for light loads. Accurate gears sultable for heavy loads are frequently machined out of lammated plastic materials. Whensuch gears are mated with metal years, they have the quality of reducingnoise. Quite accurate small gears can he made by the extrusion process. Typicallq: long lengths of rod. havlng the cross section of the desired gear, are extruded The ind~v~dual gears are then sawed from thls rod. Materials suitable for thls proccss are brass, bronze, aluminum alloys, magnesium alloys, and, occasionally, steel FIamemachlntng (oxyacetylene cuttmg) can be used to produce gears that are to be used for slow-moving applications wherein accuracy is not required. A few gears are made by the hot-roll-forming process In this process a cold mas ter gear is pressed Into a hot blank as the two are rolled together.

FINISHLING
To operate etficiently and have satisfactory hfe. gears nlust have accurate tooth profiles and the faces of the teeth must be smooth and hard These qualit~es are particularly ~mportant whi-n gears must operate quietly at hlgh speeds When they are produced rapidly and econom~cally by most processes except cold-roll forming, the tooth profiles may not be as accurate a3 des~red. and the surfaces arc somewhat rough and subject to rapid wear. Also, it IS difficult to cut gcar teeth in a hardened gear blank, and therefore economy dlctates that the gear be cut in a relatively soft blank and subsequently be heat treated to obtain greater hardness. Such heat treatment usually results in some slight distortion and surface roughness. Although most roll-formed gears have sufficiently accurate profdes, and the tooth faces are adequately smooth and frequently have sufficient hardness, this process is feasible only for relalively small gears. Consequently, a large proportion of high-quality gears dre given some type of fmishing operatton after they have recelved primary machining or after heat treatment. Most of these finishing operations can be done quite economically because only minute amounts of metal are removed and they are fast and often automatic. Gear shaving ISthe most commonly used method for finishing spur- and helical-gear teeth prior t~ hardemng.7lle gear is run, at h ~ y h speed,in contact with a shaving tool,usually of the type shown ID Figure 29-25 Such a tool 1s a very accurate, hardened, and ground gear that contains a number of peripheral serrations, thus forming a serles of sharp cutting edges on each tooth.The gear and shaving cutter are run in mesh with their

Kevwav

URE 29-25 Rotary gear .es 29 8ing cutter (see Flgu~

29-26).
Serrations

822

CHAPTER 29

Thread i d Gear Manufacturing

FIGURE 29-26 Three methods for gear finishing are (a) shaving using a special cutter-see Figure 29-25, (b) grinding using form grinding wheels, and (c) using form rolls, that is, roll forminglfinishing.

tooth faces, thus el~mlnahng any h ~ g h spots and producing a very accurate tooth Rack shaving cutters are sometimes used for shavlug small gears, the cutt procatlng lengthwise,causing the gear to roll along it, as it is moved sideways acr cutter and fed inward.

tance between the dres is reduced to cold work the surfaces and produce

If roll finishing 1s to be used, attention must be paid to the prerolled geometry. should consult the manufacturers of gear-rolling machines for specific recomm

Grindingis wed to obtain very accurate teeth on hardened gears.Two methods are used. One employs a formed grinding wheel that is trued to the exact form of a tooth by means of diamond? mountcd on :I special holder arid guid?tl hy a ltirgt. template.'llic orhcr method is invi~lute-penzratio~i ~rinding. which uses straight-sided grillding wheels that simul;ite m e side of a rack tooth.'lhc surl;tce of the %cartooth is groutld as the gcar rolls (and rcciprocntcs) p;~st the grinding whcc.1~. (irin~ling produces vzr) accurate gc:,rs. I~ut bcc;t~rx it i\ slu\v and experisivc. i t is used o n l y on rlic li~ghrst-quality. hnrdenc,d gcirs I,U[~~ call I Ia I l~ ~o he used for finishing hnrrlc1lc;t gciirs'Ihc gear tv bc finishcJ is run in contact nith une or marc ca5t iron lapping g23rs ulder a tlow c ~ i v c q fine abrasive i l . Beraube lapping ~etilovcs onlv a \t.r). smilll :amount (11' mctitl, it is usually etnin 0 ploycd on gc:lrs th;~thavr previously hccn shovdtl anJ hardcncd. This sombiniltlon of prorcsses produces gears that alo nc.irl! equal to ground geats in qu.ilit!, hut at sonsid~rably lower cost.

TION
A s wit3 an mantifastured products, gears avvt be h e c k e d to determine whether the resulting product msets the despecifications and requirements. Because of their irregular shape and the number of factors that must be me;lsured,inspeotion of gears is somswhat difficult. Among t h e facton t o be checked are d e linear tooth dimensions such as ihiskncus, spacing, depth, illid SO on; tooth prufilu: surfacc roughness: and noisc. Scvcrnl spccinl de! isoh. most of them :iutornatic or scrniauromafic. are uhcd for such insomtion C;e<rr.roorhvc,ni(v c.ulil~~.rs can IIC used tu mc:lsurc the thickness GI Se;tr tecth un thc pirch circlc (figure 33-27). CYC ge'u inspection 1n:tchincs (1igut.c 29-28) can qu~ckIy check aevcral factors, including variations in circular pitch, involute profile, lead,

FIGURE 29-28 A CNC gear

inspection machine hasX, Y, Z motions plus a r o w table.

C H A ~29 R Thread and Gear Manufacturing

tooth spacing, a n d variations in p r e s w r e angle. T h e gear is usually mounted betwe centers. T h e probe is moved t o t h e gear through X, Y, a n d Z translations. T h e gea rotated between measurements.The inset i n Figure 29-28 shows a typlcal display fo involute profile. Because noise level is important i n many applications, not only from point of noise pollution but also a s a n indicator of probable gear life,special e f o r noise measurement is quite widely used, sometimes integrated into mass-produc a s e m b i y lines

II Key Words
addendum bevel gear blanklng
bottoming tap broachlng circular prtch cold roll forming collapsible tap crown gear

dedendum dlametral pitch fluteless tap torm milhng gear finishing gear grinding gear hohh~ng gear rollmg gear shaping

gear teeth herringbone gear hob hohhmg involute curve pitch p~tch clrcle pltch diameter planetary gear

self-openmg die head spur gear taper tap thread cutting thread rolling thread tappmg worm gear

II Review Questions
How does the pltch diameter d~ffer from the major hameter for a standard screw thread7 For what types of threads are the pltch and the lead the same? What 1sthe hehx angle of a screw thread? Why are plpe threads tapered? What are three basic methods by which external threads are produced? Explain the meaning of '1," -20 UNC3A What is meant by the des~gnabon M20 x 2 5 6g6gY (What does the "x" meanq) What are two reasons fine-senes threads are berng used less now than in former years? In cutting a thread on a lathe, how 1s the pitch controlled? What a the function of a threadrng dlal on a lathe? In Chapter 22, what figure(s) showed a threadlug dial? What controls the lead of a thread when it is cut by a threading dle? What 1s the basic purpose of a self-opening dle head? What is the reason for uslng a taper tap hefore a plug tap m tapping a hole? What dlfflculuesare encountered if full threads are specified to the bottom of the dead-end hole? Can a fluteless tap be used for threading a hole in gray cast rron? Why or why not? What provlslons should a deslgner make so that a dead-end hole can he threaded? What is the major advantage of a spiral-point tap? Can a flutele$s tap be used for threadmg to the bottom of a dead-end hole? Why or why not? Is a desirable for a tappmgfluid to have lubricating quahtles? Why? How does thread rmlhng differ from thread turnlug? What are the advantages of mak~ng threads by grindmg? Why has thread rolling become the most commonly used method for making threads?

24. How can you detemlne whether a thread has been pro by rollmgrather than by cutting? 25 Why is the involute Corm used for gear teeth? 26. What is the diametral patch of a gear? 27. What 1s the relationship between the hametral pltch an module of a gear? 28 On a sketch of a gear,mdicate the pitch circle, addend cle, dedendum urcle, and c~rcular pltch. 29 What five requirements must he met for gears to oper ~sfactor~ly? Which of these are determined by the ma turing process? 30 What are the advantages of hehcal gearscomparedwlth gears? 31. What is the principal disadvantage ot hehcal gears? 32 What difficulty would he encountered m hobhing a he bone gear? 33. What is the only type of machine on whlch full-hem gears can be cut? 34. What modification in design n made to herrlngbone ge permit them to be cut by hobbmg'J 35. Why aren't more gears made by broaching? 36 What is the most rmportant property of a crown gear? 37 What are three bas~c processes for machining gears? 38. Which basic gear-mach~nlng process n utllued in a Fe gear shaper? 39. When a helical gear is machined on a m~lhng mach table lead screw and the universal dlvidmg head hav connected by a gear tram. Why? 40. Why 1s a gear-hohbmg machine much more productive gear shaper? 41. In gear shaplng and gear hobhmg. the tooth profiles ar erated.What does that mean? 42. What are the advantages of cold-roll formlng form gears?

. Why is cold-roll formingnot suitable for making gray cast iron

gears? . Under what conditions can shaving not be used for finishing gears? . What inherent property accrues from cold-roll forming of gears that may result in improved gear life?

46. Can lapping be used to finish cast iron gears? 47. What factors are usually checked in inspecting gears? 48. What are the basic methods for gear finishing?

B Problems
, Calculate the

cuttug tune needed to tap a 3i,-in.-drameter by 2-in.-deep hole usinga cutting speed of 30 sfpm.The tap has 10tpi , m e new manufacturing manager has recommended to you that chipless tapping be adopted for tappmg holes in the deep, dead-end holes on the 2-cylinder englue blocks that the cornpany makes. Chipless tappmg can run at twlce the speed of the convennonal tapping process, works well on deep holes, w~th longer tool lie.The and prov~des better quality and fin~sh tappmg process IS the bottleneck processin machining &I1the 10% of the blocks have to cast iron engine blwks. In add~tion, be scrapped due to broken taps What do you recammend? A hob that has a pitch d~amefer of76.2 mrn is used to cut a gear havrng SIX teeth. If a cutting speedof27.4 mlmin is used, what wlll be the rpm value of the gear blank? In Rgure 29-19 a form-millmg operation and cutters are shown. The gear is to be made from 4340 steel, R, SO prior to heat treat and final grmd Select the proper speeds and feeds for the job (the cutter 1s 4 m. m diameter) and compute the cutting tlme to mill this gear. Would you useup milling versus down rmllmg? A gear-broaching machne of the type shown in Figure 29-20 can do the gear m 15seconds (about 240 p a t s per hour). How many additronal gears per year are needed to cover the broach~ng tooling cost if each broaeh on the machine cost $250? Do you think the broach tool life is sufficientto handle

that number of parts? What about TIN coatmng the broaches (coat $10 00 per broach)? 6. Assume that 10,OUO spur gears, l1lS in. in diameter and 1 in. thlck, are to be made of 70-30 brass. What manufacturing method would you consider? 7. If only three gears described in problem 6 were to be made, what processwould you select? 8. K.C Stern,a design engineer for Boeing Commeru;ialBirplane Group,isdoingsome design work and has a questlou regarding the hole size that 1 s drilled before tapping Internal threads The book states, "A hole d~ameter slightly larger than the mlnor d~ameter of the thread must already exlst, made by drillingkeaming,bormg, or die castmg " Convent~onal wisdom would lead one to bekeve that a hole sl~ghtly smaller should exist befor8 tapping threads. me Mach~nery's Handbook seems to agree w i t h this. Most tap drill diameter listslzes listed are smaller than the minimum m~nor ed tor a particular thread size m MIL-S-Q08874B. K.C. has found at least one excepiion to this, but the difference was o w a couple of thousandths of anmch. K.C consultedlocalsawrces on this mattor, but no one would cammlt to the wrreekness or ~ncarrectness of the book. A number of people are scratchingthelr heads on thls one. Your mistante m this m i t e r IS greatly apprewated. Should the hole be shghtly smaller or slightly larger than the nunor diameter of the thread and why?

Bevel Gear for a Ridirzg Lawrz Mower

inch. The minimum material properties have been estimated to be a yield strength of 125 ksl and a surface hardness of Rockwell C 25.The maximum operating temperature should be less than 250F. Since the mower has a multi speed transmission, the transaxle gearing may be subject to sudden applications of load due to improper engagement of the clutch. The manuinches. While this does not translate to a Charpy test

lect an appropriate material from the alternatives dis. cussed in part 2. (Note: Casting alloys should be matched for the casting processes, high machinabiii loys would be favored for cutting applications, etc.). 4.Which of the above alternatives do you feel would be "best" solution to the problem? Why? For this system, outline the specific steps that would be necessary to duce the part from reasonable starting material. lnclu ments.

sion resistance need not be outstanding. A smooth surface finish is desirable (especially on the teeth), and the total production run has been placed between 25,000 and
1 .Based on the size and shape of the product, describe at

determine if it would be a feasible approach. If so, wo you recommendthat the forging be performed cold, ' warm, or hot? Would subsequent heat treatment or surface treatment be required? What concerns might with these processes?

least three reasonable ways in which the component could be produced. For each method, briefly discuss its relative pros and cons.

Figure CS 29

FUNDAMENTALS OF JOINING
.1 l p m o ~ u r r n o ~ ~ o
~SOLIDA~O PROCESSES N ' 0.2 CLASSIFICATION OF W ~ L D I N G AND T H E ~ A CL~ ~ ~ I N G

30.5 D ~ s r e r s F o w s m ~ ~ ~ r r o ~ s 30 7 WELDABILITY OR J O I N A B J ~ ~ Y 30.6 HEAT BE$^ 30.8 SUMMARY

Weidinghn4ttclllurgy %em1 J3feets inBrazing and %ldemg Thermal-fndvced Residual Btfects of Thermal Stresses

90.3 S ~ i C~IWMON m CONCERNS

0 . 4
OF FUSION WELDS AND TLpEs OF JOlNTS

30.1 INTRODUC~ON TO CONSOLIDATION PROCESSES

Large-size products, pioducts with a high degree of shape complexity, or products with a wide variation in required properties are often manufactured as joined assemblies of two or more component pieces These pieces may be smaller and therefore easler to handle, simpler shapes that are easier to manufacture, or segments that have been made from differentmaterials. Assembly IS an important part of the manufacturing process, and a wide variety of con6olidatranprocesses have been developed to meet the various needs. Each of the methods has its own distinctive characteristics, strengths, and weaknesses. The metallurglcal processes of welding, brazug, and soldering are usually used to joln metals and often involve the sol~dificatlon of molten material.Thc use of discrete fasteners (such as nuts, bolts, screws, and rivets) requires the creation of aligned holes and produces stress localrzation. Whiie the holes may affect performance, disassembly and reassembly can often be performed with relatlve ease. Adhes~ve bonding has grown with new developments in polymer~c materials and is being used extensively in automotive and aircraft production. Any materlal can be joined to any other material, and low-temperature joimng is particularly attractwe for composite materials. Production rates are often low, however, because of the time requlred for the adhestve to develop full strength. Lesser known joining techniques include shrlnk fits, slots and tabs, and a wlde varlety of other mechanical methods. From a technical viewpo~nt, powder metallurgy is another consolidation process, since the end product 1s bull1 up by the joining of a multitude of individual particles. Our survey of consohdation processes will begin with a spectrum of techniques known by the generic term of weldmg.Welding is the permanent lo~ning of two materials, usually metals, by coalescence,wh~ch 1s Induced by a combmation of temperature, pressure, and metallurglcal conditions. The particular combination of these variables can range from high temperature with no pressure to high pressure w ~ t h no increase in temperature. Because welding can be accompl~shed under a wide vanety of conditions, a number of different processes have been developed.Weld~ng is the dominant method of jo~ning in manufacturing, and a large frachon of metal products would have to be drastically modified it weldlng were not available. Coalescence between two metals requires sufficient proximity and activ~ty between the atoms of the pleces being joined to cause the formation of common crystals. The dealm metallurgical bond,for which there would be no noticeable or detectable interface, would requlre (1)perfectly smooth, flat, or matching surfaces; (2) surfaces that are clean

827

: H A ~ R30

Fundamenbls o f loining and free from oxides,absorbed ga internal impurities; and [4) the structure and orientation.Thme conditrans and are virtually impossible to achievein normal production. Con the various joining methods have been designed to overcame or campeusa common deficiencies. Surface rou&hnesscan be overcome flattening of the hi& pokts,or by melti ious processes also employ different approaches to cleaning the metal surfaces p welding and preventing further oxid In sohd-state welding, contaminated layers are generally re chemical cleaning prior to welding or by causing sufficient faceso that new surface is created and exlstmg impurities a 1nf.S~ welding, ~ where molten material ir produced and high tempera the reactims between themetal and its surroundings, eentaminants are ofte from the pool tif molten metal through the use offlaxingag in a tiacuum, the contamtnantsare removed much more e to achieve. In the vacuum of outer space, mating parts may weld uudet. extremely loads, even when welding is not intended. When the process requires heat that is sufficient of themetal may be significantlyaltered. Even whenu coolimg of the weldins ptoces can affect the metallur the weld anrl the adjaceat materiaL Since many of the changes are detrimenta1,th s~ble condequences of heating and cooling should be a major cnn'sideration selecliug a joining process. To produce a high-qual~ty mid, we will need (1) pressure, (2) a means of protecting or cleaning the metals to be joined, and (31 tcr avoic1,or compensatefor, hharmful motalIurgical in the sectiahs thar follow.

CLASSIIXCATION OF WELDING AND THERMAL

CUTTING PROCESSES

Wherever poS$ible, this text will utilize the nomenclature of the American Society (AWS). The various weldlng processes have h e w classified in the presented in figure 30-1, and letter symbols ha designation. The variety of praceses promde and make it possible ir, produce effective an and combinatiom of metals Chapter 31 wiU present the gas and arc we Clraptm 32 will cover re*stadce and solid-state welding; and Chaprer 33 will pr variety of other processes, including brazing and soldering. For m a y years, welding equipment (such as oxyfuel torches and electric arc has also been used to cut metal sheets and plates. DeveIoped o~ighally for salv repair work, then used for preparing plafes for welding, this type of equipmen wdely usett tocut sheets ahd platesinto desired shapes for a variew of us Laser- and electron-beamequipment annow cuthoth mewls andndumetals at to 25 d m i n [ I 0 0 0in./rnin), w~th accnrades of up to 0.25 mm [QOL in.). Figure rnarizes the commonly used t h e r d cvrting processes and provides their A W tions. Since cu~ting is often an adaptation ofwelding, the welding and cuitiug capa will be presented together as the individual processes are diiussed.

SOME C O ~ OCONCERNS N
Many oi the problems that are ihhefknt to welding and joining can lecting the proper process and considermg hoth general and process-specific ch teristics and requirements. Proper d e s i of ~ the joint is extremely critical. Heat meltin&and resolidiRmtiou can prodwe drastlo changes in the properties of baw

Solid state weld~ng (SSWI

Oxyacetylene weldmy (OAW) Pressure gas weldlng (PGWI Arc w e l d ~ n g IAW) Sh~elded metal arc weid~ng (SMAWI Gas metai arc welding IGMAW) Pulsed arc (GMAW-PI Short-oircutt arc IGMAW-S) Electrogas IGMAW-EGI Spray transfer IGMAW-ST)" Gas tungsten arc welding IGTAW! Flux cored arc weldlng IFCAWI Submerged arc w e l d ~ n g (SAW1 Plasma arc w e l d ~ n g (PAW! Stud weldlng (SW) Forge weld~ny IFOW) Cold weld~ng (CWl Frlction weldlny (FRW) Ultrason~c weld~ng (USWI Explosion w e l d ~ n g (EXW) Roll weldlng (ROW] Un~que processes Therm~t weldlng ITW) Laser beam weldlog (LBW! Electrosiag welding (ESW) Flash weld~ng (FWI lnduct~on weld~ng (IW) Electron beam weldlng (EBW)

Pro~ection weld~ng IRPW)

"Not a standard AWS d e s ~ g n a t ~ s n

Cutting processes

I
Oxygen cutting IOC) Oxyfuel gas cutting (OFC) Oxyacetylene (OFC-A) Oxynatural gas (OFC-N) Oxypropane IOFC-P) Oxyhydrogen (OFC-H) Oxygen lance cutting ILOC) Metal powder cutting (POC) Chemical flux cutting (FOC) Arc cutting (AC) Air carbon arc cutting IAAC) Carbon arc cutting ICAC) Gas tungsten arc cutting IGTAC) Plasma arc cutting (PAC1 Laser-beam cutting (LBC)

FIGURE 30-2 Class~flcat~on of

thermal cuttlng processes along w~th t h e ~ AWS r (Amencan

Weldlng Soaety) des~gnat~ons

Electron-beam cutting IEBC)

filler tnater~als. Weld metal properties can also be changed by dilution ot the filler by melted base metal, vaporization at various alloy elements, and gas-metal react~ons Var~ous types of weld detects can also be producctd.These include cracks in various forms, cavlties (both gas and shrinkage), inclusions (slag, flux, and oxides), incomplete fUSlon between the weld and base metals,rncompletepenetrat~on(msufficient weld depth), changes unacceptable weld shape or contour,arc strlkes,spatter,undesirahle metallurg~cal (aglng, grain growth, or transformations), and excessive distortion. Figure 30-3 depicts several of these detects

30.4 TYPES OF FUSIONWELDS AND TYPES OF JOINTS

F~gure 30-4 illustrates (our basic types of fus~on welds. Bead welds, or surfacing welds, are made dnectly onto a flat surface and thesefore require no edge preparation. Since the penetration depth is hm~ted, bead welds are used primarily for joinlng thin sheets of metal, building up surfaces, and depositing hard-facing (wear-resistant) materials

Incomplete fusion

Incomplete penetration

-+
Undercut
FIGURE 30-3 Some common welding defects.

Weld shape and contour

FIGURE 3 0 - 4

Four basic types

of fusion welds.

(or surfacing weld)

Groove welds are used when full-thickness strength is desired on thikker mat Some sort ot edge preparation 1 s required to form a groove between the abuttlng V, double-V (top and bottom), U, and J (one-~idedV) configurations are most co and are often produced by oxyacetylene flame cutting The specif~c type of groove ally depends upon the th~ckness of the joint, the welding process to be employed, the posltion o f the work. The objective is to obtain a sound weld throughout the tlnckness with a minimum amount of addit~onal weld metal. If possible, single-passw ~ n is g preferred, but multiple passes may be required, depending upon the thicknes the material and the weldvng process being used. As shown in Flgure 30-5, spe consumable inserts can be used to ensure proper spaclng between the mating edges good quality ln the root pass.These Inserts are particularly useful mpipeline welding other applications where welding must be performed from only one side of thew Fillet welds are used for tee, lap, and corner joints and require no special e preparation. The size of the fillet is measured by the leg of the largest 45'right tr' that can be inscribed within the contour of the weld cross section. 'Illis is sho
FIGURE 30-5

The use of a consumable backup insert in mak~ng a fusron weld. (Courtesy Arcos Industries, Mount Carmei, PA).

Insert in place

Insert tack-welded

Insert consumed

Completed weld

Excess

FIGURE 30-6 Preferred shape and the method of measuring the size of fillet welds.

Figure 30-6, which also depicts the proper shape for fiiet welds to avoid excess metal depos~tion and reduce stress concentration Plug welds attach one part on top of another and are often used to replace rivets or bolts. A hole is made in the top pldte and welding 1s started at the bottom of thls hole. F~gure 30-7 shows five bas~c types of jo~nts (loin1configurations) that can be made with the use of bead, groove, and fillet weld$ and Figure 30-8 shows some of the methods to construct these joints. In selecting the type of joint to be used, a primary consideration should be the type of loading that will be appl1ed.A large portion of what are erroneously called "welding failures" can more accurately be attributed to inadequate consideration of loading. Cost and accessibility for welding are other important factors when specifyng joint des~gn but qhould be viewed as secondary to loading. Cost is affected by the amount of requ~red edge preparation, the amount of weld metal that must be depou~ted, the type of equipment that must be used, and the speed and ease wlth which the weld~ng can be accomplished.

~utt

La P

FIGURE 30-7

Five basic joint designs for fusion welding.

Tee
Butt joints

Corner
T ioints

Edge Corner joints

VI;,
Single-pass square-groove butt jolnt

Double-pass square-groove
butt joint

Double bevel-groove T joint

S~ngle bevel-groove corner jolnt

Single V groove butt jo~nt Two-filletT joint

FIGURE 30-8 Various weld
;

Two fillet corner joint

procedures used to produce welded joints. (Courtesy Republic Steel Corporation, Youngstown, OH).

Double V-groove buttjo~nt

Large structure

20

40

60

80

100

120

140

Temperature

Effect of sire on the transition temperature and energyabsorbing ability of a certain steel. While the larger structure absorbs more energy because of its size, it becomes brittle at a much higher temperature.
FIGURE 30-9

30.5 DESIGN CONSIDERATIONS

Welding is a unique process that cannot be directly substituted for other methods of joining without proper consideration of its particular characteristics and requirements. Unfortunately, welding is also easy and convenient, and the considerations of proper design and implementation are often overlooked. One very important fact is that welding produces monolithic, or one-piece, structures. When two pieces are welded together, they become one continuous piece. This can cause significant complications. For example, a crack in one piece of a multipiece structure may not be catastrophic, because it will seldom progress beyond the single piece in which it occurs. However, when a large structure, such as a ship hull, pipeline, storage tank, or pressure vessel, consists of many pieces welded together, a crack that starts in a single plate or weld can propagate for a great distance and cause complete failure. Obviously, this kind of failure is not the fault of the welding process itself but is simply a reflection of the monolithic nature of the product. It is also important to note that a given material in small pieces may not behave as it does in a larger size. This feature is clearly illustrated in Figure 30-9, which shows the relationship between the energy required to fracture and temperature for the same steel tested as a small Charpy impact specimen (see Chapter 2 and Figure 2-20) and as a large, welded structure.'In the form of a small Charpy bar, the material exhibits ductile behavior and good energy absorption at temperatures down to 25F (4C). When welded into a large structure, however, brittle behavior is observed at temperatures as high as 110F(43C). More than one welded structure has failed because the designer overlooked the effect of size on the notch-ductility of metal. Another common error is to make welded structures too rigid, thereby restricting their ability to redistribute high stresses and avoid failure. Considerable thought may be required to design structures and joints that provide sufficient flexibility, but the multitude of successful welded structures attests to the fact that such designs are indeed possible. Accessibility, welding position, component match-up, and the specific nature of a joint are other important considerations in welding design.

WELDING METALLURGY
Heating and cooling are essentmal and mtegral components of almost all welding processes and tend to produce metallurgical changes that are often undesirable. Infuslon iveldmng, the heat is suffinent to melt some of the base metal, and this 1s often followed by a rapid cooling. Thermal effects tend to be most pronounced for this type of weldmg, but they also exlst to a lesser degree m processes where the heatmg-cooling cycle is less severe.If the thermal effects are properly considered, adverse results can usually be avoided or

FIGURE 30-10 Schemabc of a butt weld between a plate of metal A and a plate of metal 8 , wth a bask~ng plate of metal C and weld nuaaet filler o f metal D. The result~na becomes a complex alloy of i l l four m&ls.

minimized, and exceUent service performance can be obtained. If they are overlooked, however, the results can be disastrous Because such a wide range of metals are welded and a variety of processes are used, welding metallurgy is an extensive subject,and the material presented here servesonly as an introduction. I n fusion weldtng, a pool af molten metal IS created,with the molten metal coming from either the parent platealone (autogenolks welding) or a mixture of parent and filler matenal.Figure 30-10 shows a butt weld between plates of mater~al A andmaterial B. A b a e k i i ship of material C 1s used with filler metalof material D. In this situation, the molten pool is actually a wmplex alloy of all four materials.Tne moIten material is held in place by a metal "mold" formed by the surrounding solids. Since the molten pool 1s usually small compared to the surrounbg metal, fusion welding can often be viewed as e smll metal cesfing in a large metal mold,The resultant structure and tts properties ean be best understood by first analyzing the castlng and then considering the effects of the associated heat on the adjacent base material. Figure 30-11 shows a typical mlmoSlructure produced by a fusian weld. In the center of the weEd is a region composed of metal that has solidified from the molten state.The marerlal in this weld pool, or fusion ?one,IS actually a mixture of parent metal and electrode o r filler metal, with the ratio depending upon the particular process, the type of jotnt, and the edge preparation. Figure 30-12 compares two butt-weld designs, where the weld poal in the upper design would contain a large percentage of base metal and the weld pool in the lower design would be largely filler material.The metal in the fusion zone is cast material vath a microstructure reflecting the cooling rate of the weld. T h i i region cannot be expected to have the same properties and characteristics as the wrought material being welded,since then processing hrstories and resultingsttuctures

and various zones in a fus~on weld.

FIGURE 30-11 Grain structure

FIGURE 30-12 Comparison o f two but weld designs. In the top weld, a large percentage of the weld pool 1s base m In the bottom weld, most of the weld IS f~ller metal

ing upon the type and volume of weld metaland therate of cooling, but most e

chemical seyreetlon, grain size variation, grain shape prubtems, and orientation e

and a range of properties. In steels, the structures can range from hard, brtttle mate site all the way through coarse pearlite and ferrite. Because of its altered structure,the heat-affected zone may nolonger possess desirable properties of the parent material, and since it was not melted, it cannot sume the properties of the solidified weld metal. Consequently, th~sis often the we est area in the as-welded jo~nt. Except where there are obvious defects in the w deposit, most welding failures originate in the heat-affectedzone This region extends o ward from the weld to the location where the base metal has experienced too lrttle to be affected or altered by the welldingprocess. Figure 30-13presents a schematic

can experience diisiou-induced changes within the orig~nal structure. Because oftbe meltinp,solid~sation, and exposure to a range of high tempe the structure and properties of welds can be extremely complex and vaned.?hrou erconcern,however, associated problems can often be reduced or totally eliminat of the varlous p sideration should first be given to the thermal character~stics

rase of heat input.Processes with low rates of beat input (slow heating) tend large total heat content within themetal, slow cooling rate$ large heat-affected

in~tiai edge of joint

Weld metal

FIGURE 30-13 Schemattc o fa

Transformation

fusion weld in steel, presenting proper term~nology for the vartous regtons and interfaces. Part of the heat-affected zone has been heated above the transformatron temperature and will form a new structure upon cooling. The rematning segment of the heat-affected zone expertences heat alteration of the in~ttal structure. (Courtesy Sandvfk AB, Sandv~ken, Sweden)

?
Weld zone ltne

line

Parent metal

I r l
Weld zone Low Rate of Heat Input
Oxyfuet welding ( O W EleCtroslag welding (ESW) mash weldmg (FW)

Hlgh Rate of Heat Input

Plasma arc weldlng (PAW) Electron-beam weldlng (EBW) Laser wcld~ng (LBW) Spot and seam reaslance weld~ng (RW) Percus~~lon weldrng

Moderate Rate of Heat Input

Shlelded metal arc weldrng (SMAW) Flux mred arc wsld~ng (FCAW) GAS metal arc weldlng (GMAW) Submerged arcwelding (SAW) Gas tunesten arc weldnr (GTAW)

resultant structures with lower strength and hardness but higher ductility. High-heat-input processes, on the other hand, havelow total heats,fast cooling r a t q and small heat-affected i l l also increase with increased starting zones. The size of the heat-affected zone w temperature, decreased weld~ng speed, increased thermal conductivity of the base metal, and a decrease in base metal thickness. Weld geometry is also important, with fillet welds producing smaller heat-affected zones than butt welds If the as-welded properties are unacceptable, the entire welded assembly might be heat treated after welmng. Strncture variations can be reduced or eliminated, but the results are restricted to those that can be produced through heat treatment.The structures and properties associated with cold working, for example, could not be achieved. In addition, problems may be encountered in trying to achieve controlled heating and cooling (heat treatment) within the large, complex-shaped structures commonly produced by weldlng. Moreover, furnaces, quench tanks, and related equipment may not be available to handle the full size of welded assemblies. An alternative technique to reduce microstructural vanation, or the sharpness of that vanation, is topreheat e~ther the entire base metal or material at least 10 centimeters (4 inches) on either side of the joint just prior to welding. This heating serves to reduce the cooling rate of both the weId deposit and the immediately adjacent metal in the heat-affectedzone.The slower cooling produces a softer,more ductile structure and provides more time for the out-diffusion of harmful dissolved hydrogen. The weldtng stresses are dismbuted over a larger area, reducing the amount of weld distortton and the possibility of cracking. Preheattng is more common w ~ t h alloy steels and thicker sections, and it is particularly important with the Iugh-thermal-conductiv~ty metal$ such as copper and aluminum, where the cooling rate would otherwise be extremely rapid.

If the carbon content of plain carbon steels is greater than about 0.3%, the co ingrates encountered in normal weEd~ng may be suff~c~ent to produce hard, untemper martenalte, with an accompanyingloss of ductilrty. Slnce alloy steels possess higher h enability, the likehhood of martensite format~on wdl be even greater w t h these rials Speclal pre- and postweldlng heat cycles (preheat and pofthear) may be requrr when welding the higher-carbon and alloy steels. For plain-carbon steels, a preheat te perature of 200" to 400F (100" to 200C) is usually adequate. Because they can be we ed without the need for preheating or postheatlng, low-carbon, low-alloy steels extremely attractive for weldmg apphcations. In joinlng plocesses where l~ttle or no melt~ng occurs, cons~derable pressure often applied to the heated metal (as in forge or resistance welding).The weld reg1 experiences deformation, and the resultant structure exhib~ts the character~$tics o f wrought material. Since steel is the primary metal that is welded, our discussion of metallurgical tech has largely Cocused on steel. It should be noted, however, that other metals also h ~ b l heat-related t changes in therr structure and properties. The exact effects of heating and cool~ng assoc~ated w~th weldrng will depend upon the spewfic transform tions and structural changes that can occur w~thin the rnatenals being jolned

THERMAL EFFECTS I N BRAZING AND SOLDERING In brazlng and soldering. there 1s no melting of the base metal, but the jolnt still conta a reglon of solidified liqurd and heat-affected sections withm the base material Forth processes, however, another thermal efiect may become quite significant.The base filler metals are usually of radically different chem~stries, and the elevated tempera of joining also promote interd~tfusion. Intermetallic phases can form at the int and alter the properties of the joint. If present in small amounts, they can enhance bon lng and provlde strength reinforcement Most ~iatermetall~c compounds, however, a quite brittle.Too much intermetallrcmaterial can result m signlficanl loss of both streng and ductilrty. THERMAL-INDUCED RESIDUAL STRESSES Another effect of healing and coolimg 1s the introductton of restdual stresses.In weldrn these may be of two types and are most pronounced in fuslon weld~ng, where the grea est amount of heating occurs Their effects can be observed in the form of dimens~on changes, distortion, and cracking. Reridual weldcng stresses are the result of restraint to thermal expansion and co traction by the pieces being welded. Consider a rectangular bar of metal that is uni formly heated and cooled. When heated, the matenal expands and becomes larger length, width, and thickness Upon cooling, the material contracts, and each Qmensio returns to its orrginal value. Now clamp the ends of the bar in a vlse so that l e n g t h s expansion cannot take place and repeat the thermal cycle Upon heating, all of the panslon is restneted to the width and tlrickness, but the contraction upon cooling wills occur uniformly The resulting rectangle will be shorter, th~cker,and wider than the on lnal specrmen. Now apply these principles to a weld berng made between two plates, like tha illustrated in Flgure 30-14. As the weld is produced, the liqu~d region conforms tot shape of the "mold," and the adjacent materlal becomes hot and expands. The molt pool can absorb expansion of the plate perpendicular to the weld line, but expanmo along the length of the weld tends to be restramed by the adjacent plate material tha is cooler and stronger This resrstance or restraint 1s often sufficient to Induce deforma tion of the hot, weak, and thermally expandlug heat-affected zone, which now become thlcker instead of longer. After the weld poolsolid~f~es, both the weld metal and adjacent heat-affected regio cool and contract. The surroundng metal now resists t h contraction. ~ The weld reyo wants to contract but 1s restramed and forced to remaln in a"stretchrd"'cond~tion, kno as residual tensron (reg~on T).The coolrng weld,in turn, exerts forces that try to squeez the adjacent material, produc~ng regions of residual compression (regions C).Wh~le the

Weld -

FIGURE 3014

Schematic of the longltud~nal res~dual stresses In a fus~on-welded butt lornt.

(a1

T
(bl

Shrinkage o fa typical butt weld in the transverse (a) and lona~tudinal (bl d~rect~ons as ;he materiai respond; io the induced stresses. Note that restrictina transverse motion will place the entire weld in transverse tension.
FIGURE 30-15

net force must remain at zero (in keeping with the equilibrium laws of physics and mechanics), the localized tensions and compressions can be substantial. As Figure 30-14 depicts, a high residual tension is observed in the weld meta1,which becomes compressive and then returns to zero as one moves away from the weld centerline.The magnitude of the residual tension will be relatively uniform along the weld line, except at the ends, where the stresses can be relieved by a pulling in of the edges. During cooling, the thermal contractions occur both parallel (longitudinal) and perpendicular (transverse) to the weld line. Lateral movement of the material being welded can often compensate for the transverse contractions. The width of the welded assembly simply becomes less than that of the positioned components at the time of welding. Figure 30-15a depicts this reduction in width, while Figure 30-15h illustrates the longitudinal contractions that generate the complex stresses of Figure 30-14. Components being joined during fabrication typically have considerable freedom of movement, but welds made on nearly completed structures or repair welds often join components that are somewhat restrained. If the welded plates in Figure 30-15a are restrained from horizontal movement,additional stresses will be iaduced.These residual stresses are known as reaction stresses, and they can cause cracking of the hot weld or heat-affected material, or contribute to failure during subsequent use. Their magnitude will be an inverse function of thc length between the weld joint and the point of restraint, and can he as high as the yield strength of the parent metal (since yielding would occur to relieve any higher stresses).

EFFECTS OF THERMAL STRESSES Distortion or warping of the assembly can easily result from the nonuniform temperatures and thermal stresses induced by welding. Figure 30-16 depicts some of the distortions that can occur during various welding configurations. Since the causing conditions

(b)

(Cl

FIGURE 30-16 Distortions or warpage that may occur as a result of welding operations: (a) V-groove butt weld where the top o f the joint contracts more than the bottom; (b) one-side fillet weld in a T-joint; (c) two-fillet weld in a T-joint with a high vertical web.

can vary w~dely, no fixed rules can be provided to assure the absence of warping.The lowing suggestions can help, however. Total heat input to the weld should be minimized. Welds should be made with least amount of weld metal necessary to form the joint. Overweldmg. is not an ass since it actually increases residual stresses and distortion. Faster welding speeds the welding time and also reduce the volume of metal that is heated. Welding seq should be designed to use as few passes as possible, and the base material sho perm~ttedto have a high freedom of movement. When constructing a mu assembly,it is beneficial to weld toward the point of greatest freedom, such as fr center to the edge. The initial components can also be oriented out of position,so that the subse distortion will move them to the desired final shape. Another common procedur completely restrain the components during weldmg, thereby forcing some plastic in the joint and surrounding mater1al.This procedure is used most effectively on s m weldments where the reaction stresses will not be hlgh enough to cause craclung. Shll another procedure is to balance the resulting thermal stresses by deposit the weld metal in a specified pattern,such as short lengths along a joint or on altern ing sides of a plate. Warping can also be reduced by the use of peenzg. As the we bead surface IS hammered with the peening tool or material, the metal is flattened tries to spread. Being held back by the underlying material, the surface becomes pressed or squeezed. Surface rolling of the weld-bead area can have the same e In both processes, the compressive stresses induced by the surface deformation serve offset the tensile stresses induced by welding. Residual stresses should not have a harmful effect on the strength of weldme except in the presence of notches or in very rigid structures where no plastic flow c occur. These two conditions should not exist if the welds have been properly desi and proper workmanship has been employed Unfortunately,it is easy to inadvert join heavy sections and produce rigid configurations that will not permit the sm amounts of elastlc or plastic movement required to reduce highly concentrated stress In addition, geometric notcheg such as sharp interior comers, are often lncorporat into welded structures. Other harmful "notches," such as gas pockets, rough bea porosity, and arc "strikes," can serve as initiation sltes for weld failures. Thes generally be avoided by proper weldmg procedures, good workmanship, and ade supervision and inspection. The residual stresses of weldmg can also cause additional distortion when subsequ machining removes metal and upseb the stress equilibrium. For this reason, welded semblies that are to undergo subsequent machining are frequently given a stress-relief h treatment prior to that operation. The reaction stresses that contribute to distortion are more often associat with cracking dunng or immediately following the welding operation (as the we cooling).This cracking is most likely to occur when there is great restraint to the s age that occurs transverse to the direction of welding. When a multipass weld is made, cracking tends to occur in the early beads where there is insufficient weldm to withstand the shrinkage stresses.These cracks can be quite serlous if they go tected and are not chipped out and repaired, or melted and resolidified during quent passes.Figure 30-17shows the various forms of cracking that can occur as ares of welding

eld

metal crack

Toe crack

and locat~ons of crack~ng that can occur as a result of weld~ng.

FIGURE 30-17

Various types

eat-affected zone crack Weld ~nterface cra

Underbead crack

ot surface crack

To mmimize the possibility of fracture, welded joints should be des~gned to keep restramt to a minimum.The metals and alloys of the structure should be selected w~th welding in m ~ n d (more problems exlst with higher-carbon steels, higher-alloy steels, and high-strength materials), and special consideration should be given when weldlng th~cker materials. Crack-preventwn efforts can also include maintaining the proper size and shape of the weld bead.Wbile a concave fillet 1s des~rable when machining, a concave weld profile has a greater tendency to crack upon cooling, since contraction actually increases the length of the surface. With a convex profile, the length of the sudace w~ll contract s~multaneouqly wlth the volume, reducing the possibility of surtace tension and cracking Weld heads with high penetration (high depth-width ratio) arc also more prone to crackmg. Still other methods to suppress cracking focus on reduc~ng the stresses by making the coolmg more uniform or relaxing them by promotrng plasticity in the metals being welded.The metals to be welded may be preheated, and additional heat may be applied between the welding passes to retard cooling. Some welding codes also require the inclusion of a thermal stress rehef after welding but prior to use. Hydrogen dissolved in the molten weld metal can also mduce crackmg. Slower welding and cooling wdl allow any hydrogen to escape, and the use of low-hydrogen electrodes and low-moisture fluxes will reduce the likelihood of hydrogen being present.

H 30.7 WELDABILITY OR JOINABILITY

It isimportant to note that not alljoining processes arecompatible with all engineering mater~als.Whilethe terms weldability or joanabili~imply a reliable measure of a material's ability to be d d e d or joined, they are actually quite nebulous. One process might produce excellent results when applied t o a given material, whereas another might produce a dismal failure. Withm a given process, the quality of results may vary greatly with variations in the process parameters, such as electrode material, shielding gases, welding speed, and cooling rate. Table 30-2 shows the compatib~lity of the various joining promsses with some of the major elasses of engineering materials. In each case, the process is clas~ified as recommended (R),commonly performed (C),pertomed with same difficulty (D), seldom used (S), and not used (N).It should benoted, however, that the classifications are generahzations, and exceptions often exist within borh the family of materials (such as the

Oxyacetylene
Carbon and low-alloy

Beam
C

Reslstanee

Adhes~ve

Brazrng
R
R R

Soldering

R
R
C

D
C 5

C
C

R
Aluminum and
Copper and copper

C
C

C C

C C

R R
D

R
C

C
C

C
D I~iduction

N
N

S R

C R
C

840

C H A ~30 R Fundamentals of Joining

various types of stainless steels) and the types of processes (arc welding here enco passes a largevariety of 8pecificprocesses).Nevertheless, the table can serve as a gui lime to assist in process selection.

30.8 SUMMARY
If the potential benefits of welding are to be obtained and harmful side effects areto avoided, proper consideration should be given to (1) the selection of the process, process parameters, and the filler material; (2) the design of the joint; and (3) the fects of heating and cooling on both the weld and parent material. Joint design sh
consider manufacturability, durability, fatigue resistance, corrosion resistance, and s Welding metallurgy helps determine the structure and properties across the joint, a as the need for additional thermal treatments. Further attention may be require control or minimize residual stresses and distortion. Parallel considerations apply to brazing and soldering operations, with addition attention to the effects of interdiffusion between the filler and base metals. Flame a arc-cutting operations involve localized heating and cooling, and they also create tered structures in the heat-affected zone. Since many cut products undergo furt welding or machining, however, the regions of undesirable structure may not be-retain in the final product.

H Key Words
autogenous weld base metal bead weld (or surfacing weld) coalescence consolidation processes distortion fillet weld fusion weld fusion zone groove weld heat-affected zone (HAZ) incomplete fusion incomplete penetration insert intermetallic compound joint configuration monolithic peening plug weld posthcat preheat rate of heat input reaction stresses residual stresses stress relief thermal cutting weld metal (or weld poc weldability welding

Review Questions
1. What types of design features favor manufacture as a joined assembly? 2. What types of manufacturingprocesses fall under the classification of consolidationprocmses? 3. Define welding. 4. What four conditions are required to produce an ideal metallurgical bond? 5. What are some of the ways in which welding processes compensate for the inability to meet the conditions of an ideal bond'? 6. What are some possible problems associated with the high temperatures that are commonly used in welding? 7. What are the three primary aspects required to produce a high-quality weld? 8. How are welding prucesses identified by the American Welding Society? 9. What is thermal cutting? 10. What are some of the common types of weld defects? 11. What are the four basic types of fusion welds? 12. What is the role of an insert in welding? 13. What types of weld jointscommonly employ fillet welds? 14. What are the five basic joint types for fusion welding? 15. What aresome of the factors that influencethe cost of making a weldment? 16. Why is it important to consider welded products as monolithic structures? 17. How does the fracture resistanceand temperaturesensiti of a steel vary with changes in material thickness? 18. How might excessive rigidity actually he a liability in a w ed structure? 19. What is antogenouswelding? 20. In what way is the weld-pool segment of a fusion weld 1i k small metal casting? 21. Why is it possible for the fusion zone to have a chemistry is different from that of the filler metal' 22. Why is it not uncommon for the selected filler metal to chemical campositiou that is different from the material b welded? 23. What are some of the defects or problems that can occur in molten metal region of a fusionweld? 24. Why can the resulting material properties vary widely wi welding heat-affected zones? 25. What are some of the structure and property modifi that can occur in welding heat-affected zones? 26. Why do most welding failures occur in the heat-affected 27. What are some of the characteristics and consequent welding with processes that have low rates of heat inpu 28. What are some of the difficulties or limitationsencountere heat treating large, complex welded structures? 29. What is the purpose of pre- and postheating in w e 1 operations?

What heat-related metallurgical effects may produce adverse results when brazing or soldering? What are some of the undesirable consequences of residual stresses? What is the cause of reaction-type residual stresses? How are reaction stresses affected by the distance between the weld and the point of fixed constraint? What are some of the techniques that can reduce the amount of distortion in a welded structure?

35. Under what conditionsmight residual stresses have a harmful effect on load-bearing abilities? 36. Why might a welded structure warp if the structure is machined after welding? 37. What are some of the techniques that can be employed to reduce the likelihood oi cracking in a welded structure? 38. Why are the terms weldability and joinability somewhat nebulous?

II Problems
Through the 1940s the hulls of oceangoing freighterswere constructed by riveting plates of sceel together.When the defense efforts of World War I1 demanded accelerated production of freightm to supply U.S. troops ovemeas, construct~on of the hulls was converted to welding. The resulting Liberty Shrps proved qute suctessful but also drew' considerable attention when mlnor or moderate impacts (usually under low-tempera m e conditrons) produced cracks of lengths sufficient to senttle the ship, often up to 50 feet ormore.Since thematenal was process change essentially the same and the only s~gnifrcanl had been the convemon fromnvetmngto weldmg, the wetding process was blamsd for the failures. a Is this a fair assessment? b What do you think may have contr~buted to the problem? c. What evidence might you want to gather to suppart your beliefs? 'F$vo piewi oi.ZISI 1022 steel arc h-me bh~clJed-metal sr~wcldcd uslth E601? olccr~odcs. Sonic ditfi~,ultv is b z i ~ ~ experisrlccd p w~th crackingrn the weld beads and in the heat-&ectedmnes What posslhle corrective lneasures might you suggest? Figure 30-Aschematically depmts the design of a goqart frame wth cross bars and seat support.The assembly IS to he constructed from hot-rolled, low-carbon, box-channel matertdl with mitor, butt, and fillet welds at the 12 numbered joints. Due to the solrdification shrinkage and subsequent thermal contraction ofthe ~ointmatemal, the welds are best made when one or more of the heect~ons are unrestrained If thestructure is too rigd a t the time of welding,the asgociated dnneirsional changes aFe restricted, cawing the generatcon of residual stresses that can lead to distortion, crackrng, or tears. a. Consider the 12 welds in the pr~pased structure and recommend a weldmgsequence that would m i m i z e the gossibility of hot tears and ctacks due to the welding of a restraned joint. b. Your company rs dsveloping a computer-asslsked design program. Suggest one or more rules that may he programmed to aid t n the selecton of an acceptable weld sequence.

FIGURE 30-A

GASFLAME AND ARCPROCESSES

31.1 OXYPUEI-GAS WELDING Oxyfuel-Gas Welmng Processes Uses, Advantages, and Limitations Pressure Gas Welding 31.2 OXYGEN TORCH CUT~NG Processes Fuel Gases for OxyfuelGas Cutting Stack Cuttlng Metal Powder Cutting, Chemtcal Flux Cutting, and Other Thermal Methods UnderwaterTorch Cutling 31.3 FLAME STRAIG~N~NG 31.4 ARC WELDING 31.5 CONSUMABLE-ELECTRODE ARC WELDING Shxelded Metal Arc Welding Flux-Cored Arc Welding Gas Metal Arc Weldlug Submerged Arc Welding 31.8 ARCCuTnNff Air Carbon Arc Cuttlng

31.6 NONCONSUMABLEARCWELDING

Plasma Arc Weldi 31.7 WELDING EQUIPMENT Power Sources for Arc Jigs, Posit~oners, and

11 31.1 OXYNEL-GAS WELDING

OXYFUEL-GAS WELDING PROCESSES

nonferrous metals can be welded), and thc low capital investment required Acety is still the principal fuel gas. The combustion of oxygen and acetylene (C,H,) by means of a weldlng torc the type shown in Figure 31.1 produces a temperature of about 3250C (5850F)

Typical oxyacetylene weld~ng torch and cross-sect~onal schemat~c (Courtesy of Vlctor Equrpment Company, Denton, TX)
FIGURE 31-1

842

two-stage reaction. In the first stage, the supplied oxygen and acetylene react to produce carbon monoxide and hydrogen. &HZ

+ O24 2 C O + Hz + heat

reaction occurs near the tip of the torch and generates intense heat.The second stage of the reaction involves the combustion of the CO and Hz and occurs just beyond the first combustion zone. The spec~ficreactions of the second stage are: 2CO

+ O2+ZC02 + heat
-

H,

1 +2 Q2'H20 + heat

The oxygen for these secondary reachons is generally obtained from the surrounding atmosphere. The two-stage combustion process produces a flame having two dstinct regions. As shown in Figure 31-2, the maximum temperature occurs near the end of the inner cone, where the first stage of combustion is complete. Most welding should be performed with the torch positioned so that this point of maximum temperature is just above the metal being welded. The outer envelope of the flame serves to preheat the metal and, at the same time, provides shielding from oxidation, since oxygen from the surrounding air is consumed in the secondary combustion. Three different types of flames can be obtained by varying the oxygen-acetylene (or oxygen-fuel gas) ratio. If the ratlo is between 1 : l and 1.15.1,all reactions are carried to completion and a neut~alflame is produced. Most weldmg is done with a neutral flame, since it \rill have the least chemical effect on the heated metal. A higher ratio, such as 1,5:l,producesan oxzdizing&me, which is hotter than the neutral flame (about 3600C or 6000F) but simlar in appearance. Such flames are used when welding copper and copper alloys but are generally considered harmful when

Position

Possible aewlene
"feather"

FIGURE 31-2 Typical

oxyacetylene flame and the associated temperature distr~bution.

welding steel because the excess oxygen reacts with the carbon in the steel, lowe the carbon in the reglon around the weld

ization occurs, but the metal is well protected from oxidation. Flames of this type a in welding Monel (a nickel~opper alloy),high-carbonsteels, and some alloy stee for applying some types of hard-facing material. For welding purposes, oxygen 1s usually supplied from pressurized tanks m a r Because acetylene is not safe when stored as a gas at pressures above 0.1 MPa (1.5 p it is usually dissolved in acetone The storage cylmders are fiUed with a porous fil Acetone is absorbed into the voids ~nthe filler mater~al and serves as a medium dissolving the acetylene.

lene.Butane, natural gas, and hydrogen have been used in wmbmat~on with air ar

and are controlled by pressure regulators on each tank. Because mixtures of acetyle

gas velocities that m~ght blow themolt sulting ID greater heat Input without the h~gher metal from the weld puddle.Larger torch tips are used for thewelding of thcker met

USES, ADVANTAGES, AND LIMITATIONS

Almost all oxyfuel-gas welding isfuslon weldmng.The metals to be jomed are simply me1 where a weld is desired and no pressure is required. Because a slight gap often exists

F~gure 31-3 shows a schematic of oxyfuel-gas weldmg usmg a consumable welding rod.

FIGURE 31-3 Oxyfuel-gas welding with a consumable welding rod.

Pmtcctlon Elecirade Material (omed Rate of heat input Weld profile (DepthANldthJ Max. peaetrauon Assets Limltdtion~

Gases produced by combustlirn None B ~ sfur t steel and other ferrous metals Low
'I,

3 mm Cheap, slmple equqmcnt, pornblc, vezszillle Large HAZ,slow

To promote the formation of a better bondJ1ws may be used to clean the surfaces and remove conaminatmg-oxide. In addition, the gaseous shield produced by vaporizm n g flux can prevent further oxulation during the welding process, and the slag produced by solidifying flux can protect the weld pool as it cools. Flux can be added as a powder, the welding rod can be dipped in a flux paste, or the rods can be precoated. The OFW processes can produce good-quality welds if proper cautlon 1s exercised. Welding can be performed in all pomhons, the temperature of the work can be easlly controlled, and the puddle isvisible to the welder. However, exposure of the heated and molten metal to the various gases in the flame and atmosphere makes it dfficult to prevent eontam~nation. Since the heat source is not concentrated, heating is rather slow. A large volume of metal is heated, and diitort~on is likely to occur.The thickness of the material belng jolnell is usually less than 6 5 mm in.).Thus,in production applications, the flame-welding processes have largely been replaced by arc welding. Nevertheless, flame welding is still q u ~ t e common in field work,in maintenance and repairs, and in fabricating small quantities of spectalized products. Oxyfuel equipment is quite portabIe, relativelymexpensive.and extremely versatile. A single set of equipment can be used for welding, brazing, and soldering, and as a heat source for bending, forming straightening, and hardening. With the modifications to be discussed shortly, it can also perform flame cutting.Table 31-1 summarizes some of the key features of oxyfuel-gas welding.Table 31-2 shows rls compdttb'ity with some common engineering materials.

(vl

PRESSURE GAS WELDING Pressu~e gas weldtag (PGW) is a process that uses equrpment simllar to the oxyfuel-gas process to produce butt joints between the ends of ablects such as pipe and railroad rail. The enda are heated with a gas flame to a temperature below the melting point,

arbon and low,alloy steels uminum m d mqnesrum Copper and supper alloys Ntckel and nlckel alloys
Lead and nnc

Thermoplastics. thermosetb,

Di~sim~lar metal8 Metals to nonmetals Dlssrm~lar nonmptalc

Recommended for Iow-earhon and low-alloy steel$ usmg rods of the same matertal,more d~lficult fur higher carbon Common tor th~nner matenal,moce dlfficull Eor thicke Cummon for alumnum thrnner than I IQ ,drfficult fox magnesium alloys C o m n Lor most anoys.rnore difficult for some types of bronzes Common foriirckel, Monel* and InconeEs Not recommended Re~ommended Hot-gas weid~ng used for thermoplasiu, not used w~th thermosets and elastomers Seldom uscd wrth ceramics but common wtth glass D~fficult, best if melt~ng polnts are within 5Ll'F;concem for ealvanr Gofrosum Not recummended
Dlffi~ult

and the soft metal is then forced together under pressure. Pressure gas welding, fore, is actually a form of solid-state welding where the gas flame simply softe metal and coalescence is produced by pressure.

31.2 OXYGEN TORCH CUTTING

PROCESSES
Oxyfuel-gas cutting (OFC), commonly called flame cutting, is the most common the cutting process. In some cases the metal is merely melted by the flame o f t

gas torch and is blown away to form a gap, or kerf, as illustrated in Figure ferrous metal is cut, however, the process becomes one where the iron actu oxidizes) at high temperatures according to one or more of the following r

+ 0 - F e O + heat 3Fe + 20,+ Fe,04 + heat

Fe 4Fe

+ 30, +2Fe203 + heat

Because these reactions do not occur until the metal is above 815C ( oxyfuel flame is first used to raise the metal to the temperature where burning initiated.Then a stream of pure oxygen is added to the torch (or the oxygen con the oxyfuel mixture is increased) to oxidize the iron. The liquid iron ox unoxidized molten iron are then expelled from the joint by the kinetic energy oxygen-gas stream. Because of the low rate of heat input and the need for preh ahead of the cut, oxyfuel cutting produces a relatively large heat-affected zone a sociated distortion compared to competing techniques. Therefore, the used where the edge finish or tolerance is not critical and the edge materi be subsequently welded or removed by machining. Cutting speeds are relatively but the low cost of both the required equipment and its operation make the proce tractive for many applications. Theoretically, the heat supplied by the oxidation will be sufficient to progressing, but additional heat is often necessary to compensate for losses atmosphere and the surrounding metal. If the workpiece is already hot from processing, such as solidification or hot working, no supplemental heating is require a supply of oxygen through a small pipe is all that is needed to initiate and cut. This is known as oxygen lance cutting (OLC). A workpiece temperatur 1200C (2200F) is required to sustain continuous cutting.

FIGURE 31-4 Flame cutting of

a metal plate.

or metal oxide

Oxyfuel-gas cutting works best on metals that oxldize readily but do not have h g h thermal conductiv~ties. Carbon and low-alloy steels can be readily cut in thicknesses from 5 mm to in excess of 75 cm (30 in.). Stainless steels contain oxidation-resistant ingredients and are difficult to cut, as are aluminum and copper alloys.

FUEL GASES FOR OXYFUEL-GAS CUTTING Acetylene is by far the most common fuel used in oxyfuel-gas cnttmg, and the process is often referred to as oxyacetylene cuftzng (OFC-A). Figure 31-5 shows a typical cutting torch.The t ~ contams p a circular array of small holes through which the oxygen-acetylene mixture is supplied to form the heating flame. A larger hole in the center supplies a stream of oxygen and is controlled by a lever valve.The rapid flow of the cutting oxygen not only oxidizes the hot metal but also blows the formed oxides from the cut. If the torch 1s adjusted and manipulated properly, it is possible to produce a relatively smooth cut. Cut quality, however, depends upon careful selection of the process variables,including preheat ~onditions~oxygen flow rate,and cutting speed. Oxygen purities over 99.5% are requiredfor the most efficient cuttlng If the purity drops to 98.5%, cutting speed will be reduced by IS%, oxygen consumphon w~ll increase by 25%, and the quality of the cut wtll dimmish. Cutting torches can be manipulated manually. However, when the process is applied to manutacturing, the deslred path 1s usually controlled by mechanical or programmable means. Specialized equipment haa been designed to produce straight cuts in flat stock and square-cut ends on pipe. The marriage of computer numencally controlled (CNC) machines and cutting torches has also proven to be quitepopular This approach, along wlth the use of robot-mounted torches,provides great tlex~b~hty along wlth good precision and control. Fuel gases other than acetylene can also be used for oxyfuel-gas cutting, the most common being natural gas (OFC-N) and propane (OFGP).While thelr flame temperatures are lower than that of acetylene, their use is generally a matter of economics and gas ava~labllity. For certain specfa1 work, hydrogen can also be uqed (OFC-H). STACK CUTTING When a modest number of duplicate parts are t o be cut from thin sheet, but not enough to justlfy the cost of a blanking die, stack cutting may be the answer.The sheets should be flat, smooth, and free of scale, and they should be clamped together hghtly so that there are no intervening gaps that could interrupt unlform oxidation or permit slag or

Cuttrng

oxygen tube

Oxygen

control valve

"\

Tip

FIGURE 31-5 Oxyacetylene cutt~ng torch and cross-sect~onal schematic. (Courtesy of Vrctor Equ~pment Company, Denton, TX)

848

CHAPTER 31 Gas Flame and Arc Processes

Underwater cuttinq torch. Note the extra set 5 [tie no7z1eto of g i b p c n ~ n q III perlri.1 rne f ow of compresseu atr and the extra control valve (Couflesy of Bost~on-Blessmg Cornpony, Chtcogo, IL)
FIGURE 31-6

molten metal to be entrapped. Obv~ously, stack cutting will produce a less accurate than could be achieved by usmg a blanking die.

METAL POWDER CUTTING, CHEMICAL FLUX CUTTING, AND OTHER THERMAL METHODS When cutting hard-to-cut materials, modified torch techniques may be required. M powder cutting (POC) injects iron or aluminum powder into the flame to raise its cut temperature. Chemical flux cutting (FOC) adds a fine stream of special flux to the ting oxygen to increase the fluidity of the high-melting-point oxides. Both of these rnt ods, however. have largely been replaced by plasma arc cutting (PAC), which is diicur as an extension of plasma arc welding, to be presented later in this chapter. Laserelectron-beam cutting will be presented with their welding parallels in a future chap UNDERWATER TORCH CUTTING The thermal cutting of materials underwater presents a special challenge. A speci designed torch,like the one shown in Figure 31-6, is used to cut steel.An auxiliarys surrounds the main tip. and an additional set ol gas passages conducts a flow of c, pressed air that provides seconda~y oxygen for the oxyacetylene flame and expels w; from the zone where the burning of metal occurs.The torch is either ignited in the u! manner before descent or by an electric spark device after being submerged. Acetyl gas is used lor depths up to about 7.5 m (25 ft). For greater depths, hydrogen is used,si the environmental pressure is too great for the safe use of acetylene.

Flame strazghtentng uses controlled, localized upsettzng as a means of straighter warped or buckled material. Figure 31-7 ~llustrates the theory of the process If astra piece of metal is heated in a localized area, such as the shaded area of the upper diag , deformed) as it softens and trie the metal on side b will be upset ( ~ . eplastically expand against the cooler restrainmg metal. When the upset portion c a d , it will (

Straight
piece

tract, and the resulting piece will be shorter on side b, forcing it to bend to the shape in the lower diagram. If the starting material is bent or warped, as in the lower segment of Figure 31-7, the upper surface can be heated. Upsetting and subsequent thermal contraction will shorten the upper surface at a', bringing the plate back to a straight or flat configuration. This type of procedure can be used to restore structures that have been bent in an accident, such as automobile frames. A similar process can be used to flatten metal plates that have become dished. Localized spots about 50 mm (2 in.) in diameter are quickly heated to the upsetting temperature while the surrounding metal remains cool. Cool water is then sprayed onto the plate, and the contraction of the upset spot brings the buckle into an improved degree of flatness. To remove large buckles, the process may have to be repeated at several spots within the buckled area. Several cautions should be noted. When straightening steel, consideration should be given to the possible phase transformations that could occur during the heating and cooling. Since rapid cooling is used and martensile may form, a subsequent tempering operation may be required. In addition, one should also consider the residual stresses that are induced and their effect on subsequent cracking, stress<orrosion cracking, and other modes of failure.The effects of phase transformations and residual stresses have been discussed more fully in Chapter 31. Also, flame straightening should not be attempted with thin material. For the process to work, the metal adjacent to the heated area must have sufficient rigidity to induce upsetting. If the material is too thin, localized heating and cooling will simply transfer the buckle from one area to another.

With the development of commercial electricity in the late nineteenth century, it was soon recognized that an arc between two electrodes was a concentrated heat source that could produce temperatures approaching 4000C (7000F). As early as 1881,various attempts were made to use an arc as the heat sourcc for fusion welding. A carbon rod was sclected as one electrode and the metal workpiece became the other. Figure 31-8 depicts the basic electrical circuit. If needed,fillevrnetal was provided by a metallic wire or rod that was independently fed into the arc. As the process developed, the filler metal replaced the carhon rod as the upper electrode.The metal wire not only carried the welding current but, as it melted in the arc, it also supplied the necessary filler. The results of these early efforts were extremely uncertain. Because of the instability of the arc, a great amount of skill was required to maintain it, and contamination of the weld resulted from the exposure of hot metal to the atmosphere.There was little or no understanding of the metallurgical effects and requirements of arc welding. Consequently, while the great potential was recognized, very little use was made of the process until after World War I. Shielded metal electrodcs were developed around 1920. These electrodes enhanced the stability of the arc by shielding it from the atmosphere and provided a fluxing action to the molten pool. The major problems of arc welding were overcome, and the process began to expand rapidly. All arc-welding processes employ the basic circuit depicted in Figure 31-8.Welding currents vary from 1to 4000 amps, with the range from 100 to 1000 being most typical.

1
FIGURE

Power supply
DC generator DC rect~f~er 0. E

Electrode cable + ' i ; Voltmeter E l c m Ammeter d e

Electrode holder

or AC transformer

-8 Th e basic

electrica

'cuit for arc we1ding.

Workp~ece
cable

850

CnAmER 31

Gas Flame and Arc Processes

while io111zed atoms in the arc column are accelerated toward the negative elec Since the ions are tar more massive than the electrons, the heat of the arc is more (weld penetration). If the work is made negatlve and the electrode poslt~ve, theco tion is known as reverse polarity (RPDC) or DCEP, for direct-current electrode-pas

ro the dc cond~tions. Vanablepolarity power suppl~es also alternate between D DCEN condit~ons, but they use rectangular waveforms to vary the Gactlon of t ~ m each mode as well as the frequency of sw~tching Weld characteristics can now be led over a continuous range between DCEN and DCEP condit~ons. eZearode processe~)and thus supphes the metal needed to fill the jolnt. Con

some form of shi semrautomatic arc welding, but thn 1s always in conlunction w ~ t h ing and arc-stabihz~ng medium and automatic feed control devlces that maintain proper arc length. The second group of arc-weldmg processes employs a tungsten electrode, wh not consumed by the arc, except by relatively slow vaporization In these nonconsuma

d depends on the skill of the operator.Automation and robot~cs are rcducing th~s dence, but the selection and training of welding personnel are still of great impor

FIGURE 31-9 Three modes of metal transfer during arc welding. (Courtesy of Republic Steel Corporotlon, Youngstown,
OH)
Globular

Spray

short circult

Four processes make up the bulk of consumable-electrode arc welding: 1. Shlelded metal arc welding (SMAW) 2. Flux-cored arc welding (FGAW) 3. Gas metal arc welding (GMAW) 4. Submerged arc welding (SAW) These procesaes all have a medium rate of heat input and produce a fusion zone whose depth is approximately equal to its wldth. Because the fusion zone is composed of metal from both of the pieces being jomed, plus melted filler (i e., electrode), the electrode must be of the same material as that belng welded; the processes cannot be used to join dissim~lar metals or ceramics.

SHIELDED METAL ARC WELDING Shielded rnetnl arc welding (SMAW),also called st~ck weldzng or covered-electrode weldmg,is among the most widely used weldingprocesses because of its versablity and because it requ~res only low-cost equipment.'Ihe key to the process is a tmite-length electrode that consists of metal wlre, usually from 1.5 to 6.5 mm m dlameter and 20 to 45 cm ~nlength, Surrounding the wire is a bonded coating containing chemlcal components that add a number of desirable characteristics,lncludlng all or many of the following

1. Vaporize to provide a protective atmosphere (a gas shield around the arc and pool
of molten metal). 2. Provide ionizing elements to help stabilize the arc, reduce weld metal spatter, and increase efficiency of deposition. 3. Act as aflwc to deoxidize and remove ~mpuritles from the molten metal. 4. Provide a protective slag coating to accumulate impurities, prevent oxidation, and slow the cooling of the weld metal. 5. Add alloymg elements 6. Add additional filler metal. 7. Affect arcpenetratron (the depth of melting in the workpiece). 8. Influence the shape of the weld bead. The coated electrodes ale classified by the tensile strength of the deposited weld metal, the welding positlon in which they may be used, the preferred type of current and polanty (if dlrect current), and the type of coabng. A four- or flve-d~glt system of designation has been adopted by the American Welding Soclety (AWS) and is presented in Figure 31-10. As an example, type E7016 is a low-alloy steel electrode that will provide a deposit with a minimum tensile strength of 70,000 psi (485 MPa) in the non-stressrelieved condition; ~tcan be used m all positions, with either alternating or reverse-polar~ty direct current; and it has a low-hydrogen plus potasslum coatmg. To assist in identification, all electrodes are marked with colors in accordance with a standard established by the National Electrical Manufacturers Association. Electrode selection consists of determining the electrode coating, coating thickness, electrode compos~tion, and electrode d~ameter.The current type and polarlty are matched to the electrode A variety of electrode coatings have been developed.The cellulose and titanla (rude) coatings contam. S10,;Ti02; small amounts of FeO, MgO, and Na,O; and volatlk matter Upon decomposltmn, the volatile matter may release hydrogen, whlch can dissolve m the weld metal and lead to embrittlement or cracklng m the jomt.Low-hydrogen electrodes are available with compositions designed to provide shleldlng without the emission of hydrogen. Since many of the electrode coatings can absorb moisture, and this is another source of undesirable hydrogen, the coated electrodes are often baked just prior to use. To lnltiate a weld, the operator briefly touches the tlp of the electrode to the workpiece and quickly raises it to a distance that will malntaln a stable arc. The intense heat quickly melts the tip of the electrode wlre, the coating, and portlons of the adjacent base

852

CHAI~~ 3B 1R Gas Flame and Arc Processes

I
urrent Minimum tensiiestrength, in 1.000 psi, as-deposited weld metal in non-stressrelieved condition.
45 60 70 80 90 100 120

Po arity

Weiding position 1 All 2 Flat and horizonfal 3 Flat only

reversed

2 Titania

5 Law hydrogen (lime)

6 Low hydrogen +potassium 7 Cellulosic + iron powder

FIGURE 31-10

Designation system for arc-welding electrodes.

droplets w i t h a layer of liquid flux a n d slag.The fluxing constituents u n i t e w i t h an

deposition from a shielded electrode.

Bare section for contact

FIGURE 31-11 A sh~eldea metal arc weld~ng (SMAW) system.

-Electrode Protective gas from electrode coating caating

Electrode wire

FIGURE 31-12 Schematic diagram of shielded metal arc welding (SMAW). (Courtesy of American Iron and Steel Institute, Washington, DC.)

metal

Iron powder can be added to the electrode coatmy to significantly increase the amount 0 1 weld metal that can be deposited with a given sue electrode wlre and current. Alloy elements can also be incorporated Into the coating to adjust the chemistry of the weld. Special contact or drag electrodes utilize coatlngs thdt are de~lgned to melt more slowly than the f~ller wire. If these electrodes are tracked along the surface of the work, the faster-meltmg center wire will be recessed by the p ~ o p e length r to maintam a stable arc. Since electrical contacf must be maintamed w ~ t h the center wlre, SMAW electrodes are finite-length "sticks." Length is I~mited since the current must be supplied near the arc, or the electrode w~ll tend to overheat (by electrical resistance heating) and ruin the coatlng. Overheating also restricts the weld currents to values below 300 amps (generally about 40 amps per mdlimeter of electrode d~ameter).As a result, the arc temin.). Weldperatures are somewhat low, and penetration 1s generally less than 5 mm mgof mater~al thicker than 5 mm will require multlplepasses, and theslag coating must be removed between each pass. The shielded metal arc process is best used for welding ferrous meta1s:carbon steels, alloy steels, stainless stcels, and caTt irons can all be welded. Welds can be made In all pos~tions. DCEP conditions are used to obtaln the deepest poss~ble penetration, \nth alternate modes being employed when weld~ng a th~n sheet. The mode of metal transfer is e~ther globular or short clrcu~t. Shrclded metal arc weldmg is a simple, incxpens~ve, and versatde process, requlrrng only a power supply, power cables, an clectrode holder, and a small varlety of electrodes. The equipment is portable and can even be powered hy gasolinc or diesel generators Therefore, rt is a popular process ln job shops and 1s used extensively in repair operations. The electrode provides and regulates its own flux, and there is less sens~tivity to wind and drafts than m the gas-sh~elded processes. Welds can be made m all pos~tions Unfortunately, the process 1s discontinuous, produces shallow welds, and requires slag removal after each welding pass. Table 31-3 presents a process summary for shielded metal arc welding.

FLUX-CORED ARC WELDING

FIm-cored urc weldmg (FCAW) overcomes some of the linutations of the shielded metal arc process by movlng the powdered flux to the lnter~or of a continuous tubular electrode (Flgure 31-13), When the arc 1s cstabhshed, the vaporizing flux agarn produces a protectlve atmobphere and also forms a slag layer over the weld pool that w~ll r e q w e subsequent removal.Alloy additions (metal powders) can be blended into the flux to create a w~de varlety of filler metal chemistries. Compared to the stlck electrodes of the shielded metal arc process, the flux-cored electrode is both continuous and less bulky, since binders are no longer required to hold the flux in place. The continuous electrode is fed automatically through a weld~ng gun, with electrical near contact being maintamed through the bare-metal exterlor o i the wire at a pos~tion the exit of the gun. Overheating of the electrode 1s no longer a problem, and welding

**.ii\:(id.
i
,.
Hcat source Protection Eleclrodc Material joined Rate of heat input Weld profile (DIW) Current Max. penetration Assets Limitations

..~&v~lu~&aav'~~<;ded Metbl Arc Welding (SMAW &#>*..::,., ;. . , , , . . ..

Electric arc Slag irom flux ~ n gas d Irom vaporized Cdating material Discontinuous, consumable Best for steel Medium
1

c300 amps 3 4 mm Cheap, slmplc equipment Discontinuous, shallow welds; requires slag removal

854

CHAPTER 31

Gas Flame and Arc Processes

Tubular electrode contalnlng powdered metal, vapor-form~n materials, deox~dizers, and scavengers

FIGURE 31,13 The flux-cored arc welding (FCAW) process. (Courtesy of The American Weldrng Socrety, New York)

be made in all positions. Duect-current electrode-positive (DCEP) conditions ar

generated by the vaporizing flux. by combining the flux with a flow of externally supplied shielding gas, such as C O ~ 1 Table 31-4 presents a process summary of flux-cored arc welding. GAS METAL ARC WELDING If the supplemental shielding gas flowing through the torch (described above) beca the prunary protection for the arc and molten metal, there is no longer a need fo volatilizing flux. The consumable electrode can now become a continuous, solid, coated metal wire or a continuous hollow tube with powdered alloy additions id center, known as a metal-cored electrode.The resulting process, shown in Flgute3$ was formerly called metal inert-gas, or MIG, weldlng and is now known as gas me@ weldcng (GMAW).The arc is still maintained between the workpiece and the auto' ~cally fed bare-w~re electrode, which continues to provide the necessary filler rn4 ! 1

Heat source Protcctlon Electrode jo~ned Mater~al Rats of heat ~nput Weld profile (DIW) Currellr Max penatrailon
Asqets

Electr~c arc Slag and gas from Flux (optional secondary gas shield) Cont~nuous, consumable B a t for steel Med~um
<500 amps 6-10 mm Continuous electrode Requuei slag removal

LlmtatlOnS

Nozzle

Molten weld metal

\ Base metal

Electrode diameters range from 0.6 to 6.4 mm.The welding current, penetration depth, and process cost are a l l simdar to the flux-cored process. Because shielding is provided by the flow of gas,and fluxing and slag-forming agents are no longer required, the gas metal arc process can be applied to all metals. Argon, helium, and mixtures of the two are the primary shieldmg gases. When welding steel, some 0, or COzisusually added to improve the arc stability and reduce weld spatter. The cheaper CO, can also be used alone when welding steel, provided that a deoxidizing electrode wire is employed. Nitrogen and hydrogenmay also be added to modify arc characteristics. Since these shielding gases only provide proteetion and do not remove existing contamination, starting cleanlinessis critical to the production of a good weld. The specific shielding gases can have considerable effect on the stability of the arc, the metal transfer from the electrode to the work, and also the heat transfer behavlor, penetration, and tendency for undercutting (weld pool extending laterally beneath the surface of the base metal). Helium produces The hottest arc and deepest penetration. Argon is intermediate, and C 0 2 yields the lowest arc temperatures and shaIlowest penetrations. Since argon is heavier than air, it tends to blanket the weld area, enabling the use of low gas flow rates The lighter-than-air helium generally requires higher flow rates than either argon or carbon dioxide. Electromccontro~s can be used to alter the welding current, enabling further control of the metal transfer mechanism, shown previously m F~gure 31-9. Short-circurt transfer (GMAW-S) is promoted by the lowest currents and voltages (14 to 21 volts) and the use of CO, shield gas The advancing electrode makes direct contact w ~ t h the weld pool, and the short circuit causes a rapid rise in cutrent. Big molten globs form on the tip of the electrode and then separate, forming a gap between the electrode and workpiece. This gap reinitiates a brief penod of arcing, but the rate of electrode advancement exceeds the rate of melting in the arc, and another short circuit o c c m The power conditions oscillate between arcing and ?hart circuiting at a rate of 20 to 200 cycles per second. Short-circuit transfer is preferred when joining thin materials andcan be used in all weldmgpositions,but it suffers from a higli degree of spatter. If the voltage and amperage are increased, the mode becomes one of globular tramfer.The electrode melts from the heat of the arc, andmetal drops form, with a diameter approximately equal to that of the electrode wire. Gravity and electromagn~tic forces then transfer the drops to the workpiece at a rate of several per second. Since gravity plays a role in metal transfer, there is a definite limitation on the positions of welding. With even higher currents and voltages (25 to 32 volts and about 200 amps), argon gas shielding, and DCEP conditions, spray transfer (GMAW-ST) occurs. Small droplets emerge from a pointed electrode at a rate of hundreds per second.Because of their small size and the greater electromagneticeffects, the droplets are easily propelled across the arc in any direction,irrespective of the effects of gravity. Spray transfer is accompanied by deep penetration and low spatter.The biggest problem with out-of-position welding may be keeplug the rather large molten weld pool in place untiI it solidifies. Pukecispray lransfr (GMAW-P) was invented in the 1960sto overcome some ofthe lnnibtions of conventionalspray transfer. In this mode, a low welding current i s first used

to create a molten globule on the end of the fdler wire.A burst of high current plodes" the globule and transfers the metal across the arc m the form of a spray nating low and high currents at a rate of 60 to 600 tlmes per second, the fille transferred in a succession of rap~d bursts, similar to the emlsslons of a rap~dly aerosol atomizer. W ~ t h the pulsed form of deposition, there is less heat input to and the weld temperatures are reduced.Thinner material can be welded, d duced, workpiece discoloration is minimized, heat-senutive parts can be we1 duchvlty metals can bejoined, electrode hfe is extended, electrode cooling tec not be required, and fine microstructures are produced in the weld pool. We required topmduce a weld is lower than with other methods (re Controls can be adjusted to alter the shape of the weld pool and vary the pen In general, the gas metal arc process is fast and economical and curren for over half of all weld metal deposition.There is no frequenl change of elec with the shieldedmetal arc process. No flux is required, and no slag forms over Thus, multiple-pass welds can be made without tbe need for intermediate process can be readily automated, and the lightweight, compact weldingunit I to robotic manipulation. A direct-current electrode-positive (DCEP) arc is used because of its deep penetration,spray transfer, and the ability to pr welds with good profile. Process variables include type of current, current shielding gas, electrode diameter, electrode composition, electrode stickout beyond the gun), welding speed, welding voltage, and arc length. Table 3 process summary for gas metal arc welding. I n a process modification known as advanced gas metal arc welding (AG a second power source is used to preheat the filler wire before it emerges fr welding torch. Less arc heating is needed to produce a weld, so less base metal ed, producing less dilution of the f i e r metal and less penetration. Another recent modification is the use of flat electrode wire, typi rectangular cross section of about 4 mm by 0.5 mm. By having a larger surface ar ticipating in the arc, deposition rate is similar to a two-wire feed with o delivery system. The arc is also asymmetric. Orienting the wire perpe weldseam produces a wide, shallow weld pool, suitable for bridging gaps inating the need to weave during dep0sition.A narrower, deeper weld pool results the wire is parallel to the we1d.Varying the angle between parallel and perpendicul erates a spectrum of weld-pool geometries.
gy or

SUBMERGED ARC WELDING No shielding gas is used in the submerged arc welding (SAW) proc Figure 31 -15. Instead, a thick layer of granular flux is deposited just ahead of a s wire consumable electrode, and the arc is maintained beneath the blanket of flux a few small flames beingvisib1e.A portion of the flux melts and acts to remove from the rather large pool of molten metal, while the unmelted excess provides

.. -.. . &w@ga+w) . , . .
XII.,
-&"-".~.-'~,

rs.;LII,

I ) .
=

, ,

Heat source Protection Electrode Material joined Kate of heat input Weld profile (DIW) Currant Max. penetration Asneth Limitations

Electric arc Externally supplied shielding gas Continuous,consumabic All common metals Medium 1 c500 amps 6-lOmm No slag to remove More costly equipment than SMAW or FCAW

"ex-

( niter.

~eial is
beezed ,weld,

1I

Welding electrode I

Solidified flux

Flux tube

Vee if required

Plate metal Granulated flux Workpiece

if required

Molten

Granular

Direction of

FIGURE 31-15 (Top) Basic features of submerged arc welding (SAW). (Courtesy of Linde Division, Union Carbide Corporation, Houston, TX) (Bottom) Cutaway schematic of submerged arc welding. (Courtesy of Americon Iron ond Steel institute, Woshington, DC.)

weld metal shie1ding.De molten flux solidifies into a glasslike covering over the weld.?his layer, along with the flux that is not melted, provides good thermal insulation. The slow cooling of the weld metal helps to produce soft, ductile welds. Upon further cooing, the solidified flux cracks loose from the weld (due to the differential thermal contraction) and is easily removed. The unmelted granular flux k recovered by a vacuum system and reused. Submerged arc welding is most suitable for making flat-butt or fillet welds in lowcarbon steels (0.3% carbon). With some preheat and postheat precautions, mediumcarbon and alloy steels and some cast irons, stainless steels, copper alloys, and nickel alloys can also be welded.The process is not recommended for high-carbon steels, tool steels, aluminum, magnesium, titanium. lead, or zinc.The reasons for this incompatibility are somewhat varied, including the unavailability of suitable fluxes, reactivity at high temperatures, and low sublimation temperatures. Because the arc is totally submerged, high welding currents can be used (600 to 2000 A). High welding speeds, high deposition rates, deep penetration, and high cleanliness (due to the flux action) are all characteristic of submerged arc we1ding.A welding speed of 0.75 mlmin in 2.5-cm-thick steel plate is typical. Single-pass welds can be made with penetrations up to 2.5 cm (1.0 in.), and greater thicknesses can be joined by multiple passes. Because the metal is deposited in fewer passes than with alternative processes, there is less possibility of entrapped slag or voids, and weld quality is further enhanced. For even higher deposition rates, multiple electrode wires can be employed. Limitations to the process include the need for extensive flux handling, possible contamination of the flux by moisture (leading to porosity in the weld), the large volume of slag that must be removed, and shrinkage problems due to the large weld pool. The high heat inputs can produce large grain size structures, and the slow cooling rate

CHAPTER 31

Gas

FIGURE 31-16

Diagram of a

stud weld~ng gun. (Courtesy of American Mochinnt )

FIGURE 31-17 (Left) Types of studs usedfor stud weldlng CCenter) Stud and ceramic ferrule.
(Right) Stud after weid~ng and a section through a welded stud. (Courtesy ofNelson Stud Wld~ng Co, Elyr~a,OH)

the stud and the workpiece.After a small amount of metal is melted.the two pieces are brought together under light pressure and allowed to solidify. Automatic equipment controls the arc, its duration, and the application or pressure to the stud. Figure 31-17 shows some of the wide variety of sluds that are specially made for this process. Many contain a recessed end that is filled with flux. A ceramic ferrule,such as the one shown in the center photo of Figure 31-17,may be placed over the end of the stud before it is positioned in the gun. During the arc, the ferrule serves to concentrate the heat and isolate the hot metal lrom the atmosphere. It also confines the molten or softened metal and shapes it around the base of the stud, as shown in the photo on the right of Figure 31-17.After the weld has cooled, the brittle ceramic is broken free and removed. Since burn-off or melting reduces the length of the stud, the original dimensions should be selected to compensate. Stud welding requires almost no skill on the part of the operator. Once the stud and ferrule are placed in the gun and the gun positioned on the work, all the operator has to do is pull the trigger.The cycle is executed automatically and takes less than one second.Thus the process is well suited to manufacturing and can be used to eliminate the drilling and tapping of many special holes. Production-type stud welders can produce over 1000 welds per hour.

1 31.6 NONCONSUMABLE-ELECTRODE ARC WELDING

GAS TUNGSTEN ARC WELDING Gas tungsten arc weldmg (GTAWI was formerly known a7 tungsten inert-gas (TIG) weldmg. or Hellarc weldlng when helium was the sbield~og gaa A nonwncunaabl. tunp sten electrode provides the arc but not the filler metal Inert gas (argon, helium, or a mixtute of them) flowsthrough the electrode bolder to provide a protective shield around the electrode, the arc, the pool of molten metal, and the adjacent heated areas (Note: CO, cannot be used m this process slnce it provides inadequate protection for the hot tungsten electrode ) While argon is the most widely used gas, and produces a smoother, more stable arc,helium may be added to increase the heat Input (h~gher weld~ng speeds and deeper penetration) Helium alone may be preferred for overhead weldmg since it 1s lighter than air and flows upward. The composition,diameter,length,and tip geometry jhalled,pointed, or truncated cone) of the tungsten or tungsten-alloy electrode are selected based on the material being welded, the thickness of the material, and the type of current b e ~ n g nsed.The

Welding torch used in nonconsumableelectrode, gas tungsten arc welding (.GTAW), showing feed lines for power, cooling water, and inert-gasflow. (Courtesy of Linde Division, Union Carbide Corporation, Houston, TX)
FIGURE 31-18

behind thistube

out

Electrode

Wi I

tungsten is often alloyed with thorium oxide, zirconium oxide, cerium oxide, c thanum oxide lo provide better current-carrying and electron-emission characte and longer electrode life. Since tungsten is not consumed at the temperatures of tl the arc length remains constant, and thc arc is very stable and easy to mai Figure 31 -18 shows a typical GTAW torch with cables and passages for gas flow,^ and cooling water. In applications where there is a close lit between the pieces being joined,nc mctal may be needed. When filler metal is required, it is usually supplied as a se] rod, about a meter (3 i t ) long and in various diameters, as illustrated in Figure 31-1 filler metal is generally selected to match the chemistry andlor tensile strength mctal being welded. When high deposition rates are dcsired, a separate resistance h circuit can bc provided lo preheat the filler wire.As shown in Figure 31-20, the d tion rate of heated wire can be several times that of a cold wire. By oscillating tht wire from side to side while making a weld pass, the deposition rate can be f l increased.The hot-wire process is not practical when welding copper or aluminum ever, because it is difficult to preheat the low-resistivity filler wire. With skilled operators, gas tungsten arc welding can produce high-quality that are very clean and scarcely visible. Since no flux is employed, no special clr or slag removal is required. However, the surfaces to b e welded must be clean an Tungsten electrode
I

Welding torch

,Shielding gas

Filler

D~agramof gas tungsten arc weld~ng(GTAW) (Courtesy of Amencan Iron and Steel Inst~tute,Wash~ngton, DC.)
FIGURE 31-19

Base metal

20

Hot wire w ~ t h oscrllat~on

14

m
c
0

12

Hot wlre

10 8
4

Cold w~re

4 Arc

energy, KW.

FIGURE 31-20 Comparison of the metal deposition rates in GTAW with cold, hot, and oscillating-hot filler wire (Courtesy of Welding journal.)

of oil, grease, paint, and rust, because the inert gas does not provide any cleaning or fluxing action. It is also important to control the arc length throughout the process. Since the arc is somewhat bell-shaped, decreasing the standoff distance will decrease the melt and heat-affected widths on the workpiece. However, if the hot tungsten electrode comes into contact with the workpiece or molten pool, it will contaminate the electrode. All metals and alloys can be welded by this process, and the use of inert gas makes it particularly attractive for the reactive metals, such as aluminum, magnesium, and titanium, as well as the hi@-temperature refractory metals. klaximum penetration is obtained with direct-current electrode-negative (DCEN) conditions, although alternating current may be specified to break up surface oxides (as when welding aluminum). DCEP or reverse-polarity conditions are used only when welding thin pieces where shallow penetrations are desired. Weld currents should be kept low, since this mode tends to melt the tungsten electrode. Weld voltage is typically 20 to 40 volts, and weld current varies from less than 125 amps for DCEP to 1000 amps for DCEN. A high-frequency, high-voltage, alternating current is often superimposed on the regular AC or DC welding current to make it easier to start and maintain the arc. The pulsed arc gas tungsten arc welding (GTAW-P) modification oflers all of the advantages previously cited for pulsed gas metal arc welding, including the ability to weld thinner materials due to the lower heat input and lower temperatures. GTAW costsmore than SMAW and is slower than GMAW.However, it produces a high-quality weld in a very wide range of thicknesses, positions, and geometries.The process has a medium rate of heat input and the welds have a depth that is approximately equal to the width.The materials being welded are generally thinner than 6.5 mm ('/,in.). Table 31-7 provides a process summary.

t,. f i c; ... , : .; .<: : :,:li4-~es.%&6aryi .. . Arc o Welding (CTAW) ,. .. C& ~ " n ~ i t c

,

, ,

.. ...

..

...

Heat source

Protection

Etcnrode Material jolned Rate of hedl lnput

Weld prohle (DIW) Current

Elennc arc Exto~nally supplied shjeldlng gas Nonconsumable All common metals Medium
1
<500 amp9

Max penetration Assets Llrnitaoons

3 mrn High-qual~ty welds, no slag to be removed Sloww than mosumnhle electrode GMAW

862

CHA~E 31 R Gas Flame and Arc Processes

Process schematic of spot welding by the inert-gas-shielded tungsten arc process.

FIGURE 31-21

GAS TUNGSTEN ARC SPOT WELDING A variation of gas tungsten arc welding can be used to produce spot welds betwee1 pieces of metal where access is limited to one side of the joint or where thin shc being attached to heavier material.The basic procedure is tllustrated in Figures : and 31-22 A modified tungsten inert-gas gun is used with a vented nozzle on the The nozzle is pressed firmly against the material, holding the pieces in reasonably contact. (The workpieces must be sufficiently rigid to sustain the contact pressure.) gas, usually argon or helium, flows through the nozzle to provide a shielding atmosp Automatic contsols then advance the electrode to initiate the arc and retract it tt correct distance for stabilized arcing. The duration of arcing is timed automatics produce an acceptable spot weld.The depth and size of the weld nugget are conh by the amperage, time, and type of shielding gas In arc spot weldmg, the weld nugget begins to form at the surface where thc makes contact.This is in contrast to the more standardresistance spot-weldmg met where the weld nugget forms at the interface between the two members. Each tech has its characteristic advantages and disadvantages.

PLASMA ARC WELDING

In plasma arc welding (PAW) the arc is maintained between a nonconsumable elecl and either the welding gun (nontran~ferredarc) or the workp~ece (transferred arc), lustrated m Figure 31-21. The nonconsumable tungsten electrode is set back withi "torch" in such a way as to force the arc to pass through or be contained wlthin as diameter norzle.An inert gas (usually argon) ISforced through this constricted arc,* it is heated to a high temperature and forms a hot,fast-movingplasma The emergin then transfers its heat to the workpiece and melts the metal.This flow is called th~ fice gns. A second flow of inert gas surrounds the plasma column and provides d mg to the weld pool. When filler metal is needed, it 1s provided by an external fee, Figure 31-24 presents a comparison of the nonconstricted arc of the G' process and the constricted arc of plasma arc welding, and shows the differenc temperature distribution. Plasma arc welding is characterized by a high rate of input and temperatures on the order of 16,500C (30,00OoF).This in turn offer welding speeds, narrow welds with deep penetration (a depth-to-width ratio of abo a narrow heat-affected zone, reduced distortion, and a process that is insenslti variations in arc length since the plasma column is cylindrical.Welds can be made pos~tlons, and nearly all metals and alloys can he welded.

WGURE 31-22 Making a spot weld by the inert-gas-sh~elded tungsten arc process. (Courtesy of A I Reducbon ~ Company inc., New York, NY)

FIGURE 31-23 Two types of plasma arc torches. (Left) Transferred arc; (r~ght) nontransferred arc.

Transferred arc

Nontransferred
arc

Torch body

Constricted arc (plasma arc welding) Temperature, K.

010,000 to 14,000
14,000 to 18,000

18.000 to 24,000 24,000 and up

FIGURE 31-24 Comparison of the nonconstr~cted arc of gas tungsten arc weld~ng and the constricted arc S M InternaDonal, of the plasma arc process Note the level and dlstr~but~on of temperature. (Courtesy A Matenols Park, OH.)
W~th a low-pressure plasma and currents between 20 and 100 amps, the metal simply melts and flows Into the jornt. At higher pressures and currents m excess of 100 amps a "keyhole" effect occurs m which the plasma gas creates a hole completely through the sheet (up to 20 mm thlck) that 1s surrounded by molten metal As the torch is moved, 11qu1dmetal flows to fill the keyhole. If the gas pressure is increased even further, the molten metal 1 s expelled from the reglan and the process becomes one of plasma cutting, whlch is discussedlater m thls chapter. Many plasma torches employ a smal1,nontransferred arc within the torch to heat the orifice gas and ionize it The iomzed gas then forms a good conductive path for the main transferred arc.This dual-arc technique permits instant ignition of a low-current

. .. . A& ~ e ~ h i(PAW) &

Heat source Protection Electrode Mdterral jo~ned Rate of heat rnput Weld prohle (DIW) Current Max penetralrou Asxts Limitations
P l a s m arc Externally supplied shielding gas Noncon~umable All common metals

HI& 3
4 0 0 amps 12-18 mm Can have long arc length High initial equipment cost, large torches may limit accessibilitv

arc. which is more stable than that of an ordinary plasma torch. Separate D C p m

POWER SOURCES FOR ARC WELDING Arc weld~ng requires large electrical currents, often in the range of 100to 1000 amps.7 voltage is usually between 20 and 50 volts Both D C and AC power suppltes are av;

In the past, most direct current units were gasoline- or diesel-powered mot generator set$ and these are still used when welding is to be performed in remote locatio Most welding today, however, uses solid-state transformer-rectifier machines, such the one shown in Elgure 31-26. Oaeratine on a three-phase electrical line, these machi

nected to the various phases of a three-phase supply to help balance the load.

Welding current (amps)

Welding current (amps)

FIGURE 31-26 Rect~fier-type AC and DC weld~ng power supply (Courtesy of L~ncoln Electnc Company, Cleveland, OH)

Inverter-based power supplies, introduced in thc 1980s. prov~de great flex~bllity. Through solid-state electron~cs, these AC mach~nes can quickly modify the shape and frequency of the pulse wavelorm or momentarily change the power output. The squarewave technology currently belng employed prov~des improved arc starts and more stable arcs. The percent of time in electrode negatlve or electrode positive can be adjusted, dnd output frequency can be vaned from 20 to 250 Hz.Through feedback and loge control, the power supply can actually adjust to compensate for changes in a number of process variables.

JIGS, POSITIONERS, AND ROBOTS

Jigs or fixtures (also calledpositconers) are frequently used to hold the work in production welding. By positioning and manipulating the workpiece, the welding operations can often be performed in a more favorable orientation. Parts can also be mounted o n numerically controlled (NC) tables that position them with respect to the welding tool. Industrial robots have replaced humans for many welding applications. They can operate in hostile environments and are capable of producing high-quality welds in a repetitive mode.

Whlle oxygen torch cutting has been d~scussed earl~er in this chapter, and laser cuttlng will be covered in a future chapter, there are a number of arc-cutting methods.Virtua1ly all metals can be cut by some form of electnc arc. In these processes the material is melted by the intense heat of the arc and then permitted, or forced, to flow away from the reglon of the sllt or notch (kerf). Most of the techniques are simply adaptations of

;as Fla me and Arc Processes the arc-welding procedures discussed m this chapter. Each has its inherent characte tics and capabilities, including tolerance,thickness capability,kerf width, size of the heat-affected zone, and cost. Selection depends upon factors such as t requirements, the subsequent processes that will be performed on the cut part, end use of the product

CARBON ARC AND SHIELDED METAL ARC CUTTING The carbon arc cutting (CAC) and shielded metal arc from a carbon or shielded metal arc electrode to from the cut by gravity or the force of the arc itself.These processes are generally ed to small shops, garages, and homes, where there is limited investm AIR CARBON ARC CUlTING In air carbon arc cuttmg, the arcis again maintained between a carbon electrode and workpiece, but high-velociry jets of air are directed at the molten metal horn holes in electrode holder. While there is some oxidation, the prlrnary function of the air blow the molten material from the cut. A u carbon arc cutting 1s particularly effectiv cuttlng cast Iron and preparing steel plates for welding. Speeds up to 0.6 mlmin are rlble, but the process is quite noisy, and hot metal particles tend to be blown over a stantial area. OXYGEN ARC CUlTING In oxygen arc cutting (AOC), an electric arc and make the cut.The electrode is a coated ferrous-m establish a stable arc,while oxygen flows through the bore and is directed on the are incandescence With easily oxidized metals, such as steel, the arc preheats the base me which then reacts with oxygen, becomes liquef~ed,and is expelled by the oxygen stre GAS METAL ARC CUTTING If the wire feed rate and other variables of gas metal arc weldmg (G ed so that the electrode penetrates completely through the workplece, cutting ra than weld~ng will occur, and the process becomes gas metal arc cutting (GMAC) wire feed rate controls the quality of the cut, and the voltage determines the wid the slit or kerf. GAS TUNGSTEN ARC CUTTING Gas tungsten arc cutting (GTAC) employs the same basic circuit and shielding g used 1n gas tungsten arc welding, wlth a high-velocity jet of gas added to expel molten metal. PLASMA ARC CUlTING The torches used in plasma arc cutting (PAC) produce the highest temperatures from any practical source.With the nontransferred type of torch, the ar pletely withtn the nozzle, and a temperature of about 16,500C (30, With the transfer-type torch, the arc is maintained between the electrode and piece, and temperatures can be as hlgh as 35,000"G (60,000" these temperatures and near supersonic speeds are capa electrically conductive material simply by melting it and blowi Early efforts to employ this technique showed that the speed, eratlng cost were far superior to those of the oxyfuel cuttrng met early systems could not constrict the arc sufficiently to produce the qual' to meet the demands of manufacturing. Therefore, plasma arc cuttin limited to those materials that could not he cut by the oxidation

FIGURE 31-27 Cutting sheet a plasma torch. metal w~th (Courfesy o f GTE Sylwm~u,

techniques. In the 1970s radial impingement of water on the arc was found to produce the desired constriction. It provldes an intense, highly focused source of heat,and waterinjected torches cannow cut virtually any metal in any position.Mapetic fields have also been used to constrict the arc and can produce high-quality cuts without the need for water impingement. Compared to oxyfuel cutting, plasma cutting is more economical (cost per cut is a fractlon of oxyfuel), more versatile (can cut all metals as easily as mlld steel), and much faster (typically, five to eight tlmes faster than oxytuel). Cutting speeds up to 7.5 d m i n have been obtained in 6-mm-thlck alummum and up to 2.5 mlmin in 12.5-mm-thick steel.The combination of the extremely high temperatures and jetlike action of the plasma produces narrow kerfs and remarkably smooth surfaces. nearly as smooth as can be 2' of vertical, and surface axobtalned by sawing.Plasma-cut surfaces are often with~n idation is nearly eliminated by the cooling effect of the water spray. In addition, the heat-affected zone in the metal is only one-third to one-fourth as large as that produced by oxyfuel cutting, and a preheat cycle is not requiredm the cutting of steel. Heat-related distortion is extremely small. Transferred-arc torches are usually used for cutting metals, while the nontransferred type are employed with the low-conductivity nonmetals Ordinary air or inexpensive nitrogen can be used as the plasma gas for the cutting of all types of metal. Oxygen plasma systems were introduced in the 1980s and are used on carbon and lowalloy steel products with thicknesses rangmg from 2 to 32 mm (up to lit4in.). When cutting thick sections (greater than 12 mm), an argon-hydrogen mixturemay be preferred to provide a deeper-penetrating arc. A secondary flow of shielding gas (nitrogen, air, or carbon dioxide) may be used to help cool the torch, blow the molten metal away, shield the arc, and prevent oxidation of the cut surface. The arc-constricting water flow can also serve as a shleIdiig medmm. During the 1990s hzgh-density, or preciriolt plasma, systems began to appear. Various designs are used to restrict the orifice (i e., superconstrict the plasma), producing vertical edges (less than 1" taper), close tolerances ('I3 that of conventional), and drossfree plasma cutting of thin materials.The lower-amperage torches (10 to 100 amps) are limited to cutting carbon and low-alloy steels less than 16 mm ( 1 ' , in.) thick and higherperformance metals (such as stainless and high-strength steels, nickel alloys, titanium, and aluminum) less than 12 mm ('I, in.) thick. In addition, the cutting speeds are slower than conventional plasma cuttmg,but there 1s no reduction in the size of the heat-affected zone Pulsedplasma arc cutting,another recent development, can reduce heat input to the workpiece while producrng kerfs that are 50% narrower and cleaner edges on the cuts. Combining a plasma torch with CNC mampulation can provide fast, clean, and accurate cutting, hke that shown in Figure 31-27. In this process, the cutting table may be placed underwater as a means of reducing noise, air pollution, dust, and arc glare (dyes are placed in the water). Plasma arc torches can also be incorporated into punch presses to provide a manufacturing machine with outstandrng flexibility in producing cut and punched products from a variety of materials. Plasma arc cutting IS also suitable for robot application. A single robot system can be used for both cutting and welding of intricate shapes and contours. Water constriction of the manipulated arc is a problem, however, making it important to select the right process parameters and type of gas for the particular application. Table 31-9 compares the features of oxyfuel cutting, plasma arc cutting, and laser cutting.

31.9 METALLURGIC
When used for cutting, the flame and arc processes expose materials to high localized temperatures and can produce harmfulmetallurgical effects.If the cut edges will be suhsequently welded, or if they will be removed by machining, there is little cause for concern.When the edges are retained in the finished product, consideration should be given to the effects of cutting heat and then Interaction with the applied loads. In some cases, additional steps may be required to avoid or overcome harmful consequences.

'lame and Arc Processes

$wf .. %T #M . . ~> % . ' .; # ~ . i j.. mAJ< .i ..and . taser

te.11ure
I'rcli.rrr.d
- .

2"

."""

I
" '

O\?fu:l

('UII~II~

I'Iu$n~.t Ar. C I I I I I I ~ ~ metals S~mlar to oxyEu~l

A h s t as good as laser on thm material

I.:iscr Cuili~t$ Good quahty-bsst plate materrel Im Fod to 1inch

ma1er1aIs <I cut Qual~ty Thidrness range

1 Ste~t 2. Stamless

Avsiage

3 1 1 6 tnch lo unhrmted not used

3.Alummum
Cutt~ng speed

not used Long preheat is requtrcd

26 ga. to 3 loch 26 ga to 5 ~ n e h 22 ga. to 6 ~nch Fast cuttmg

or time

For carbon steels with less than 0.25% carbon, thermal cutting does not pr serious metallurgcal effects,However, m steels of higher carbon content, the me gical changes can be quite significant, and preheating andlor postheating may quired. For alloy steels, additional consideration should be glven to the effects various alloy elements. Because of the low rate of heat input, oxyacetyl large heat-affected zone.The arc-cutting methods produce quite s~milar to those of arc welding. Plasma arccutting is s calired, that the origrnal properties of a mela1 are only mo All of the thermal cutting processes produce 5om face generally in tenslon. Except in the case of thin sh However, if subsequent machining removes only a portion of the cut surface, or d penetrate to a sufficient depth, the resulting Imbalance distortion. It may be necessary to remove all cut surfaces to a substantial depth to good dimensional stability. Thermal cutting can also introduce geometrical features into the edge. or arc-cut edges are rough to varying degrees and thus contain notches that stress raisers and reduce the endurance or fracture st jected to high or repeated tensile stresses, the cut su should be removed by machiningor at least subjected

111 Key Words

acetylene altcrndtlng currenl arc arc cutting arc ueldlng hulk weldmg c~rburlz~ng flame consumable electrode procebs DCEN DCEP electrode 11llermelal flame stralghtenlng
flux

flux-cored arc weldng fus~on weldlng gas metal arc weldlng gas tungsten arc welding globular transler heat-affected zone
14

kerf
MAPP

neutral flame nonconsumable-electrode process nontramferred arc orlftce gas

oxldiztng flame oxyfuel-gas cutttng oxyfuel-gas weldrng oxygen lance cuttlng penetrdtlan plasma plasma arc weld~ng posltioner power supply pulsed arc pulsed spray transfer res~dual stresses reverse polarily shtelded metal arc weld~ng

short c~rcult transfer

stack cutttng straight polarlty stud weldmg

transferred arc

Review Questions

869

II Review
Why does an ozyfuel.gas weld~ng torch usually have aflame with two dist~nct reglonsq of the maxrmum temperature m an oxy What sth he locat~on acetylene flame? What function or funNons are performed by the outer zone of the weldingflame? What three types of flames can be produced by varying the oxygen-fuel ratio? Wh~ch type of oxyfuel flame 1s most wmmonly used? What are some of the attractlve features of MAPP gas? Why mmght a welder want to change the tlp rive (or onfice diameter) In an oxyacetylene torch7 What is filler metal, and why might it be needed to produce a jo~nt? what is the role of a weldingflux? Oxyfuel-gaswelding has a low rate of heat mput. What aresome of the adverse features that result from the slow rate of heaung? What are some of the more attractlve features of the oxyfuelgas process? How does pressure gas welding differ from the oxyfuel-gas process? In what way does the torch cuttrng of ferrous metals d~ffer from cutting nonoxrdaing metals? Why ml&t ~t he possible to use only an oxygen lance to cut hot steel strands as they emerge Gum a contmuous castirrg operatmon? How does an oxyacetylene crrtung torch &fer from an oxyacetylene weldlng torch') What are some of the ways in which cuttlng torches can be man~pulaled? What modificat~on mu$! beincorporated mto a cuffmg torch to permit it to cut metal underwate?:l? If a curved plate ts t o be strazghtened by flame str~ghtemII& should the heat be apphed to the longer or shorter surface of the arc? Why? why does the flame stra~ghtenmng process not work for thin sheets of metal7 What sortsof problems plagued early attempts to develop arc
welding?

Whatare the three baslc types Of current and polarity that are used 1n arc weldIng7 What is the dlfterence between a consumable and nonwn sumable electrode7 For which processes does a filler metal have to be added by a separate mechanism? What are the three types O f metal transfer that can bmur during arc weldmg? What are some of the process vaccables that must be s p e c i f i when setting up an arc-weldiny process? What are the four primary conaumahle electrode arc-weldmg processes? What are some general properties of the consumable.electmde arc-welding pmcesses? What are someof the functions of the electrode coa~ngs used in shlelded metal arc weldmg? How are welding electrodes eommouly class&ed, and what information does the destgnanon usuaUy provide? Why are shieltled metal arc electrodes often baked just pnor to welding? What is the functron of the slag coating that f o m s over a shelded metal arc weld?

What benefit can be obtained by placing iron powder in the coating of shrelded metal arc eleclrodes that w~ll be used to weld ferrous metals? Why are sh~eldadmeral arc electrodes generally limited in length, forcing the proctss t o be one of mnterrmttent operation'? Why is the shielded metal arc-wddtng process limired to low weldmgcurrents and shallow penetration? What are some of the attracovefeatu~m of the shieldedmetal arc-weldmgprocess? What is the advantage of placing the flux in the center of an electrode (flux-cored arc weldmg) as opposed to a watmg on the outside (shielded metal arc weldtng)? What feature enables the weldmg current in F%4W to be hmgher than in %MAW7 What are some of the advantages of gas metal arc welding compared i o the shielded metal arc process? Describe the relanve performance of argon, hebum, and carbon dioxide gases m creating a hmgh-temperature areand promqtmg weld penetration. W h ~ of h the metal transfer mechanisms n most used m arc welding? Describe the metal transfer that occurs during pulsed arc gas metal arc welding. Whatare some of thebenefits that can be obtained by thereduced heating of the pulsed arc process? What are Some of the prrmary process uan&bies in the gas metal arc-weldmg process? What benefits can be gamed by using a rectangular cross-seeflon electrode wae as opposed to a t m d one? What are some of the iunctions of the flux in submerged arc welding? What are some of the attractive features of submerged arc welding? Major litrltions? What is the primary goal or objective m bulk welding? of stud welding? What IS the primary ob~ectlve What IS the funct~on of the ceramic ferrule placed over the end of the stud m stud weldins? What is thecurrent (or proper) designation fosMIG wezdmg? T I 0 welding? Heliarc welding? SO. What types of shieIding gases are used in the gas tungsten arc process? 51 What can be done to increase the rate of filler metal deposltion duringgas tuugslen arc welding? 52. What are some of the attractwe features of gas tungsten arc welding? 53. How are h e spat -++el&produced by g a s ~ n g s t m arcspot weldIn$ dmfferentfmm those made by convent~onal resrstance spot
&.

Ho!! IS [he hc.~linguf lh:. \\orkpic~e dur~nz [,l.~.;nla d c wcldlox dutzr~nl ttom the hcet~ng~n utller arc-wcld~i~tcchr~iqucs~ M'h.it :Ire the 1u.o dilfcr.-nl pa:, l1,rus lo pl.tsrna .qrc ucl.li,~~! What are sorneof the attractwe features of plasma arc weid~ng? what IS the primary difference between plasma arc weld~ng and plasma arc cutbng? What are the attractwe features or benefits of an inverterbased power supply? What 1s the kerfin thennal cutting operations? What is the purpose of the oxygen i n oxygen arc cutting?

870

C a m 31 Gas Flame and Arc Processes

61 Why is plasma arc cuttmg irn attn.eri~eway of cuttrng hi&aelr~ng-pomt matenalc.7 62. Whet Lcchnrquesran be used t o constrict the arc m plasma arc cutting, pmducinganarrower, tllorecontrolled cuf? 63. Compared t ooxyfael cutting, what are some af the attractive fcntures of the plasma tmhnique? 64. Dessrib the relative size of the heat-affected BOW for the various cutling processes oxyfuel, arc, and plasma

$5. Why might tha residual stmses lndumd durmg cuthng o

ations beabjectiaahle?

66. Why m1&1 rt be wise to machlne away the thermally and heat-affectedzonc of metal that wdl be usedma
machme part?

Bicycle Frame Construction aad Repair

s a new employee in a b~cyde shop, you learn that
customers will frequently seek advice regarding varirepair. Benf or broken frames appear o m typesof b~cycle to be a common problem, and you learn that inexpensive bieydes generally have frames made from welded low-oarban steel tublng. Th~s material oan be heated to straghten and iseasily repar welded using a wrde varrety of welding not quiTe as simple. m dfracture? Waa the weld in any way defective? Was the second failure relatedto the welding repar? 2. If the tubing had been strengthened by an age-hardenrng hsat treatment, could the same results hav9 occurred? Explain is there a means of repairing the original fratture? W a t would have been your recommmdatlon?

aluminum, and the repair seemed to be of good quattty, Shortly thereafler, however, the frame broke agaln. This time the fracture was adjacent to the repar weld and the characteristics of the breakwere different. Whlle the f~rst fraotwe was somewhat brittle m miure, the second ape more ductile, with evidence of metal fbw pearedto b failure was not related to the attempted repair and that the

assemble the frame? d, Composde materials can be used to produce tailored

relnfarced tubing, such as graphite fiber-ra~nfofcd epoxy, how would you join the assembly?

RESISTANCE AND SOLID-STATE

32.1 INTRODUC~ON 322 T~IEURY OF RESISTANCJZ W~mmo OF RESrSTANcE WELDING Heating 32.5 SOLID-STAE WELDIMG Pressure PROCES~ES Current and Current Control Forge Welding Power Supply ForgeSeam Welding WEtDlNG ~ O C E S S E S 32.3 RESISTANCE Cold Welding Resistance Spot Welding Reststance Seam Welding Projection Weldmg Am GE LIMITATIONS S 32.4 A D V ~ ~ ~ T A Roll Welding or Roll Bonding Friction Welding and Inertla Weldlng Friction-Stir Welding
Ultrasonic Welding

D~ffus~on Welding Explosive Welding REPAIRTO Case Study. FIELD A POWER TRANSFORMER

As indicated in the lists of Figures 30-1 and 30-2, there are a number of welding and cutting processes that u t b e heat sources other than oxyfuel flames and electric arcs, and some use no heat source at all. We begin this chapter w ~ t h a group of prucesses that use electrical resistance heatlng to form the joint. A second group of processeb, known as solid-state welding proclsses, create joints without any mclting of the workpiece br filler matenal.

1 32.2 THEORY OF RESISTANCE WELDING

In reszstance welding, heat and pressure are combined to induce coalescence. ElecPodes are placed in contact with the material, and electrial resistance heamg is used to raise the temperature of the workpieces and the interface between them.The same electrodes that supply the current also apply the presure, which 1s usually vaned throughout the weld cycle. A certain amount of pressure is applied init~ally to hold the wor*kpiecesin contact and thereby control the electrical resistance at thc i n t e r f a . When the proper temperature has been attained, thepressure is increased toinduce coalescence. Because pressure is uhlized, coalescence occurs at a lower temperature than that required for oxyfuel gas or arc welding. In fact, melting of the base metal does not occur in many resislance-welding operations Res~stance-welding processes might well he considered as a form of solid-statc welding, although they are not officially classified as such by the Amer~can Welding Soc~ety. In some resistance-weldmg processcq additional pressure a applied immediately after coalescence to provide a certain amount of forging action Accompanying the deformation is a certain amount of grain refinement. Additional heahng can also be employed after welding to promde tempering and/or stress relief. The required temperature can often be attained, and coalescencecan be achieved, in a few seconds or less. Resistance welding, therefore, is a very rapid and economical process, extremely we11 suited to automated mannfacturing No filler nlctal is requcred, and the tight contact malnthed between the workpieces excludes air and eliminate? the need for fluxes or shleldiig gases

nce and Solid-State Weld~ng Processes

HEATING
The heat for resistance weld~ng is obta~ned by passang a large electrical current thmu the workpieces for a short per~od of time The amount of heat Input can be determ~n by the basic relationship: H = 12Rt where: H = total heat input ln joules

I = current in amperes
R
= electrical resistance of

the circuit in ohms

r = length of time during w h ~ current h is flowing in seconds

It is important to note thatthe workpieces actually form part of the electrical e wit, as illustratedin Figure 32-1, and that the total resistance between the electrodesco sksts of three distinct components:

1. The bulk resistance of the electrodes and warkpieces-the upper electrode, upp workpiece, lower workpiece and lower electrode 2. The contact resistance between the electrodes and the workpiecesbetween upper electrode and upper workpiece and the lower electrode and lower workpre 3. The reslsfance between the surfaces to be joined, known as the fayrngsurfaces
Since the maximum amount of heat is generated at the point of rnaxlmum res tance,itis desirable to have this be the location whe~e. the weld is to be made.Therefor it 1 s essential to keep components 1and 2 as low as possible with respect to resistam The bulk resistance of the electrodes is always quite low, and that of the workpieces determined by the type and thickness of the metal being joined. Bemuse of the lar areas involved and the relat~vely high electrical conductivity of most metals, the w o piece resistances are usually much less than the contact or interface values Resistant (the resistance between the electrodes and workpieces) can be minimized by us electrode materials that are excellent electrical conductors and by controlling the and shape of the electrodes and the applied pressure. Any change 1n the pressure tween the electrodes and workpieces, however, also affects the conract befween tb faying surfaces. Therefore, only limited control of the electrode-to-work resistance be obtained by pressure vanation. The final resistance, that between the fayingsurfaceq is a function of (1)the ity (surface fimsh or roughness) of the surfaces; (2) the presence of nonconductive dirt,or other contaminants; (3)the pressure;and (4) the contact area.These factors all be controlled if d o r m resstance welds are to be produced. As indicated in Figure 32-2, the objectxve of resistance welding is to bring both the faying surfaces to the proper temperature, while simultaneously keep~ng the r ma~ning material and the electrodes relatively cool. Water cooling is usually used keep the electrode temperature low and thereby extend their useful life.

Pressure
cylinder

......................

I I I

1
I

Transformer Electrodes
FIGURE 32-1

The basic

resistance welding circuit.

--___-_._ 1

I FIGURE 32-2 The des~red I

lperature dlstribut~on across electrodes and workn~eces during resistance weldir;g.

Water

Temperature

PRESSURE Because the appl~ed pressure promotes a forging action, resistance welds can be produced at lower temperatures than welds made by other processes. Controlling both the magnitude and timing of the pressure, however, is very important. If too little pressure IS used, the contact resistance will be high and surface burning or pitting of the electrodes may result. If excessive pressure is applied, molten or softened metal may be expelled from berween the faying surfaces or the electrodes may mdent the softened workpiece. Ideally, moderdte pressure should be applied to hold the workp~eces in place dnd establish proper resistance at the interface pnor to and durlng the passage of the welding current The pressure should then be increased considerably just as the proper weldmg temperature is attained.This completes the coalescence and forges the weld to produce a he-grained structure. On small, foot-operated machines, only a srngle spr~ng-controlled pressure is used. On larger, production-type welder%the pressure is generally applied through controllable air or hydraulic cylinders. CURRENT AND CURRENT CONTROL Wlnle surface condit~ons and pressure are important variables, the temperature achieved during resistance weld~ng is primarily determined by the magnitude and duration of the weldmg current.The various resistances change as current flows and the material heats. The hulk resistances of metal increase as temperature nses, and the contact resistances decrease as the metal softens and pressure improves the contact. Smce the best conditions are the inirial ones, high currents and short time intervals are generally preferred. The weld location can attain the desired temperature wh~le min~miz~ng the amount of heat generated in or diss~pated to the adjacent material. In production-type welders, the magnitude, duration, and timing of both current and pressure can he programmed to follow specified cycles Figure 32-3 shows a relatively slmple cycle for a resistance weld that mcludes both forging and postheating operations.

Current

Pressure

FIGURE 32-3 A tv~ical current

I

pressure cycle idr resistance

d~na. Thls cvcle includes

>rgingandp&heating operations.

Time

874

CHArlen 32

Resistance and Solid-State Weldlng Processes

Force

Weidelectrodes

FIGURE 32-4 The arrangement of the electrodes and workpieces in resistance spot weldlng

Force

The quality of the final weld, therefore, often depends morc on the development a proper qchedule and the subsequent setup, adjustment, and maintenance of equinmi than it does on operator skill

POWER SUPPLY S ~ n c the. e overall resistance in the welding c~rcuits can be quite low, high curreura generally requlred to produce a resrstance weld. Power translormers convert the hi voltage, low-current llne power to the hirh-currenl (un to 100.000 amus). low-volta

phasc power. These machines reduce the current demand per phase, give a balanc load, and produce excellent welds

RESISTANCE SPOT WELDING spot weld~ng (RSW)is the simplest and most w~dely used form ofresistar Re~lstunce overhp~ed materials that v weldin~, prov~dmg a fasl, economical means of join~ne
steel-bodied automobile contains hctween 2000 and 5000 spot welds. Figure 32-4 prese

FltURE 32-5 A zpot-weld

nugget between two sheets of 1.3-mm (0.05-in ) aluminum alloy The nugget 18 not symmetr~cal because the radius of the upper electrode IS greater than that of the lower electrode (Courtesy Lockheed Martin Corporntron, Bethesda, MD)

FIGURE 32-6 Tear test of a satisfactory spot weld, showing how failure occurs outside of the weld.

of the metal under the electrodes. As shown in Figure 32-6, the strength of the weld should be such that in a tens~le or tear test, the weld will remain Intact wh~le failure occurs in the heat-affected zone surrounding the nugget Souild spot welds can be obtained with excellent consistency if proper current density and timing, electrode shape, electrode pressure, and surface coud~tions are maintamed,
Spot-Weldhg Equipment. A varlety of spot-welding equipment 1s ava~lable to meet the needs of product~onoperations. For light-production work where complex current-pressure cycles are not required, a simple rocker-arm machlne is often used. The lower electrode arm is stationary,while the upper electrode, mounted on apivot arm, is brought down into contact with the work by means of a spnng-loaded foot pedal. Rocker-am machines are available with throat depths up to about 1.2 m (48 in.) and transformer capacit~es up to 50 kVa.They are used primarily on steel Larger spot welders, and those used at high production rates, are generally of the press type, as shown in Figure 32-7. On these machines, the movable electrode is controlled by an air or hydraulic cylinder, and complex pressure cycles can be programmed. Capacit~es up to 500 kVa with a 1.5-m (6Qin.) throat depth are quite common. Special-purpose press-type welders can employ multlple welding heads to make up to 200 simultaneous spot welds in less than 60 seconds Quite often, the desired products are too large to he manipulated and positioned on a welding machine. Portable spot-welding guns have been instrumental in extending the process to such applications. The guns are connected to a stauonary power supply and control unit by flexible air hoses, electrical cables, and water-cooling Zlnes.They can be used in a manualfash~on or installed on industrial robots where programmed positioning enables quality spot welds to be produced in a hlghly automatedfashion Robotic spot welding is currently the most common means of jorning sheet metal components in the automotive industry Electronic advances m the late 1980s enabled the weld~ng transformer to be integrated into the welding gun. By transforming the power ~mrnedlatelyadjacent to the area of use, the small lntegral transformer guns, or traflsguns,offer reduced power losses and enhanced process efic~ency. However, if accurate posit~oning is required in an art~culated system like an lndustnal robot, the added weight of the ~ntegral transformer may become a disadvantage Servomotors have also been incorporated into a variety of spot-welding machines to control the electrode posit~oning, speed of closure, level of appl~ed torque or pressure, and rate at which the load is applied.

32-7 Slng'e-phase, a~r-operated, press-type resistance welder wlth microprocessor control. (Courtesy Sc~oky Inc, Ch~cago,IL )

Electrodes. Resistance spot-welding electrode5 inust conduct the welding cutrent to the work, set the current density at the weld location, apply force. and help dissipate heat dur~ng the noncurrent port~onsof the welding cycle. Electr~caland thermal conductwity properties are Important considerations for electrode selection. Hot compressive strength must be sufficient to resist electrode deformat~on during the application of pressure In addition, the electrode should not melt under welding conditions and should be of a composition that does not alloy with the material belng welded-a

Alum~num Brass
coapef

x
Y

x
X
Y

x
2

x
X

Falvanlzed Iron Iron (Wrought) Monel Ntchrome N~ckeI N~ckel Silver

x
I:

x x
x

x x
x
x

x
x
X

x
x x

x
x

x
x

Steel
Tin Plate
Ztnc

k
x
X

x
X

x x
X

x
x

phenomenon that promotes sticking or galling and electrode wear. The Resista Welder Manufacturers Association (RWMA) has standardized vanous electrode geo trles and has approved a variety of electrode materials, including copper-base aU refractory metal$ and refractory-metal composrtes.
Spot-Weldable Metals and Geometries. While steel is clearly the most comr

metal that is spot welded, one of the greatest advantages of the process is that virtu all of the commercial metals can be joined, and most of them can be jomed to c other In only a few cases do the welds tend to be brittle.Table 32.1 shows some o f many combmat~ons of metals that can be successtully spot welded While spot weldmg is primarily used to jorn wrought sheet material, other fa of metal can also be welded.Sheets can be attached to rolled shapes and steelcasti as well as some types of nonferrous die castings Most metals requlre no special paration, except to be sure that the surface is free of corrosion and is not badly pit For best results, aluminum and magnesium should be cleaned immediately prior tow ing by some form of mechanical or chem~cal technique. Metals that have high elecb condumvity requre clean surfaces to ensure that the electrode-to-metal resistanceis enough for adequate temperature to be developed withm the metal itself. Silver copper are especially difficult to weld because of their h ~ g h thermal conductiv1ty.H er welding currents coupled with water coolmg ot the surrounding material ma. required if adequate welding temperatures are to be obtained. When the two pieces belng joined are of the same thickness, the practical limi spot welding is about 3 mm ('IR in.) for each sheet. Sheets of differingthickness can be joined, and thin pieces can be attached to material that is considerably thicker 1 3 mm. When metal? of dtfferent thickness or different conductivity are to be welded,l ever, a larger electrode or one with higher conductiwty is often used against the t h er or higher-resistance material to ensure that both workpieces wtll be brought ta desired temperature in a simultaneous fashion.

RESISTANCE SEAM WELDING Resisrnnce s e m welds (RSEW) can be made by two distinctly different processe the first process, sheet metal segments are joined to produce gas- or liquid-tight ves such as gas tanks mufflers, and simple heat exchangers.The weld is made between a lapping sheets of metal, and the seam is simply a sesles of overlapping spot welds,

Seam welds rrlapping spots ng. (Courresy Corpwat~on,

power

Schemat~c of the process.

Overlappkng welds

those shown in Figure 32-8 The basic equipment is the same as for spot welding, except that the electrodes now assume Lhe form of rotating disks, hke those shown schematically m Figure 32-9. As the metal passes between the electrodes, tlmed pulses of current form the overlapping welds. The timing of the wclds and the movement of the work are controlled to ensure that the welds overlap and the workpieces do not get too hot.The weldmg current is usually a bit h~gher than in conventional spot welding (to compensate for the short circuit of the adjacent weld), and the workpiece IS often cooled by a flow of air or water, In a variation of the process, a continuous current is passed through the rotating electrodes to produce a continuous seam. This form of seam weldlng is best suited for thin materials, but metals up to 6 mm ('I, in.) can be joined A typ~cal welding speed is about 2 mlmin for t h ~ n sheet. The second type of resistance seam welding, known as resutance butt welding, is used to produce butt welds between thicker metal plates.The electr~cal resistanceof the abutting metals is still used to generate heat, but high-frequency current (up to 450 kHz) is now employed to restnct the flow of current to the surfaces to be joined and their immediate surround~ngs. (Note This is slmilar to the rcsults obtained in the parallel proccss of high-frequency induct~on heating.) When the ahuttlng surfaces attain the desired temperature, they are piessed together to form a weld. Resistance butt weldmgis used extensively in themanufacture of plpes and tubes, structural as illustrated in Figure 32-10, but the process is also used to construct s~mple shapes from sections of plate. Mater~al from 0.1 mm (0.004 in.) to more than 20 mm (3/q in ) In thickness can b e welded a( speeds up to 80 mlmin (250 fpm).The combimation of high-frequency current and high welding speed produces a very narrow heat-affected zone. Almost any type or combination of metal can be welded,including difficult dissimilar metals and the high-conductivity metal$, such as alummum and copper.

PROJECTION WELDING
In a mass-production operation, conventional spot weldmg 1s plagued by two s~gnificant Imitations Because the small electrodes provide both ihe h~gh currents and the required pressure, the electrodes generally require frequent attention to maintain their geometry. In addition, the process is designed to produce only one spot weld at a tune. When increased strength is required, multlple welds are often needed, and this means

e and Sol~d-State Welding Processes

FIGURE 32-10 Using highfrequencyAC current to produce a resistance seam weld in buttwelded tub~ngArrows from the contacts ~ndicate the path of the hlgh-frequency current

multiple operations. Prolection weldtng (RPW) provides a means of overcoming th liimitahons. Figure 32-11 illustrates the principle of project~on welding A dimple 1s emboss into one of the workpieces at the locat~on where a weld is desired.The two workpie are then placed between large-area electrodes in a press machme, and pressure and c rent are applied as in spot welding. Srnce the current must flow through the polnts ofco tact (i.e ,the dimples),the heating is concentrated where the weld 1s desved.As the me heats and becomes plastic, the pressure causes the d~mple to flatten and form a we Because the projections are press formed Into the sheet, they can often be produ during previous blanking and forming operations with virtually no addit~onal cost. Mo over, the dimples or projections can be made m almost any shape-round, oval, clrcular ring-shaped-to produce welds of shapes that optlmlze a glven design. It important, however, that the shape be such that the weld will form outward from t center ot the projection Since the heat tends to develop m the piece w~th the pmjectio it isbest to inwrporate theminto the heavier of the two pieces to be ~ o ~ n or ed the met wth the higher conductivity. Multiple dmples can be incorporated into a sheet, enabling multiple welds to produced at one time. The number of projections is lim~ted only by the machine to provide the requ~red current and pressure and the need to u tr~bute both. If more than three projections are to be made at one tune, height of all project~ons should be uniform to ensure umform contact and hea An attractive feature of projection welding is the fact that it does not re cia1 equipment. Conventtonal spot-welding machines can be converted to welding simply by changing the size and shape of the electrodes. In addition,proj welding leaves no indentation mark on the exterior surfaces, a definite advantage o spat welding when good surface appearance is required. In a variation of the process,proje&on weld~ng can also be used to attach bolts nuts to other metal pnrta Contact is made at a projection that has been machined forged onto the bolt or n u t Current is applied, and the pleces are pressed together form a weld.

FIGURE 32-11 Principle of

Projections

project~on weld~ng (a) prlor to appiicatlon of current and pressure and (b) after format~on of the welds

&@
Electrodes
nuggets Weld

(b)

(a)

SECTION 32.5

Sotrd-State Weld~ng Processes

879

#,

11 32.4 ADVANTAGES AND LIMITATIONS OF RESISTANCE WELDING

The resistance-welding processes have a number of distinct advantages that account for their wide use, particularly in mass-production operations: 1. They are very capid. 2. The equipment can often be fully automated. 3. They conserve material, since no filer metal, shielding gases, or flux is required. 4. There is minimal distortion of the parts being joined. 5. Skilled operators are not required. 6. Dissimilar metals can be easily joined. 7. A high degree of reliability and reproducibility can be achieveo. The primary limitations of resistance welding include: 1 . The equipment has a high initial cost. 2. There are limitations to the thickness of material that can be joined (generally less than 6 mm or 'I, inch) and the type of joints that can be made (mostly lap joints). Lap joints tend to add weight and material. 3. Access to both sides of the joint is usually required to apply the proper electrode

1 I

force or pressure. I. Skilled maintenance personnel are required to service the control equipment. 5. For some materials, the surfaces must receive special preparation prior to welding. The resistance-welding processes are among the most common techniques for bigh-volume joining. The rapid heat inputs, short welding times, and rapid quenching by both the base metal and the electrodes can produce extremely high cooling rates in and around the weld.While these conditions can be quite attractive for most nonferrous metals, untempered martensite can form in steels containing more than 0.15% carbon. For these materials, some form of postweld heating is generally required to eliminate possible brittleness. Table 32-2 provides a process summary for resistance welding.

I'

32.5 SOLID-STATE WELDING PROCESSES

FORCE WELDlNC
Being the most ancient of the welding processes,forge welding (FOW) has both historical and practical value, as it helps us to understand how and why the modern welding practices were developed. The armor makers of ancient times occupied positions of prominence in their society, largely because of their ability to join pieces of metal into single, strong products. The village blacksmith was a more recent master of forge welding. With his hammer and anvil, coupled with skill and training, he.could create a wide variety of useful shapes from metal.

Heat source Proteetlon Matenal ~Orned Ratc of heat lnput Weld profile (UfW) MzxunUm penetration
Assets

Electrical res~stmce heatlng mth hlgh current None, isolabon of wcld sue 1s adequate AII common metals (steel, aiummum, and copper)
High

Does not apply

Does nor apply High speed, Smdll HAZ, no t l u ~ , f ~ i imetal, er or sh~eld~n gas g requ~red, adaptable lu mass production

CHAI'ILR 32

Resistance and Solid-State Welding Processes Using a charcoal forge, the blacksmith heated the pieces to be welded to aprai cal forg~ng temperature and then prepared the ends by hammenng so that they COI be properly fitted together.The ends were then reheated and dipped into a boraxfl Heating was cont~nued until the blacksmith judged (by color) that the \vorkpiecesw at the proper temperature for weld~ng.They were then withdrawn from the heat and ther struck on the anvll or h ~by t the hammer to remove any loose scale or impurib The ends to be joined were then overlapped on the anvll and hammered to the degi necessary to produce an acceptable weld. As the lwo pieces reduce m tkckness, they spread in width,resulting m the creab of new, fresh, uncontamnated metal surface. As these surfaces are being created, I hammer blows also provide the necessary pressure to produce instant coalescence Th by the correct combmation of heat arid deformation, a competent blacksm~th could p duce joints that might be every bit as strong as the origmal metal. However, because the crudeness of the heat source, the uncertainty of temperature, and the difficulty maintaluing metal cleanhness, a great amount of sk~ll was required and the results WI h~ghly vanahle.

FORCE-SEAM WELDING Although forge weldmg has largely been replaced by other join~ngmethods; aId amount of forge-seam welding is still used in the manufacture of pipe. As previou presented m Chapter 17, a heated strip of steel is first formed into a cylinder, and1 edges are simply pressed together in e~ther a lap or a butt configuration Weldingis I result of pressure and deformation when the metal is pulled through a conical weld bell or passed between welding rolls. COLD WELDING Cold welding is a variation of forge welding that uses no heating but produces me1 I lurgical bonds by means of room-temperature plastic deformation. The surfaces to joined are first cleaned and placed in contactThey are then subjected to high locali pressure,,sufficientto cause about 30 to 50% cold work.While some heating will oo due to the severe deformation, the primary factor in producing coalescence is the If pressure acting on newly formed surface material.The cold-welding process is generr confined to the joining of small parts made fro111soft, ductile metal, such as the electj connections shown in Figure 32-12. ROLL WELDING OR ROLL BONDING In the roll-welding or roll-hondmg (ROW)process, two or more sheets or plates of rn: are joined by passing them simultaneously through a rolling mi1l.A~ the materials, reduced in thckness, the length andlor width must mcrease to compensate.The nef

FIGURE 32-12 Small parts jo~nedby cold weldlng. (Courtesy of Koldweld Corporatron, W~lloughby, OH.)

SLCTION 32.5 Sol~d-State Welding Processes

881

FIGURE 32-13 Exarn~les of channels that have been formed between the roll-bonded sheets. (Courtesy O11nBrass, East Alton, I1 )

created uncontaminated interfaces are presed together by the rolls and coalescence is produced. Roll bonding can be performed either hot or cold and can be used to jo~n either similar or dissimilar metaIs (such as the Alclad aluminums-a &in of highcorrosion-r.esistance aluminum over a core of high-strength aluminum-or conventional steel with a stainlesssteel cladding).Theresulting bond can be quite strong, as evidenced by the roll-bonded "sandwh" matenal used in the production of various U.S. coins. By precoating select portions of one interfacesurface with a material that prevents bonding, the roll-bonding process can he used to produce sheets that have both bonded and nonbonded areas. Subsequent heating in an oven or furnace can cause the no-bond soaring to volatilize.The resultmg pressure expands the no-bond region6,producmg flaw paths for gases or liquidaA common example of t h ~ technique s is in the manufacture of refrigerator freezer panels, like those shown in F~gure 32-13, where inexpensive sheet metal is used to produce sh.ucturalpanels that also serve to conduct the coolant.

FRICTION WELDING AND INERTIA WELDING

In f P i c m * weldzng (FRW) the heat required to produce the joint is generated by friction heating at the i n t e r f a a m components to be jorned are first prepared to havesmooth, square-cut surfaces. ASshown in Figure 2-14, one piece is then held stationary whrle the other is mounted in a motor-driven chuck or collet and rotated against the hrst piece at high speed. A low contact pressure may be applied initially to permit cleaning o f the suryaces by a burnishing action. The pressure is then increased, and contact friction quickly geberates enough heat to soitea both camponen& and raise the abutting surfaces to the weldmg temperature. As soon as this temperature ia reached, rotation 1s stopped and the pressure 1 s further increased t o complete the weld. The softened metal is squeezed out to the edges of the joint, forming a flash, which can be removed by subsequent machining. Glean, uncontarmnated matertal is left on the mterface,and the force steel bars create a "forged" structure in thejoint.Friction welding has been used to u p to 20 cm (8 in.) in diameter ftnd tubes of even larger diameter.The process is also ]deal for welding dis~imllar metals with very different melting temperatures and physical

om

882

CHAPIER 32

Resist

z and Sol~d-State Weld~ng Processes

FIGURE 32-15 Schematic diagram of the equ~pment used (Courtesy ol for fr~ction weld~ng. Materials Eng~neering.)

rotating part

properties, such as copper to aluminum, titanium to copper, and nlckel alloys to st F~gure 32-15 shows a schematic of the equtpment requ~red for friction welding

kinehc energy of the flywheel isconverted mto frictional heat at the interface between

are requued.

angular reciprocating motrons,friction weldmg can be extend to noncircular shape$ as square and rectangular bars. F~gure 32-17 shows yome typxcal fncbon-welded par
FIGURE 32-16

Schematic f rhe vdrlour representat on o steDs ~n inrrtla weldlnq. The rot'ating part is now attached to a large flywheel.

FRICTION-STIR WELDING

advancing probe, and coalesces to form a solid-state bond The most common ap tion is the formatlonof butt welds,usually between plates of the lower-melting-poin als (both wrought and cast alloys) or thermoplast~c polymers

FIGURE 32-17 Some typical friction-welded parts. (Top) Impeller made by joining a chrome-moly steel shaft to a nickel-steel casting. (Center) Stud plate with two mild steel studs joined to a square plate. (Bottom) Tube component where a turned segment is joined to medium-carbon steel tubing. (Courtesy of Naycar Boy City, Division of Newcoc Inc., Royal Oak, MI.)

Weld quality 1s excellent.The extensive plastic deformation creates a refined gram structure with no entrapped oxides or gas porosity. As a result, the strength, ductihty, fatigue life, and toughness of the resulting weld are all qulte good Welds m aircraft aluminum are 30 to 50% stronger than those formed by arc weldmg. Since no material is melted, both wrought and cast alloys can be jomed, and they can be jolned to each other. No filler material or shielding gas is required, and the process is environmentally friendly (no fumes, weld spatter, or arc glare). Because of the high energy efficiency, total heat input and associated distort~on and shrinkage are all low. Joint preparation 1s minimal, and surface oxides need not be removed. Welding can be performed in any position and requires access to only one side of the plate. Gaps up to 10% of the material thickness can be accommodated with no reductton in weld quality or performance. Weld speed, however, is slower than in most fusion processes. As shown in Figure 32-18, the key process variables Include probe geometry (dlameter, depth, and profile), shoulder diameter, rotation speed. downward force, travel speed, and poss~ble tilt to the tool. Friction-st~r welding has been used to weldnearly all of the wrought aluminum alloys, including some that are classified as "unweldable" by fuslon processes. Aluminum plates up to 65 mm (2'1- in.) thick have been successfully welded from a single slde, and a l u m u r n up to 75 mm (3 in ) thlck has been welded by a two-sided process. Copper, lead,magneslum, btanium, and zinc have all been successfully welded, along with thln steel plate and some combinat~ons of d~sstmilar materials As an ~ndlcation of process capability, the Echpse 500, a six-passenger, short-hop air taxi, built by Eclipse Aviatlon of Albuquerque,New Mexico, contans ovex 135 meters (5300 inches) of friction-stir welds on the aluminum fuselage and w ~ n g panel assemblies, eliminatmg over 60% of the usual rivets. Various thermoplastics, mcludiug polyethylene, polypropylene, nylon, polycarbonate, and ABS, have also been succe58fully welded, with several exhib~ting as-welded strengths exceeding 95% of the tens~le strength of the base material. In an extension known as friction-stcrprocessmg, the thermomechanical features of friction-stir are used for purposes other than creatlng a jomt. By tracklng the stir tool through the mnater~al w~th overlapping passes, ultrafine grain size ( 4 0 pm) can be produced that enables superplastic forming at comparatively high stram rates. While sumaterial can now b e made perplasticity is usually limited to thin sheets. th~cker superplastic, and by stirrlng only selected locations, large parts can be made from less expensive conventional materials, with enhanced formability in only the needed locatlons. In other applications, key surfaces of large castings can be enhanced by the p a t sage of the probe.The cast structure, w ~ t h possible micropo~os~ty and segregation, is replaced by a fine, homogeneous,wrought microstructure. Strength, ductihtp,corrosion resistance, and fatigue resistance are all improved. Fusion welds can be stirred to replace the cast stmcture w t h a fine, worked structure,removing any weld defects and euhancing properties. Reinforcement particles can be sbrred ~ n t o a material to create a Downward force

of weld

rotating probe generates frictional heat, whlle the shoulder provides add~tlonal frict~on heating and prevents expulsion of the softened mater~al from thejolnt. (Note To provide addltlonal forging action and conf~ne the softened material, the tool may be tilted so the trailing edge IS lower than the leadlng segment.)

FIGURE 32-18 Schematic of the friction-stir weldrng process. The

Processes re and Solid-State Weld~ng

FIGURE 32-19 (a) Top surface of a friction-stirweld joining 1.5mm- and 1.65-mm-thick aluminum sheets with 1500-rpm pin rotation. The welding tool has traversed left-to-rightand has retracted at the right of the photo. (b) Metallurgical cross section through an alloy 356 aluminum casting that has been modified by friction-stir processing.

(bl

partrcle-re~nforced composite surface a n a standard alloy substrate. Powder produ of unique composition can be brought to full denaity with attracttve strength and d t~lityIn a modification known asfrict~on-strr channeling, a slight upward hellx is I porated onto the surface of the rotating probe.As the proberotates, material is drsp upward, creating a conXmuous subsurface channel, such as those used for conveyi cooling water. One of the photos in Figure 32-19 shows a Friction-stir weld tn alumin plate, viewed from the top.The other shows the structural changes induced by a fri stir pass through an alurn~num-alloy casting. N a e the significant refinemnt in sttuct Table 32-3 summarizes some of the aftrachve features O f the friction-stir process.

qiSfW&%g$ir Welding
i : , , : : ;%4 ' ; : ,
Metallurgical benefits Excellent weld qoality range ~Crnatonals,includmg some "noilwelddble' by Appl~cable lo a w~de Cu\lon rnelhds Sol~&tate process Low d~srortlon of ihe rvorkpiece High julnt strength No 10% of allay elements F~ne microstruetule No crack~ng Environmental benefits
No shield~nggaa la required No surfan, dranlng IS required No solvent degreaung 1 s used No fumes, @ses, or smoke ISproduced Postweld Onlsh~ng is often unnecessary No arc gidre ur reilected laser beams

"

Energy benehts

Welds praduced with farless energy than in wher processes Enables we,ght redliciinn in auctaft. automobiles and ships

Force

DC polarization

m 1
Reflector ( sonotrode

AC power

FIGURE 32-20 Diagram of the equipment used in ultrasonic welding.

ULTRASONIC WELDING In ultrasonic welding (US%'), coalescence is produced by the localized application ot high-Gequency (10,000 lo 200,000 Hz) shear vibrations to surfaces that arc held together under rarher light normal pressure. Allhough there is some heating a1 the faying surfaces, the interface tempcrature rarely exceeds one-half of the melting point of the material (on an absolute-temperature scale). Instead, it appears that the rapid reversals of stress along the contacl interface break up and disperse the oxide films and surface contaminants, allowing the clean metal surfaces to coalesce into a high-strength bond. Figure 32-20 depicts the basic componenls of the ultrasonic welding process. The ultrasonic transducer is essentially the same as that employed in ultrasonic machining. 1t is coupled to aforce-application system that contains a wclding tip on one end, either stationary for spot welds or rotating for seams. The pieces to be welded are placed between this tip and a reflecting anvil, thereby concentrating the vibratory energy. Ullrasonic welding is restricted to the lap joint welding of thin materials-sheet, foil, and wire-or the attaching of thin sheets to heavier structural members. The maximum thickness is about 2.5 mm (0.1 in.) for aluminum and 1.0 mm (0.04 in.) for harder metals. As indicated in Table 32-4, the process is particularly attractive because of the number of metals and combinations of dissimilar metals that can be joined. It is even possible to bond metals to nonmetals. such as aluminum to ceramics or glass. Because the temperatures are low and no arcing or current flow is involved, the process is often preferred for hcat-sensitive electronic components. Intermetallic compounds seldom form, and there is no contamination of the weld or surrounding area.The equipment is simple and reliable, and only moderate skill is required of the operator. Surface preparation is less than for most competing processes (such as resistance welding), and less energy is needed to produce a weld. Typical applications include joining the dissimilar metals in bimetallics, making microcircuit electrical contacts, welding refractory or reactive metals, bonding ultra-thin metal, and encapsulating explosives or chemicals. Ultrasonic welding has also been used to produce spot and seam welds on thin plastics and to seal Coil or plastic envelopes and pouches. Compared to joining methods that employ solvents or adhesives, the ultrasonic method is considerably faster and results in products with cleaner surfaces. DIFFUSION WELDING Diffusion welding (DFW)or diffusion bonding occurs when properly prepared surfaces are maintained in contact under sufficient pressure and time at elevated temperature.

C H A 32. ~ Resistance and %lid&ate Welding Processes

GermaRium

Gold flolybdenum

NieBel
Piatinurn

In kmmt to the deformation-welding methods, plastic flow is W e d and d

pal @ondingm&ehanism is.2ttomicdiffusian.A well-prepared interface can be a p l q u i grain boun:dar$ yithinn3iveaitlgvoids afidimpurities. Under lowpre

Plates to be welded

steel and stainlerssteel FIGURE 3.2-21 (Left) Schematic of the explosive welding process. (Right) Explosive weld between m~ld showing the characteristic wavy interface.

two layers of metal and detonated in a progressive fashion, beginning where the surfaces touch. A compressive stress wave, on the order of tliousands of megapascals (hundreds of thousands of pounds per square inch), sweeps across the surface of the platesSurface films are liquefied or scarfed off the metals and are jetted out of the interface.The clean metal surfaces are then thrust together under high contact pressure.The result is a Lowtemperature weld with an interface configuration consisting of a series of interlocking ripple's. Since the bond strength is quite high, explosively clad plates can be subjected to a wide variety of subsequent processing, including further reduction in thickness by rolling. Because it is a solid-state welding process, numerous combinations of dissimilar metals can he joined,

W Key Words
flash forge welding forge-seam weldimg friction welding friction-stir processing friction-stir welding inertia welding lap joint nugget projection welding resistance welding resistance seam welding resistance spot welding rocker-arm machine roll bonding solid-state welding spot-weldinggun transgun ultrasonic welding

I I Review Questions
'

21. What two limitations of spot welding can he overcome by using the projection approach? 22. What limits thenumber of projection welds that can he formed in a single operation? 23. What are some of the attractive featuresof resistance weldimg when viewed from a manufacturing standpoint? 4. Why is thereno need for fluxes or shielding gases in resistance 24. What are some of the primary limitations to the use of resistance welding? ich termis most signifi- 25. What type of metallurgical problem might be encountered when spot welding medium- or high-carbon steels? the total 26. What were some of the limitations that made the forge welds of a blacksmith somewhat variable in terms of quality? an be taken to reduce the resistance between 27. What features promote coalescence in cold welding? 28. Describe how the roll-bonding process can be used to fabricate products that contain pressure-tight, fluid-flowchannels that once required the use of metal tubing. during the resistance-welding cycle? Too much pressure? 9. What is the ideal sequence for pressure application during 29. How is inertia welding similar to friction welding? Different resistance welding? from Maion welding? ! 10. Why do the resistance-welding conditions become less fauor- 30. How are surfaceimpurities removedin the friction- and inertiawelding processes? able as the material heats and softens? ' 1 1 . What magnitude of current may he used to produce resistance 31. What are some of the geometric limitations of friction and inertia welding? 12. What is the simplest and most widely used form of resistance 32. How does friction-stir welding differ from friction welding? 33. What are some of the attractive features of friction-stir welding? 34. What types of material or property modifications can be induced through friction-stir processing? 35. How do ultrasonic vibrations produce a weld? 15. What is the major advantage of spot-welding guns? ,,16.What are some of the properties that must be possessed by 36. What are some of the geometric limitations of ultrasonic , resistance-welding electrodes? welding? 37. What are some of the attractive features of ultrasonic welding? !17. What is the most common metal that is spot welded? (18. What is the practical limit of the thicknesses of material that 38. What are the conditions necessary to produce high-quality diffusion welds? can be readily spot welded? 39. How are surface contaminants removed during explosive welding? methods used to produce resistance seam 40. If the interfaceof a weld is viewed in cross section,what is the distinctive geometric feature of an explosive weld?
1. What are the two primary functions of the electrodes in resistance welding? 2. What are [he two major roles of the applied pressure in resistance welding? e welding be considered as aform of solid-

CHAM'ER 32

Resistance and Sol~d-State Welding Processes

I I I Problems
materials, require the joining of dissimilar materials. Select several o i the processes discussed in this chapter and investigate the capability of the process l o join dissimilar materials and the associated iimitations. stir welding. Can you identify other examples of wher ing or joining process i s currently being used for p other than those for which i t wasinitially developed.

Field Repair to a Power Transformer

oil. Overheating occurred, and a repair is now necessary. cooling in their design. Large transformers are often submerged in oil-filled reservoirs, where the volume of oil provides a noncorrosive heat sink. In addition, it is not uncommon for additional features, such as horizontal cooling fins, to be added to the design to aid in dissipating heat from the reservoir. The figure shows the exterior of a large transformer that has been installed in a rural, somewhat remote location. The reservoir housing has been constructed by welding low-carbon steel plates 9 and 12 mm and in.) thick. While the transformer was in use, a service vehicle accidentally backed into the cooling-fin assembly, producing cracks in several of the fillet areas and a resulting loss of

cedure and make the necessary arrangements.

for this task, 2. For each of the candidate processes, identify its primary advantages and limitations, 3. Which the candidate processeswould You

(i

recommend? Why?

4 . Describe the procedure that you would outline for

such a repair. Are there any special concerns or precautions?

33 1 INTRODUCTION Thermal Spray Coating or Metalllz~ng AND CUTTING 33.2 OTHER WELDING PROCESSES 33.4 BRAZING Thermit Welding Nature and Strength of Brazed Jomts Electroslag Welding Design of Brmed Jomts Electron-Beam Welding Fffler Metals Laser-Beam Weldmg Fluxes Laser-Beam Cuttmg Applymg the Braze Metal Laser Spot Welding HeatingMethods Used in Flash Welding Brazing 33 3 SURFACE MODIFICATION BY Flux Removal and Other WELDING-RELATED PROCESSES Postbraze Operations Surfacing (IncludingHard Facing)

Fluxless Braz~ng Braze Welding 33.5 SOLDERING Design and Strength of Soldered Jolnts Metals to Be Joined Solder Metals Soldering Fluxes The Soldering Operation Flux Removal Fluxless Soldeting

We have already surveyed gas-flame and arc welding (Chapter 31), as well asreslstance of weld~ng and yolid-state joining processes (Chapter 32) Other processes wllhin the real~n rnclude some that are quite old (thermit weld~ng) and others that are among the newest In manutactunng (laser and electron beam).These and several others will be presented here, along w~th a briel section devoted to the application of welding and welding-related processes to surfacing and thermal spray coating. There are also many joining or assembly operations where welding may not be the best cho~ce. Perhaps the heat of welding is objectionable, the materials possess poor weldabil~ty, welding 1s too expensive, or the joint involves t h ~ n or dissimilar mater~als. In such cases low-temperature joinins methods may be preferred. These include brazing, soldering, adhesive bondmg, and the use of mechanical Iasteners Blazing and soldering will be explored in this chapter, while adhesive bonding and mechanical fasteners are deferred to Chapter 34.

WELDING AND CUTTING ~'ROCESSES 1 1 1 33.2 OTHER

THERMIT WELDING Therm~t welding (TW) is an extremely old proceqs in which superheated molten metal and slag are produced from an exothermic chemical reactlon between a metal oxlde usudlly refers to a mechanical mixture and a metallicreduc~ng agent.The name therm~t of about one part (by weight) finely divided alumlnum and three parts Iron oxide,plus possible alloy additions. When this mlxture is ignlted by a magnesium fuse (the ignition temperature is about 1150C or 2100eF), it reacts according to the following chemical equation:
8A1 + 3Fe30,

9Fe

+ 4A120, + heat

The temperature nses to over 2750C (5000F) in about 30 seconds, superheating the molten iron, which then flows by gravity Into a prepared joint, provid~ng both heat and filler metal. Runners and risers musl be provided, as in a casting, to channel the molten metal and compensate for solidification shrinkage.

890

CHAPTER 33

Other Welding Processes, Brazing and Soldering Steels and cast irons can be welded using the process described above. Cop brass, and bronze can be joined using a starting mixture of copper oxide and alumin more exotic metals.

in thick sections of materia1,particularly in remote locations or where more sophist

castings that have broken or cracked.

ELECTROSLAG WELDING

the passage of electrical current through a pool of electrically conductive liquid

in./mi move upward at a rate that is typically between 12 and 40 mmimin ('I, to 11i2

is the best of all the welding processes for making welds in thick plates. The thickn

Completed weld
Weld
(a)

metal

(bl

FIGURE 33-1 (a) Arrangement of equipment and workpieces for making a vertical weld by the electroslag process. (b) Cross section of an electroslag weld, looking through the water-cooled copper slide.

Sccno~ 33.2 Other Weld~ng and Cutting Processes

&

891

of the plates can varv from 12 to 900 mm iLi, to 38 in.). weld > ,, and the leneth of the -~ (amount of vertical travel) is almost unlimited. Edge preparation is minimal, requiring only squared edges separated by 25 to 35 mm. Applications have included building - construction, shipbuilding, machine manufacture, he&-pressure vessels, and the joining of large castings and forgings. Solidification control is vitally important to obtaining a good electroslag weld, since slow cooling tends to produce a coarse grain structure. Cracking tendencies can be suppressed by adjusting the current, voltage, slag depth, number of electrodes, and electrode extension to produce a wide, shallow pool of molten metal. A large heataffected zone and extensive grain growth are common features of the process Whiie these are undesirable metallurgical features, the long thermal cycle does serve to minimize residual stresses, distortion, and cracking in the heal-affected zone. Subsequent heat treatment of the welded structure may be necessary, however, if good fracture resistance is required.
~U
~~

ELECTRON-BEAM WELDING In the electron-beam welding (EBW) process, the metal to be welded is heated by the impingement of a bean1 of high-velocity electrons. Originally developed for obtaining ultra-high-purity welds in reactive and refractory metals, the unique qualities of the process have led to a much wider range of applications. Figure 33-2 presents the electron optical system.An electric current heats a tungsten filament to about 2200C (4000"F), causing it to emit a stream of electrons by thermal emission. By means of a control grid, accelerating voltage, and focusing coils, these electrons are collected into a concentrated beam, accelerated, and directed to a focused

FIGURE 33-2 Schematic d~agram of the electron-beam welding process

Weld~ng Processes, Brazing and Solder~ng

cannot travel well through air.To he effective as a welding heat source, the be generated and focused in a very high vacuum, typically at pressures (1 X lo-' rnm Hg) or less. Inmany operations, the workpiece must also be enclosed ~nthe h~gh-vacuu her, wlth prov~sion for positioning and manipulation.The vacuum then ensures

restriction on productivity As a consequence,electron-beam weld~ng machines h

the parts can be viewed directly through viewing ports

ing speeds are common; no shielding gay, flux, or filler metal is required; the can he performed in all positions; and preheat or postheat 1s generally unneces

steel. (Courtesy of Ham~lton Standa D~vis~on of Unrted Technologies Corporation, Hartford, CT)

5 .

w$j:$eay
.i 't,

..

..

Welding (EBW)

Heat noure?

Prowhon
Eleftrode Material jo~nzd Rate of heat Input Weld Praftle (DAN) Martmum pencrratlon Assets

ffigh-en6rsy ekectrod heam Vaouum

None
All common metals
High

20
175mm(7rn) Htgh preclsnn, hl& qual~ty, deep and netrilwwclds. smell HAZ,low dismtian, fast welding s~eed. bean is easily postioned and defleclcd: no filler metal, fly=,

orsh~eld~ngga% required

Barn and target must be within avacuum,vcry

exp@n%ve equipment. W o r k piece slzc is limited by the vaeuum chamber:slgmhcanf edge prepiratron and protectran from alignment regulred,requ~rcssaEety X-ray and vwble radlallon

O n the negat~ve side, the equpment is quite expensive,and extensive joint prepmatlon is required. Behause of the deep and n m o w weld pro*, joints must be stralght and precisely aligned over the entrre length of the weld. Machining and fixtunng tolerances are often quite demandiing.Thevacuumtequirements terrd to limit production rate,and the size o f the vacuum chamber may restrict the size of thc wofkpiece that can be welded. The electron-beam process 19 best employed where welds of extremely high quality are required or where other processes will not produce the daired results. Electronbeam wrlds often exhibit joint strength 15 to 25% greater than arc welds in the same material.The u q u e capabilities have resulted in its routine use in a number of applicalions, particularly in the automotive and aerospace industries. Table 33-1 provides a process summary for electron-beam wlding.

LASER-BEAM WELDING
Laser beams can be used as a hmt source for weldingghole makin&cutting, cladding,and heat treating a wide variety of engineering metals. When used for laser-beam wcldrng (LBW), the beam of coherent light can be focused to a diameter of 0.1 to 1.D mm, pmviding a power density in excess of 10' wattdmm2.The high-intenstty beam can be used t o simply melt the material at the joint, but more often, it produces a very narrow column of vapofized meral (a "keyhole") with a surmund~ng liquid pool. As the beam traverses, the liquid flows Into the joint to produce a weld with a depth-to-width ratio generally greater than 51. Because of the narrow weld-pool geometry, high travel speed of the beam sally several meters perminute],and low total heat input, the molten met&lsolidiiesquickly, producing a very Ulin heat-affected zone and little thermal distorhon. Finishing costs are quite low. Since welds require only one-side access,many differentjomt configurations are possible. Laser-beam welding is most effective for simplc iuslon welds without filler metal (aatugennus weids),but careful joint ppreparatlon is required to produce the narrow gap 1 and necmsary Level of ~leanliness Filler metal can be added if the gap is excessive,and a law-velwty flow of inert gas(genera1ly hehum or argon) may be used to protect the weld pool from oxidarion F~gure 3 - 4 s h m s a typical lager-beam weld, and Dble 33-2 provides a process summary for laser-be&mwelding. As shown in Rgnre 33-5, laser-beam weldrng and electron-beam welding both offersome of the highest power densities of the welding p~ocesscs.?hewell-Collimated beam of intense laser energy can produce deep penebahon welds that are similar to electron-beam wrlds, but the lasw-beam technique oKers several distinct advantages: HGURE33-4 wr butt weld Of 3-mm (0.125-1n.) a 1. The beam can be transmitted through air (i.e.,a vacuum environment isnot required). steel, made at 1.5 rn/mln w~th There isno physical contatact between the welding equipment and the workpiece.The 1250-watt laser. (Courtesy of originating laser can be a considerable distance removed. Coherent, IK., Sanfo Clara, 4 3)

(m-

894

C H A ~ 33 R Other Welding Processes, Braz~ng and Soldering

Heat source Protecf~on Electrode Matenal~oined Rate of heat input Weld piofile (DIWJ Muxlmum penetratron Assets

Laser llght None, or externally supphcd gas

None All common metals

Hlgh 5 ZSmm(1m) High heat transfer efficiency, can weld any locat~on that 1s light-a~cessble, small HAZ, low d~~tortran,can accurately focus the beam wdh ilght aptics, h ~ g h welding speed Posslble problcrns wlth reflect~vity of some metals, good posltronmg and fit-up requ~red

Llmddt~ons

s 2
a ,

iY

be made qu~ckly and at high travel spwds

w/rnrn2

2. No harmful X-rays are generated. 3. Thelaser beam can be easily shaped, directed,andfocused with both transmission reflective optics (lenses and mirrors), and some beams can be transmitted thro fiber-optic cables. 4. The only restriction on weld location is optical accessibility.Welds can be made difficult-tureach places, and materials can be joined within transparent cvntaine a vacuum tube. such as ~nside
A laser-welding system consists of an industrial laser, a means of gulding and f cusing the beam, and ameans of posit~vning and manipulating the parts to be welde

mount and maneuver a heavy, bulky laser and, by reducing weight, enhances the and accuracy of both positioning and manipulation. Cutting, drilling, welding, an 'eatingcan all be performed with the same unit, and multiple axes of motion can p a htgh degree of mobility and accessibility.

The equipment cost for a C 0 2or Nd:YAG laser-beam welding system is quite high, but this cost can be somewhat offset by the faster welding speeds, the ability to weld without filler metal, and low distortion, which enables a reduction in postweld straightening and machining. Caution should be used with such equipment, however,since reflected or scattered laser beams can be quite dangerous, even at great distances from the welding site. Eye protection is a must. Because laser welds do not significantly reduce sheet metal formability, they have been used to produce tailored blanks for the production of sheet products. Different types of steel or different thicknesses can be joined to produce single-piece products with different properties at different locations. Laser welding has made great progress in the welding of aluminum alloys and has replaced gas tungsten arc welding or riveting in a number of applications. With a sharply focused beam and short exposure times, laser welds can be very small and have a low total heat input, often on the order of 0.1 to 10 joules.These conditions are ideal for use in the electronics industry, and laser welding is frequently used to connect lead wires to small electronic components. Lap, butt, tee, and cross-wire configurations can all be used. It is even possible to weld wires without removing the polyurethane insulation. The laser simply evaporates the insulation and completes the weld with the internal wire. Lasers have also been used in hybridprocesses that combine laser welding with arc welding (GMAW, GTAW, or PAW), with both operating in one process zone and producing one weld pool.These hybrids combine the deep penetration, low distortion,and high-welding-speed features of laser welding with the wider pool, gap-bridging capability of arc welding.The resulting weld pool is wide and shaIlow at the surface, transitioning to deep and narrow. In addition to the unique and flexible weld-pool geometry, another benefit is the enhanced arc stabilization provided by the material that the laser evaporates. Laser power can be reduced from that required for lasers operating alone, and welds can be made faster than with just the arc-welding processes. The shielding gas from the arc-welding process protects the entire weld pool.

LASER-BEAM CUTTING
Cutting small holes, narrow slots, and closely spaced patterns in a variety of materials, or producing small quantities of complex-contoured sheet or plate, is another widely used application of industrial lasers. Laser-beam cutting (LBC) begins by "drilling" a hole through the matcrial and then moving the beam along a programmed path. As shown in Figure 33-6, the intense heat from the laser is used to melt andlor evaporate the material being cut. A stream of assist gas blows the molten metal through the cut, cools the workpiwe, minimizes the heat-affected zone, and may participate in a combustion reaction with the material being cut. Oxygen is the usual gas for cutting mild steel. The laser heats the metal to a temperature where the iron and oxygen combine in an exothermic reaction. The molten iron and iron oxide. have a low viscosity and are easily blown away by the flow of assist gas In this exothermic cutting process, the assist gas actually contributes additional heat. High cutting speeds are possible, the speed being limited by the rate of material burning. Nitrogen is used with stainless steel and aluminum, and, because of its high reactivity, titanium requires an inert gas, such as argon. Inert gases or air are used when cutting nonmetallics. The latter processes are ones of endothermic cutting, since the gas actually absorbs energy as it is heated. Cutting speed is set by the rate at which the laser can melt and/or vaporize material. Exothermic cutting produces an oxidized edge, while endothermic cutting (or clean cutting) results in oxide-free surfaces. Clean, accurate, square-edged cuts are characteristic of the laser cutting process, and the kerf (typically as small as 0.25 mrn) and heat-affected zone are narrower than with any other thermal cutting process. No postcut finishing is required in many applications, even though the process does produce a thin recast surface. Figure 33-7 shows the edge of carbon steel that is 6 mm (0.25 in.) thick, laser cut at 1.8mlmin with a 1250-W laser.

Laser beam

d~rectlon
Cut edge

cutting

Molten

2 Molten mater~al
ejected by gas flow

FIGURE 33-6 Schemat~c of laser-beam cuttmg. The laser prov~des the heat, and the flow

o f assist gas propels the molten droplets from the cut.

FIGURE 33-7 Surface of 6-mm-thick carbon steel cut with a 1250-wattlaser a t 1.8 mlmin. (Courtesy of Coherent, Inc., Sonta Cloro, CA.)

CO, andYAG lasers have beenused in both continuous and pulsed modes cutting speed depends on the material being cut and its thickness, it is greates continuous mode that is preferred for straight and mildly contoured cuts. The mode is preferred for thin materials and enables tight corners and intricate de be cut without excessive burning. Metal plates as well as a variety of nonmetals cut in thicknesses up to 30 mm (1'1, in.). Gutting temperatures can be in ex 11,OOO"C (20,000F), and cutting speeds as high as 25 mlmin have been observe some nonmetals. In addition to very common robotic applications, N e r s have also bee ed on CNC-type machines or combined with traditional tools, such as punch to produce extremely flexible hybrid equipment. Because no dedicated dies ing are required to produce a cut and there is no setup time, the laser is an e cal alternative to blanking or nibbling for prototype or short-run product as plastics,~ materials that are difficult to cut by conventionalme~ods,such composites. Since lasers can cut a wide variety of metals and nonmetals, laser cuttin come a dominant processin the cutting of composite materials.The more un' thermal characteristics o l the components, the better the cut and the less damage to the material. Kevlar-reinforced epoxy cuts easiest and gives a heat-affected zone. Glass-reinforced epoxy is more difficult because of the thermal differences, and graphite-reinforced epoxy is even worse because of dissociation temperature and thermal conductivity of the graphite. By the graphite has absorbed sufficient cutting heat, the epoxy matrix will have posed to a significant depth. The use of lasers for machining is discussed f u Chapter 27.

LASER SPOT WELDING

Lasers have also been used to produce spat welds in a manner that offers vantages when compared to the conventional resistance methods. A sma force is applied to ensure contact of the workpieces, and a fine-focused beam t the area of the weld.Welding is performed in the keyhole mode, where the lase a small hole throueh the molten wuddle. As the beam is moved,molten metal the hole and solidifies,forming a fusion-type nugget

Laserspot weldzng can be performed with access to only one side of the joint. It is a noncontact process and produces no indentat~ons. No electrodes are mvolved, so electrode wear is no longer a production problem Weld quahty is independent of m a t e r d resistance, surface mistance, and electrode condition, and n o water cooling is required. The total heat input is low, so the heat-affected zone is small. Speed of welding and strength of the resulting joint are comparable to resistance spot welds.

FLASH WELDING
Flash welding (FW) is a process used to produce butt welds between similar or dissimilar metals in solid or tubular form.The two pieces of metal are first secured in currentcarrying grips and lightly touched together.An electric current may be passed through the joint to provide optional preheat, after which the pieces are withdrawn slightly.An intense flashing arc forms across the gap, which melts the material on both sutfaces.The pieces are then forced together under high pressure (on the order of 70 MF'a, or 10,000psi), expelling the liquid and oxides, and upsetting the softened metal.The electric current is turned off, and the force is maintained until solidification is complete. If desired, the upset portion can then be removed by machining.Figure 33-8 shows a schematic of the flash-welding process, including both the equipment setup and the completed weld. To produce a high-quality weld, it is important that the initial surfaces be flat and parallel so that the flashing is even across the area to be joined. The flashing action must be continued long enough to melt the interface and also soften the adjacent metal. Sufficient plastic deformation must occur during the upsetting to transfer the impurities and contaminants outward into the flash.The equipment required is generally large and expensive, but excellent welds can be made at high production rates. Percussion welding (PEW) is a similar process, in which a rapid discharge of stored energy produces a brief period of arcing, which is followed by the rapid application of force to expel the molten metal and produce the joint. In percussion welding, the duration of the arc is on the order of 1 to 10 ms. The heat is intense but highly concentrated. Only a small amount of molten metal is produced, little or no upsetting occurs at the joint, and the heat-affected zone is quite small. Application is generally restricted to the butt welding of bar or tubing, where heat damage is a major concern. Upset welding (UW) is also similar to flash welding,but there is no period of arcing. The equipment and geometries are similar, but the heating is achieved through electrical resistance. The parts are clamped in the machine, pressure is applied, and high current is passed through the joint. When the abutting surfaces have been heated to a suitable forging temperature, the current is stopped and an upsetting force is applied to produce coalescence. The initial conditions of interface flatness, finish, and alignment must create uniform contact if a good-quality weld is to be produced.

G U R E 3 M Schematic tgmm of the flash-welding ocesess. (a) Equipment and Wp; (bJ completed weld.

mam
Clamps

-----

. . . . .

--.--

-----

Stationan/

Movable

Power
(a)

supply

111 33.3 SURFACE MODIFICATION BY WELDING-RELATEV PROCESSES

SURFACING (INCLUDING HARD FACING) Surfacing or overlaying is the process of depositing a layer of weld metal ou uv or edge of a different-composition base material.The usual objectives are too proved resistance to wear, abrasion,heat, or chemical attack, without having to entire piece from an expensive material, one that is difficult to fabricate, or would not possess the desired bulk properties. Since the deposited surfaces ar ally harder than the base metal, the process is often called hurd facing.This is n true, however,for in some cases a softer metal (such as bronze) is applied to base material.
Surfacing Materials. The materials most commonly used for surfacing (1)carbon and low-alloy steels; (2) high-alloy steels and irons; (3) cobalt-base (4) nickel-based alloys, such as Monel,Nichrome, and Hastelloy; (5) copper-has (6) stainless steels; and (7) ceramic and refractory carbides, oxides, borides,silic', similar compounds. Surfacing Methodr and Applications. Since some of the base metal melts d deposition, surfacing is actually a variation of fusion welding and can he perfo nearly all of the gas-flame or arc-welding techniqucs, including oxyfuel gas, metal arc, gas metal arc, gas tungsten arc, submerged arc, and plasma arc.Arc is frequently used for the deposition of high-melting-point alloys. Submerged ing is used when large areas are to be surfaced or a large amount of surfacing is to be applied.The plasma arc process further extends the process capabilities; of its extreme temperatures. To obtain true fusion of the surfacing material. ferred arc is used and the surfacing material is injected in the form of a po nontransferred arc is used, only a mechanical bond is produced, and the process a form of metallizing. Lasers can also be used in surfacing operations.

I 1

j j
1
:.

{
xla

i! . 8

THERMAL SPRAY COATING OR METALLIZING The thermal spray processes offer a means of applying a coating of high-perf material (metals, alloys, ceramics, intermetallics, cermets, carbides, or even pla more economical and more easily fabricated base metals. A wire or rod of the material is fed into a gas flame or arc, where it melts and becomes atomized by of gas, such as argon, nitrogen, combustion gases, or compressed air.The gas stre pels the 0.01- to 0.05-mm (0.0004- to 0.002-in.)-diameter particles toward the t face, where they impact ("splat"), cool, and bond. Very little heat is transfer1 substrate, whose peak temperatures generally range from 100 lo 250C (200 to As a result, thermal spraying does not induce undesirable metallurgical changes cessive distortion, and coatings can be applied to thin or delicate targets or sensitive materials such as plastics.The applied coating can range in thichess to 12 mm (0.004 to 0.5 in.). Several of the thermal spray processes use adaptations of oxyfuel equipment. Figure 33-9 shows a schematic of an oxyacetylene metal spraying signed to utilize a solid wire feed.The flame melts the wire and a flow of compr disintegrates the molten material and propels it to the workpiece. An alterna of oxyfuel gun uses material in the form of powder, which is gravity or pressure the flame, where it is melted and carried by the flame gas onto the target.'Ibe feed permits the deposition of material that would be difficult to fabricate such as cermets, oxides, and carbides. In addition, the droplet size is control e powder, not by the factors that control atomization. The lower temperatures and lower particle velocities of the oxyfuel d methods result in coatings with high porosity and low cohesive strength. An of the process known as high-velocity oxyfuel (HVOF) spraying propels the dr a supersonic stream of hot gas. Because the particles impact with high kinetic, the resulting coating is dense and well bonded.

a 5

haraeteristic iurninous white cone

FIGURE 33-9 Schematic diagram of an oxyacetylene metal-spray~ng gun. (Courtesy of Solref IWetco, Wjflterthur,
Switzerland.)

Prepared base matenal

4 8

The simplest of the electric am:methods is wire are or electric arc spraying. Two oppositely charged electrode wires are fed through a gun, meeting at the tip, where they f o r m an arc. A stream of atom~zing gas flows through the gun, stripping off the molten metal to produce a high-velocity spray. Since all af the input energy is used to melt the metal this process is extremely energy efficient. Plasma spray metdlizing,illustrated in Figure 33-10, is a moresophisticated techniquc..L\ plasmn-forming gas >cr\,c>ss but11 the hcat suurcr: and propelling ngcnt for the co.lt~~igmictcri;~l.\vhich is usoallp!cd in the form of ~(~\vder.'lhc molten r~nrticlcs attain high velocity and thzrefore produce a dense, strongly bonded coatiug:since temperatures can reach 16,500C (30,00OoF), plasma spraying can be used to deposit materials with extremely high melting pants. Metals, alloys, ceramics, carbides, cennets, intermetall~cs, and plastic-based powders have all been suocessfully deposited. While thermal spraying or metall~zing is similar to surfacing and is often applied for the same reasons, the coatings are usually thinner and the process is more sultable for irregular surfaces or heat-sensitive substrates. The deposition guns can be eithes handheld or machinedriven. A standoff distance of 0.15 to 0.25 m (6 to 10 in.) 1s usually maintained between the spray nozzle and the workpiece. Table 33-3 compares the features of five methods of thermal spray deposition. Unlike surfacing, metallizing does not melt the base metaI. Adhesion is entirely mechanical, so it is essential that the base metal be prepared m a way that promotes good mechanical interlockmg. The target material must f~rst be clean and free of dirt,moisture, 0x1, andother crmtaminants.The surface is then roughcncd by one of ;I \,ari:ty of rncrhuds to crc:lie minut', crc\iic\ that can anchor the gril is the most common techbolidifying p.irticlcs (in1 hlasting oith ;I shdtp, dbrasiv~ I I ~ ~ L and I ~ a . surtcicc n~ughncss of 3.5 to 7.5 microns is adequate for most applicnlions.
Surface Preparation for Metalllrfng.

Spray powder suspended In carrier gas, Circutatlng coolant \ Plasma gas \

Circulat~ng

Sprayed material , 'lasma flame

coolant

of2power

%A
Electrode Prepared base material

FIGURE 33-10 Diagram of a plasma-arc spray gun. (Courtery of Sulzer Metco, Winterthur, Sw~tzerlond.)

2 3 3

I
?$&
f!

Method
Flame spray
Wire

Source.

Heat Ternperature Deposited

Impact

("cl

Materials
Metals Metal$ ceamm. plastta Metdl$ carb~der

Velocity (misec)
180 30
M]&IM)O

Strength
Medium Low

Ox$uel
Oxyfuel

3000
30W

Powder

High-velocity
myfuel

CYxyfuel

3100

(HVOF)
WIIP arc

Plasma spray

DC arc DC arc

550a 5500

Metals only All

250
250-1200

ChamcterlstIcs and Applications of Sprayed Metals. During deposit' atomized, molten, or semimolten particles mix with av and then cool rapidly u pact with the base metal. The resultant coatings consist of bonded particles that range of sue, shape, and degree of melting. Some particles become oxidized and ticle voids can become entrapped. Compared to conventional wrought material, ~ngs are harder, more porous (0.1 to 15%porasity1,and more brittle.'Ibemal spray c add little, if any, additional strength to a part, since the strength of the par usually between one-third and one-half of its normal wrought strength. therefore, generally look to the coating to provide resistance to heat, wear, erosion, corrosion, or to rwtore worn parts to original dimensions and 8pecIfications.Som cal appIications include:

1. Protective coatwzgs, Zinc and aluminum are sprayed on iron and steel to provid rosion resistance-a process that may well extend the lifetime of bridges, bwl and other infrastructure items.The interior surfaces ol power boilers can be w~th high-ehromum alloys to extend wall life by providing both heat-resat corrosion resistance. 2. Buzldzng up worn surfaces. Worn parts may be salvaged or their life exte adding new metal to the depleted regions. The repair and restoration o f air gine components is pmbably the largest single use of thermal spraying. 3. Hardsuxfacmg. Although metal spraying should not be compared to hard-fa posits that are applied by welding techniques, it can be used when thin coa considered to be adequate.'&pical applicattons might include automobile cylin ers and piston rings; thread guides in terctile plants; and critical parts witinn bear~ngs, and seals. 4. Applying coatings of expensive metals. Metal spraymg provides a simple metho applying thin coatings of noblemetals to surfacs where conventional plating w not be economical. 5. Elecrrzc~properfies. Because metal can be deposited on almost any surface, spraying can be used to apply a conductive surface to an otherwise poorc or nonconductor. Coppez, aluminum, orsilver 18 frequently sprayed on glass or tics for this purpose. Conversely, sprayed alumina (A1203can be used to im 1 insulating or dielectric properties. 6. Reflectingsurfaces. Aluminum, $prayed on the back of glass by a special fusion pro makes an excellent mirror. 7. Decorative effects. One of the earliest and still important uses of metal spraying w to obtain decorative effecfr Because sprayed metal can be treated in a variety of such as bnffefed, wire brushed, or left in the as-sprayed condition,it is frequently spe for finishing manufactured products and architectural materials.

8. TailoredsurJace characteristics. Porous coatings of cobalt or titanium alloys, or certain ceramic materials, have been applied to medical implants to help promote adhesion and in-growth of bone and tissue.

In brazing and soldering, the surfaces to be joined are first cleaned, the components assembled or fixtured,and a low-melting-pointnonferrous metal is then melted, drawn into the space between the two solids by capillary action, and allowed to solidify Brazing is the permanent joining of similar or dissimilar metals or ceramics (or composites based on those two materials) through the use of heat and a filler metal whose melting temperature (actually, liquidus temperature) is above 450C (840F)' but below the melting point (or solidus temperature) of the materials being joined.The brazing process is different from welding in a number of ways:

1 . The composition (or chemistry) of the brazing alloy is significantly different from
that of the base metal. 2. The strength of the brazing alloy is usually lower than that of the base metal. 3. The melttagpoint of the brazing alloy is lower than that of the base metal, so none of the base metal is melted. 4. Bonding requires capillary action to distribute the filler metal between the closely fitting surfaces of the joint. The specific flow is dependent upon the viscosity of the liquid, the geometry of the joint, and surface wetting characteristics. Because of these differences, the brazing process has several distinct advantages: 1 . A wide range of metallic and nonmetallic materials can be brazed.The process is ideally suited lor joining dissimilar materials, such as ferrous metal to nonferrous metal, cast metal to wrought metal, metals with widely different melting points, or even metal to ceramic. 2. Since less heating is required than for welding, the process can be performed quickly and economically. 3. The lower temperatures reduce problems associated with heat-affected zones (or other material property alteration), warping, and distortion.Thinner and more complex assemblies can be joined successfully.Thim sections can be joined to thick. Metal as thin as 0.01 mm (0.OU04 in.) and as thick as 150 mm (6 in.) can be brazed. 4. Assembly tolerances are closer than for most welding processes, and joint appearance is usually quite neat. 5. Brazing is highly adaptable to automation and performs well when mass-producing complex or delicate assemblies. Complex products can also be brazed in several steps using filler metals with progressively lower melting temperatures. 6. A strong pennanent joint is formed. Successful brazing or soldering requires that the parts have relatively good fit-up (i.e., small joint clearances) to promote capillary flow of the filler metal. The parts must he thoroughly cleaned prior to joining, and many parts will require flux removal after joining. It is also important to remember that any subsequent heating ofthe assembly can cause inadvertent melting of the braze metal, thereby weakening or destroying the joint. Another concern with brazed joints is their enhanced susceptibility to corrosion. Since the ,filler metal is of different composition from the materials being joined, the brazed joint is actually a localized galvanic corrosion cell. Corrosion problems can often be minimized, however, by proper selection of the filler metal.

NATURE AND STRENGTH OF BRAZED JOINTS Brazing, like welding. forms a strong metallurgical bond at the interfaces. Clean surfaces, proper clearance, good wetting, and good fluidity will all enhance the bonding.
'Ths temperature
2s

an arb~trary one, selected to d ~ s t m g u ~ s brazing h from solder~nB

CHAPTER 33 Other Welding Processes, Brazing and Soldering

FIGURE 33-11 Typical

variation of tensile strength with clearance in a butt-joint braze. (CourTesy of Handy & Harmon,
Rye, NK)

.009"

,012"

,015''

Joint clearance in

inches

portant. If the joint is too tight,it may be difficult for the braze metal to flow in gap (leaving unfied voids), and flux may not be able to escape (remaining in loc that should be filled with braze material). There must be sufficient clearance fo

capillary forces may be unable to draw the material info the joint or holdit in place ing solidification. Figure 33-11 shows the tensile strength of a butt-joint braze as a f tion of joint clearance.

metal, such as nickel. When clearances range between 0.075 and 0.13 mm (0.00 0.005 in.), however, acceptable brazing becomes somewhat difficult,and joints with in excess of 0.13 mm (0.005 in.) are almost impossible to braze. It should be noted the speciried gap should be maintained over the entire braze area-braze surfaces sh be parallel.

and tensile stresses can induce cracking.

SCTION 33.4 Brazing

903

Room temperature
FIGURE 33-12 When branng

Brazing temperature

dissimilar metals, the initial jornt clearance should be adjusted for the dlferentthermal expansions (here bras3 expands more than rteel). Proper brazing clearances should exist at the temperature where the filler metal flows.

Force f ~ t

nickel or copper to enhance brazing Fluzdity 16 a m a s u r e of the flow characterist~cs of the molten brazemetal and is a function of the meta1,its temperature, surface cleanlinesq and clearance.

DESIGN OF BRAZED JQINTS

Because thestrength of a braze filler metal is generalIy less than that of the metals being joined, a good jomt design is reqnurd if one is to obtain adequate mechan~cal strength. The desired load-carrying ability is usually obtained by ( 1 ) ensuring proper joint dearance and (2) providmg sufficient area for fhe bond. Figure 33-13 depicts the two most common types of brazed joints: bun and Lap. Butt jyints do not require additional thickness in ihe vicinity of he jomt; they are most aften med where the strength requiref the ments arenot that crdcal.The bonding area is limited to the cross-sectional area o thinner or smaller member in contrast, lap joints can provide bondmg areas that are mnstderably largtr than ihe bbutt configuration; they are often preferred when maximam strength is deared. If the joints are made ver)i carefully, a lap of I to 1 1 ' , times the material thickness can develop strength equal to that of the parent metal Fur joints that are made by routineproduction,it is best to use a lap of 3 to 6 times the material thickness to ensure that failure w~ll occur in the base metal, no2 in the brazed joint. Variations of the two basic jolnt designs rnelude the butt-lap and scatf confgurations, shown in Figure 33-14.The butt-lap design is an attempt to combine the advantage of a uniform thickness with a large bonding area and compan~on high strenglh. Unfortunately, it also requires a higher degree of joint preparation. The scarf joint maintains uniform thickness and increases bonding area hy tilting the butt joint interface. Careful joint preparation and component alignment is requlred to maintain the desired dimensions of joint clearance. F~gure 3514shows relalively simple butt, lap, butt-lap, and scarf

Butt joint

Lappnt

FIGURE 33-13 The two most common types of braze jolnts are butt and lap. Butt offersun~form th~ckness across the jo~nt, area and whereas idp offers greater bond~ng higher strength.

904

CHA~E 33 R Other Weldmy Processes, Brazing and Solder~ng

Butt

Flat pans
Tubular pans

Flat parts

Tubular parts

Butt-lap

Scarf

Flat pans

Flat parts

FIGURE 33-14 Variattons of the butt and lau confiuurations include the buk-lap and scarf. The four tvues are shown for both flat &d tubular parts.

joints for both flat and tubular parts. Figure 33-15 shows good brazing designs far variety of joint configurations. The materials being brazed also need to be considered when d~signing a bsaze joint.Table 33-4 summarizes the compatibility of various engineering materials withth brazrng process.

FIGURE 33-15 Some common loint des~gns for assembling parts by brazing.

Cast iron Carbon and law-alloy steels Stanlrss steel Alummum and magnesium Copper and copper alloys

Nkckel and nickel alloys

Tmmum Lead and zrne ~ e m o p l a $ t i c s . t h ~ r m ~and s e t elastomers ~ Oeramics and glass D~ssun~lar metals Metals to nonmetals D ~ s ~ m i lnonmetals ar

Somewhat &fficult Recommended for low and medrum-carbon matenab, dfficult for highsdrbnn materials: seldom used for heat-treated aUoy steeis brazing allays are Recommended,s~lver and n~ckel preferred Common far alummunl alloys and some alloys oi magnerjum Recommended for copper and hlghcopper brasses; somewhat vanable w ~ t h bronzes Rewmmended D~ffroult, notrecommended Not r$commended Not recummended Not rmmmended Recommended, but may be difficult, dependmg on degree of dastmilanly Not recommended Not recommended

FILLER METALS
The filler metal used in brazing can be any metal that melts between 450C (840F) and the melting point of the material being joined.Actua1 selection, however, considers a variety of factors,including compatibility with the base materials, brazing-temperature restrictions,restrictions due to service or subsequent processing temperatures, the brazing process to be used, the joint design, anticipated service environment, desired appearance, desired mechanical properties (such as strength, ductility, and toughness), desired physical properties (such as electrical, magnetic, or thermal), and cost. In addition, the material must be capable of flowing through small capillaries,"wetting" the joint surfaces, and partially alloying with the base metals.The most commonly used brazing metals are copper and copper alloys, silver and silver alloys, and aluminum alloys. Many of the brazing alloys are based on eutectic reactions (see Chapter 4), where the material melts at a single temperature that is lower than the melting points of the individual metals in the alloy.Table 33-5 presents some common braze metal families, the metals they are used to join, and the typical brazing temperatures. Copper and copper alloys are the most commonly used braze metals. Unalloyed copper is used primarily for brazing steel and other high-melting-point materials, such as high-speed steel and tungsten carbide. Its melting point is rather high (about llOOC), and tight-fitting joints are required (gaps less than 0.075 mm). Copper-zinc alloys offer lower melting points and are used extensively for brazing steel, cast irons, and copper. Copper-phosphorus alloys are used for the fluxless brazing of copper since the phosphorus can reduce the copper oxide film. These alloys should not be used with ferrous or nickel-based materials, however, since they form brittle compounds with phosphorus and the resulting joints may be brittle. Manganese bronzes can also be used as filler metal in brazing operations. Pure .silver can be used for brazing titanium. Silver solders (alloys based on silver and copper) have brazing temperatures significantlybelow that of pure copper and are used in joining steels, copper, brass, and nickel. While silver and silver alloys are expensive, only a small amount is required to make a joint, so the cost per joint is still low. Alz~minurn-silicon alloys, containing between 6 and 12% silicon, are used for brazing aluminum and other aluminum alloys. By using a braze metal that is similar to the base metal, the possibility of galvanic corrosion is reduced.These brazing alloys, however, have melting points of about 610C (113D"F), and the melting temperature of commonly brazed aluminum alloys, such as 3003,is around 670C (1290F).Therefore,control of the brazing temperature is critical. ln brazing aluminum, proper fluxing action,surface cleaning, andior the use of a controlled atmosphere or vacuum environment is required to assure adequate flow of the braze metal. Nickel- and cobalt-based alloys are attractive for joining assemblies that will be subjected to elevated-temperature service conditions and/or extremely corrosive environments.The service temperature for brazed assemblies can be as high as 1200C(2200F). Gold and palladium alloys offer outstanding oxidation and corrosion resistance, as well as good electrical and thermal conductivity. Magnesium alloys can be used to braze other types of magnesium.
: .. #mq.ble@ F~~n:iies, etai is They & Used to join,

iny T w r a t u f m
Bra7e Metal Fam~ly
Alummum-sd~con Copper and copper alloys Copperphosphorus Silver allnys Precious metals (gold-based) Magnesium Nickel alloys

'

.:

Materials Commonly Jolned

Alurnmum alloys Various ferrous metals as well aa copper and nickel alloys and stainless steel Copper and copper alloys Ferrous and nonferrous metals, except aluminum and magnesium Iron, nickel,and cobalt alloys Magnesium alloys Stainless steel, nickel, and coball alloys

Typlcal Brazlng Temperature (DC)

565-520

925-1150
7U0-925

620-980 900-llW 595420 924-1200

906

C H A33 ~

Other Weldlng Processes, Brazing and Soldering

A variety of brazing alloys are currently available in the form of amorpho formed by cooling metal at extremely rapid ratesThese foils are extremely thin ((a or 0.0(115 in. being typical) and exhibit excellent ductility and flexrbility,even whe are made from alloys whose crystalhne form is quite brittle Shaped Inserts can or stamped from the foil and insested into the joint region. Since the braze mat fully dense, no shrinkage or movement occurs dunng the brazing operation One amorphous alloy, composed of nickel, chromium, iron, and borou,is u produce assemblies that can withstand high temperatures. During the bra tlon, the boron dlffuses into the base metal, raismg the melting point of the t filler.The brazed assembly can then be heated to temperatures above the me1 i of the original braze alloy, and the brazed joint will not melt.

FLUXES
In a riolmal atmosphere, the heat required to melt the brazmg alloy would the formation of surface oxides that oppose the wettmg of the surface and s bonding. Brazcng fluxes, therefore, play an important part m the process b solvrng oxrdes that may have formed on the surfaces prioi to heating, (2) pre formation of new oxides during heahng, and (3) lowering the surface tensio the molten brazing metal and the surfaces to be joined, thereby promoting thefl the m ~ l t e n material into the joint. Ideally, the flux will melt and become acti temperature below the solldus of the filler metal yet remain active throughout tire range of temperatures encountered while makmg the braze. Surface cleanliness 1s one of the most significant factors affecting the quality uniformity of brazed joints. Although fluxes can dissolve modest amounts of oxides, are not cleaners.Before aflux is applied, dirt,grease, oil, ru$t,and heat-treat scale sh be removed from the surfaces that are to be bra'zed. Cleaning operations water- or solvent-based techniques; high-temperature burn-off of oils, greas residues, acid pickling; grit blasting wlth selected medid, other mechanical method exposure to h~gh-temperature reducing atmospheres The less cleamng the flux h do, the more effcct~ve it will be during the brazing operation. Because thepresen surface graphite impairs wetting, cast iron materials often require special treat Graphite removal by chemical etching may be required before cast iron can he br Brazing fluxes usually take the form of chemical compounds in which the common ingredients are borates, fused borax, fluoroborates, fluorides, chlorides, alkalies, wetting agents, and water The particular flux should be selected for bility with the base metal being brazed and the particular process being used.Pas are utdized for furnace,induction, and dip brazing, and they are usually applied by b ing. E ~ t h epaste r or powdered fluxes can be used with the torch-brazing process, w applicationis usually achieved by dipp~ng the heated end of the filler wire into the material.

APPLYING THE BRAZE METAL The braling filler metal can be applied to jomts in several ways.Thealdest method ( srill a common technique when torch braung) uses bnzmg metal in the form of a or wire The joint area is first heated to a temperature high enough to melt the braze and ensure that it remains molten wh~le flowing rnto the joint The torch is then used melt the braze metal, and capillary action draws it into the prepared gap. The above method of braze metal application requires considerable labor care must be taken to ensure that the filler metal has flowed into the inner porti the joint.To avoid these difficulties, the braze metal is often ~nserted into the join to heatmg, usually in the form of wires, foils, shims, powders, or preformed rings, w ers, disks, or slugs. Rings or shims can also be fitted into internal grooves in the10 before the parts are assembled. When using preloaded joints,care must be exercised to ensure that the filler me is not drawn away from the htended surface by the cap~llary action of another sur of contact. Capillary action will always pull the molten braze metal into the sma clearance, regardless of whether that was the intended location. In addition, the flow

SECTION 33.4 Brazing

907

Tier metal must not be cut off by inadequate clearances or the presence of entrapped or escaping air. Elllets and grooves within the joint cat1 also act as reservoirs and trap the filler metal. Yet another approach is to precoat one or both of the surfaces to be joined with the brazing alloy. Simply placing the materials in contact and heating forms the desired bond.By having the braze matzrial already in place over the full area of contact, the joining operation does not have to rely on capillary action and metal flow. More complex assemblies can be produced than with conventional methods, and the thickness of the braze material is precisely controlled to provide maximum strength to the joint. All of the components must maintain fixed positions during the brazing operation, and some form of restraint or fixturing is often required. Alignment and clearances can often be maintained by tack welding,riveting, staking, expanding or flaring, swaging, knurling, or dimpling. Shims, wires, ribbons, and screens can also be employed to assist in locating pieces or maintaining fit. For more complex components, special brazing jigs and ,fixtures are often used to hold the components during the heating. When these are used, however, it is necessary to provide springs that will compensate for thermal expansion, particularly when two or more dissimilar metals are being joined.

HEATING METHODS USED IN BRAZING Since molten metal tends to flow toward the location of highest temperatnre,it is important that the heat sources used in brazing control both the temperature and the uniformity of that temperature throughout the joint. In specifying the heatingmethod, anumber of factorsshould be considered, including the size and shape of the parts being brazed,the type of material being joined, and the desired quantity and rate of production. A common source of heat for brazing is a gas-flame torch. In the torch-brazing procedure, oxyacetylene, oxy-hydrogen, or other gas-flame combinations can be used. Most repair brazing is done in this manner because of its flexibility and simplicity, but the process is also widely used in production applications where specially shaped torches speed the heating and reduce the amount of skill required. Local heating permits the retention of most of the original material strength and enables large components to be joined with little or no distortion.The major drawbacks are the difficulty in controlling the temperature and maintaining uniformity of heating, as well as meeting the cost of skilled 1ahor.A protective fluxisusually required, and the flux residue must be removed after brazing. If the flux and filler metal can be preloaded into the joints and the part can endure uniform heating, a number of assemblies can be brazed simultaneously in controlledatmosphere or vacuum furnaces, a process known asfurnace bruzing. If the components are not likely to maintain their alignment, brazing jigs or fixtures must be used. Fortunately, for most assemblies that are to be furnace brazed, a light press fit is usually sufficient to maintain alignment. Figure 33-16 shows some typical furnace-brazed assemblies. Because excellent control of the furnace temperature is possible and no skilled labor is required, furnace brazing is particularly well suited for mass-production operations, with eithe.r batch- or continuous-type furnaces being used. Furnace brazing heats the entire assembly in a uniform manner and therefore produces less warpage and distortion than processes that employ localized heating. Extremely complex assemblies can be produced, with multiple joints being formed in a single heating. A variety of furnace atmospheres can be utilized to reduce oxide film and prevent both the base and filler metals from oxidizing during the brazing 0peration.A chemical flux may no longer be needed, and the parts emerge clean and free of contaminants. When reactive mgterials are to be joined or the joint must meet the highest of standard$ a vacuum furnace may be preferred. A third type of heating is salt-bath brazing, where the parts are preheated and then dipped in a bath of molten salt that is maintained at a temperature slightly above the melting point of the brazing metal. This process offers three distinct ad\santages:
1 . The salt bath acts as the brazing flux, preventing oxidation and enhancing wettabdlty. 2. The work heats very rapidly because 11ism complete contact wlth the heating medium

Ott

:lding Processes, Brar~ng and Solder~ng

FlZClRE 33-15 Typ~cal furnace-brazed assemblies. (Courtesy of Pac~fic Metals Company, a division of Reliance Steel 6r Aluminum, Lor Angefes, CA. )

3,

Temperature can be accurately eonrrolled, so thin pieces can be attached to thi pieces without danger of overheating,This last feature makes the process well ed for brazhg aluminum, where precise temperature control is often requited.

In salt-bath brazing, the parts must be held in jigsor fixture6 {or be pre some manner), and the brazing metal must kpreloaded into the joints. To thebath remains at the desired rempera%ure during the immersion process, its vol must be substantiallylarger than that of the arisembliesto be brazed. In dip brazing the assembha ltre immersed in rd bath o f molten braking metal. bath thus pro~ides bath the heat and the metal for the joint Since the braze metal usually coai the entire workpiece, it is a somewhat wasteful process and is usually ployed Only f o r small products. In&&ctiun bralag ufiillZeS high-i~equency induction currents as the source of and is therefore limited to the joining of electrically conductive materials A va high-frequency-AC power supplies is available m large aad smdl capacities T h coupled to a simple heating coil deskned to fif around the joint. The heating mi generally formed from copper tubing and typrwlly carry a supply of cooling Although the filler metal can be added to the joint manually after it is heated,& practica is to usepreloaded jomSto speed the operation and produce more unifom Induction brazing offers the following advantag&, which account for its extensive use.
1 ,The. oomplete heating cycle is vczy rapid, usually only a few seconds in duration.

2. The @peration can be made semiautamahcso that only semiskilled labar is requir 3, Heazing can be confined to fhe specific area oafrhejoint through useof spiallp signed c o i l s , fYequency control, and short heating times, This minimizes softening distortion and reduces p m b b m assodafed with &ale and discoloration. 4. Uaiform r e d & me easily obtained5 . By making new and relakiwly simple heating coils, a wide variety of work c brazed with asingle power supply.
&ststance braziw can be used to produce relatively simple johts in metals high electrical conductivity.Tht: pads to be joined are presed between two elects and a current is passed through. Unlike resistance welding,the carbon ar:graphte e trade pro~ide most of the r&~ance in resistance brming,and tlmheating of thej isprimariiy by conductionhm the hot electrodes.

~EGTION 33.5

Soldering

909

Infrared heat lamps, lasers, and electron beams can also be used to provide the heat required for brazing. Recent stud~es have also shown microwave energy to be an efficient heat source. Silicon carbide plates are positioned around the joint and ate heated by the microwavesHeat is then transferred to the joint by radiation.

FLUX REMOVAL AND OTHER POSTBRAZE OPERATIONS Since most brazing fluxes are corrosive, the flux res~dne should be removed from the work as soon as brazing is completed. Rapid and complete flux removal is particularly important in the case of aluminum,wherechloridescan be particularly detrimental.Fortunately.many brazing fluxes arewater soluble,and an immersion in a hot-watertank for a few minutes will often provide satisfactoryresults. Blasting with g r ~or t sand isanother effective method of flux removal but this procedure may not be attractive if a good surfacefinish is to be maintained.Fortumtely,such drastictreatment IS seldom necessary, Other postbraze operations may include beat keating,cleanmg,and inspection. A visual examination is probably the simplest of the inspection techniquesand 1 s most effemve when both sides of a brazed joint are accessible for examinati0n.A proof test can be performed by subjecting the joint to loads in excess of those expected during service. Leak tests or pressure tests Can assure gas- or irqu~d-t~ghtness. Cracks and other flaws can be detected by dye penmant,magneticpartlcle, ultrasonic,or radiograph~c exm a r i o n . Destructtvr-forms of evaluation include peel tests, tension or shear tests, and metaUographic examination. FLUXLESS BRAZING Since the application and removal of brazing flux involves significantcosts, particularly wherecomplex joints and assemblies are involved, a large amount of work has been devoted to the development of procedures where a flux is not required. Controlled furnace atmospheres can make a flux unnecessary by reducing existlng oxides and preventing the f o m t ~ o n of new ones. Vacuum furnace^ can also be used to create and preserve clean brazing surfaces Special brazing metals have bbeen developed with alloy addition% such as phasphorus, that can atso fuIfill the role of a flux.

BRAZE WELDING Braze welding differs from straight brazing m thaz capillary actlon is not required to di~ttibute the filler metal. Here the molten filler i s simply deposited by gravity, as in and warpoxyacetylene gas welding, Because reIatively l O y temperatures are requ~red 18Very effective for the repair of steel products and fering is minimized, braze weld~ng rouscastings. It is also attractive forjoining cast Irons since the low heat does not alter the graphite shape, and the process does not require good wetting characterist~cs. Strength is determlnedby the braze metal bevlg u s d and the amount applied. Considerable buildup may be required if full strength is to be restored to the repairedpart B r a e welding is almost always done with an oxyacetylene torch.The surfaces are first "tinned" with a thin coat~ng of the brazing metal, and the remainder o f the filler metal is then added. Figure 33-17 shows a schematic of braze welding.

SulJcri~ly is a hra~ing-I! IIL' operation whcrc thc iill~~r nizt~l I):!> .I inclt111g lc.nip~,l.atur? (or liquidlh tcI11lx!~ilfLlrc II thi. ;ilh~y h:ls a ircczingranpc.J I x l r 1.50 C' (<%(PF). I1 is ty~ictllv u ~ d

for joining thin metals, connecting electronic components, joining metals while aviiding
F~ller metal

Weld~ng torch
I

Base meml
FIGURE 33-17 Schematic of the braze-weld~ng process

910

CHA~I'ER 33

Other Welding Processes, Brazing and Soldering exposure to high elevated temperature$ and fiUmg surface flaws and defe generally involves six nnportant steps: (1)design of an acceptable~oinf; (2) sel correct solder for the job; (3) selection of the proper type of flux; (4) cleaning of the s faces to be joined; (5) apphcation of flux,solder,and suffic~ent heat to allow the moltens der to fill the joint by capillary action and solidify; and (6) removal of the fluxresidue.

DESIGN AND STRENGTH OF SOLDERED JOINTS Soldering can be used to join a wide vanety of sizes, shapes, and thickne ployed extensively to provide electrical coupling or gas- or llquid-tight seals. Whiie low joining temperatures are attractive for heat-sensitwe materials, soldered lo~nts dom develop shear strengths in excess of 1.75 MPa (250 ps~). Consequ strength is requited,soldered jomts should be avoided, the contact area or some form of mechanical joint, such as a rolled-seam lock, should be made prio soldering. Butt joints should never be used, and designs where peeling action is pos ble should be avoided Figure 33-18 shows some of the more common solder joint des~ including lap, flanged butt, and mterlock. As with brazing, there IS an optimal clearance for best performanc der joints, a clearance of 0.025 to 0.13 mm (0.001 to 0.005 in.) provides for of the solder, expuls~on of the flux, and reasonable joint strength. In a should be held firmly so that no movement can occur until the solder has below the solidification temperature Otherwise, the resulting joint may contain crac and have very little strength. METALS TO BE JOINED Table 33-6 summarizes the compatibility of soldering with a variety of engineering terials. Copper, silver, gold, and tin-plated steels are all easily soldered. Aluminum a strong, adherent oxide that makes soldering difficult Specla1fluxes and modified t niques may be required, but adequate joints are indeed po~sible, as shown by the la number of soldered alunnnum radiators currently in automotive use SOLDER METALS Soldering alloys, the filler metals for soldering, are generally combinations of I melting-temperature metals, such as lead, tin, blsmuth, indium, cadmium, silver, and germanium. Because of their low cost, acceptable mechanical and phys~cal pr ties,and many years of use, the most commonsolders are alloys of lead and tin WI addition of small amounts of antimony, usually less than O.S%.The threemost alloys contain 60,50, and 40% tin, and all melt below 240C (465F). Because penslve, those alloys havmg higher proportions of tin are used only where t fluid~ty, higher strength, and lower melting temperature are desired. For wi and for filling dents and seams, where the primary desire is appearance and litt is required,solders containing only 10 to 20% tin are preferred. Jolnts made

Flanged butt

Flush lap

Lb

Flanged edge

P~pe lo~nt

FIGURE 33-18 Some common des~gns for soldered joints.

SECTION 33.5 Soldering

911

Mater~al CasL rron

Carbon and low alloy steels Stmless steel Alummum and magnesium Copper and copper alloys N~ckel and mckei alloys

Soldering Recommendation Seldom used mce qraphte and slllcon inh~bit bondmg . D~fficult for low-carbon materials, scldom used for highcarbon materials
Common for 300 serres;dlfficult for 400 senes

Seldom used, however,speaal solders are available

Recommended for copper, brass, and bronze Commonly performed usmg hlgh tm solders Seldom used Recommended, but mubf usc low-meltmg-temperature solders Not recommended Not recommended

T~tan~um Lead and z~nc Thermoplast~cs, thermosets, and elastomers Ceramics and glass Dlssimtlar metals Metals LO nonmetals Dissunllar nometals

Recommended.but w ~ t h cconsideratlnn fm galvaluc corrosion Not recommended

Not recommended

tin alloy require h~gher temperatures to psoduce but will withstand servlce temperatures as high as 150C (300F). Other soldering alloys may be specified for special purposes or where environmental or health concerns dictate the use of lead-free ~oints. Lead and lead compounds can he quite toxic Since 1988,the use of lead-containing solders in drinking water lines has been prohibited m the United States, and concern has been expressed regarding other applications and industries. Japan andrhe European Union have banned the use of lead-containing solders in electronic equipment. If subst~tute solders are to be acceptable, however, they should not only be harmless to the environment, but should also exh~hit deslrable characteristicsm the areas of melting temperature,wettahiliTy,electrical and thermal conductivity, thermal-expansion coefficient. mechanical strength,ductility, creep resistance, thermal fatigue resistance, corrosion res~stance, manufacturability,and cost.At present, none of the lead-free solders meet all of these requirements, and most are deflclent in more than one area Compatible fluxes must also be identified, and assembly methods may need to he modifled. Most of the alternative solders have been proposed from other eutectic alloy systems.T~n-antimony and tin-copper alloys are useful in electrical applications and have good strength and creep resistance but htghmelting points. Bismuth alloys have very low melting points and good fluidity but suffer from poor wettability Indium alloys offer low meltlng points, duct~lity that is retained even at cryogenic temperatures, and rapid creep that allows joints between dissimilar metals to adjust to changes in temperature without generating internal stresses.Tin-indium alloys have been used to join metal to glass and glass to glass.They have very low melting points and good wettabillfy, hut they are expensive and can be somewhat brittle Aluminum is often soldered with tin-zinc, cadmium-zinc, or alummum-zinc alloys.Tin-silver and tin-gold offer possibilities when a somewhat higher melting potnt 1s desired (typically above 205OC, or 400F) coupled with good mechanical strength and creep resistance, but both systems are limited by the h~gh cost of their components. Lead-silver and cadmium-silver aUoys can also be used for higher-temperature service. The three-component tm-sllver-copper system has received considerable attenhon for electronics ao~licatlons. L ~ k the e filler metal used in brazing and braze welding, solders are available as wue f standard and special preshaped forms.Table 33-7 and paste, as well as in a variety o presents some of the more common solder alloys with their melting and typ~ c aapplicat~ons. l

..

SOLDERING FLUXES
As in brazing, soldering requires that the metal surfaces be clean and free of omde so that the solder can wet the surfaces and be drawn mto the joint to produce an effective bond Soldering flrrxes are used to remove surface oxides and prevent oxide formation

98 Pb-2 Sn 322 90 Pb10 Sn 302 SO Pb-ZO Sn 277 70 Ptt30 Sn ZS5 60 Ph-40 Sn 238 50 Pb50 Sn 216 40 Pb-60 Sn 190 Silver Folders 97 5 P b 1 Sn-1.5Ag 308 36 P M 2 Sn-2 Ag 189 96 Sn-4 Ag 221 Other alloys 124 45 -55 Bi 138 43 Sn-57 Ei 95 Sn-5 Sb 240 125 50 Sn-50 In 37 5 Pb-25 In-37 5 Sn 138 95 5 3 - 3 9 Ag-0 6 CO 217

316
268

183 183 183 183 183

308 179 221 124 138 234 117 138 217

6 34 94 72 55 33 7 0 10
0

Side seams m three-p~ece can Coating and jo~n~ngmetals Fdmg and seammg auto bodies Torch aoldenng W~prng solder, rad~ator cores, heater unlts General purpose Electronrc (low temperature)
ffigher-temperature semce Eiectr~cal

Electrical
Low temperature Low temperature Electrical Metal-to glass Low temperature Eleca~cal

O
0

6
8 0
0

during the soldering process, but it is essential that dirt, oil, and grease be remov before the flux is applied.This precleaning or surface preparation can be performed a variety of chemical or mechanical means,including solvenr or alkalme degreasers, a immersion fpickling), grit blwtm$ sanding, wire brushing, and other me~hanical sion techniques Soldering fluxes are generally classified as cqrrostve or noncorrosrve. The common nmcorrasive Bux is rosjn (the residue after distilling turpentine) dlssolve alcohol. Rosin fluxes are suitable for making joints to copper and brass, and to tin-, m~um-, and silver-plated surfaces, provided thaz the surfaces have been adequa cleaned prim to soldering.Aniline phosphate is a more active noncotrostve flux, bu has limited use h e w s &it emits toxic gaseswhen heated.The wide variety of corro fluxes provide enhanced cleaning action but require complete removal after the salde operation to prevent corrosion problems during service.

THE SOLDERING OPERAnON

Solderingrequires a source of sufficient heat and a means of transferring it to them als being joined. Any metbod of heating that is suitable for brazing can be used for s dering, but furnace and salt-bath heating are seldom used. Most hand soldering is done with soldering irons or small oxyfnel or air-fuel (acetylene, propane,butane, MAPP) torches Induction heatmg is used when large numbers of identical parts are be soldered.For low-melting-point @lders,infrared heat sources can also be employ The joints can be preloaded with solder, or the filler metal can be supplied from a The particular method of heating ltsually dictates which procedureis used. Wave soldering, depicted in Figure 33-19, is a process used to solder wire e such as the multiple connectorsthat protrude through holes in electronic circuit b Molten solder is pumped upward through a submerged nozzle to create a wave o in a pool of moltenmetal.The b u i t boards are then passed across this save at a where each of the pins sees contact with the molten metal. Wetting and capillary a pulls solder Into each joint, and numerous connections are made as each board p across the wave.

FIGURE 33-19 Schematic of wave

soldering.

In the vaporphase solderzng process, a product w ~ t h preposrtioned solder is passed through a chamber containing hot, saturated vapor.The vapor condenses on the cooler product, transferring 1t6 heat of vaporlzatlon This results in rapid and uniform heatrng, with excellent temperature control and the possibility of an oxygen-free environment The soldering temperature is lim~ted only by the b o ~ l ~ n pg a n t o l the flu~d, w~th current materlals operating in the range of 100" to 265C (212"to 510F).While the process can be used to cure epoxles and stress relieve metals, its pr~mary application 1s the soldering of surface-mounted components to substrate materlals. Because of the prense temperature control, mult~pass soldering is possible, using up to three different solder compositions wlth three different meltlng temperatures. Because the solder is preposrhoned, this process is also known as vapor-phase reflow saldenng. Dlp saldenng, where the entire piece is Immersed m molten metal, has been used to produce automobile radiators and "tinned" coatings.

FLUX REMOVAL
After soldering, the flux residues should be removed fi-om the finished joints, either to prevent corrosion or for the sake of appearance. Flux removal is rarely difficult,provided that the type of solvent in the flux is known. Water-soluble fluxes can be removed with hot water and a brush. Alcohol will remove most rosin fluxes However, when the flux contarns some form of grease, as in most paste fluxes, a grease solvent must be used,followed by a hot-water rinse. In the past, solvents contain~ng chlorofluorocarbons (CFCs) were the cleaners of choice, but since they have been implicated in the depletion of atmospheric ozone, an alternat~ve means of flux removal should be employed or the process converted to fluxless soldering.

FLUXLESS SOLDERING
SeveralfZunless-solderlag techniques have been developed using controlled atmospheres (such as hydrogen plasma), thermomechanical surface actrvatlon (such as plasma gas Impingement), or protective coatings that prevent oxide formakon and enhance wettlng Additional successes have been reported with both laser and ultrasonic soldenng.

I I Key Words
mist gas

autogenous weld braze weldlng braz~ng butt loint capillary actlon corroylve flux drp brazing drp soldering electron-beam weldrng electroslag welding filler metal

flash weld~ng fluldlty flux fluxless bravng fluxless solder~ng furnace brazing hard fia~flg hybrid proce$ses induction brazing jigs and fixtures joint clearance kerf

lap jomt laser spot weld~ng laser-beam cultlng laser-beam welding lead-free solder metalhzmng mold~ng plates noncorrosive flux percuss~on welding reflaw solder~ng resistance brazrng TOSln

salt-bath brazing sdver solder soldermg

surfacing

thermal spray therm~t therrmt weld~ng torch brazing upset weldrng vapor phase soldering wave solderrng wettablhty

914

C H A ~ E33 R

Other Welding Processes, Brazing and Soldering

I I Review Questions
1. In what ways is a thermit weld similar to the production of a casting? 2. What is the source of the welding heat in thermit welding? 3. For what types of applications might thermit welding be attractive? 4. What is thesource of the welding heat in electroslag welding? 5. What are some of the various functions of the slag in electroslag welding? 6. Electroslag welding would be most aliractive for the joining of what typt's of geometries and thicknesses? 7. Why is a high vacuum required in the electron-beam chamber of an electron-beam welding machine'! 8. What types of production limitations areimposed by the highvacuum requirements of electron-beam welding? What compromises are made when welding is performed on pieces outside the vacuum chamber? 9. What are the major assets and negative features of highvoltage electron-beam welding equipment? 10. What are some of the attractive features of electron-beam welding? Negative features? 11. What is unique about the fusion zone geometry of electronbeam welds? 12. What are some of the ways in which laser-beam welding is more attractive than electron-beam welding? 13. Which type of laser light can be transmitted through fiberoptic cable'? 14. Why is laser-beam welding an attractive process for producing tailored blanks for sheet metal forming? For use on small electronic components? IS. What are thc attractive properties of hybrid processes that combine laser and arc welding? 16. What is the function of the assist gas in laser-beam cutting? 17. How do the cut edges differ with endothermic laser cutting and exothermic laser cutting? 18. What features have made lasers a common m a n s of cutting compositt: materials? 19. What are some of the attractive features of laser spot welding? 20. In the flash-weldingprocess, why is it important to have a sufficient duration of arcing and sufficient amount ofupsetting? 21. What me some common objectives of surfacing operations? 22. What types of materials are applied by surfacing methods? 23. What are some of the primary mcthods by which surfacing materials can b e deposited onto a metal substrate? 24. What are some of the techniques that can be used to apply a thermal spray coating? 25. How is thermal spraying similar to surfacing? How is it dfferent?

26. Why issurface preparationsuch acritical featureofmetalluq 27. What are some of the more common applications of sprayt coatings? 28. Providea reasonabledefinition of brazing. 29. What are some key differences between brazing and fusir welding? 30. Why is brazing analtractive process for joining disslmilarm terials? 31. What advantages can be gained hy the lower temperatnres~ the brazing process? 32. Why do brazed joints have an enhanced susceptibilitytoconosia 33. What is the most important factor contributing to thestreng of a brazed joint? 34. How does capillary action relate to joint clearance? 35. Why is it necessary to adjust the initial room-tcmperatu clearance of a joint between two significantly different metal: 36. What is wettabiiity? Fluidity? How does eachrelate to bradnl 37. What are the two most common types of brazed joints and ti attractive features of each'? 38. What are some important considerations when selecting brazing alloy? 39. What are some of the most commonly used brazing metals: 40. What specialmeasures should he taken when brazing alumin4 41. What arc the three primary functions of a brazing flux? , 42. Why is it important to preclean brazing surfaces before q plying the flux? 43. In what ways might braze metal he preloaded into joints? 44. What is the purpose of brazing jigs and fixtures? 45. What is the primary attraction oIlurnacebrazing operatio@ 46. Why might reducing atmospheres or a vacuum he emplo$ during furnace-brazing operations? 47. Why is dip hrazing usually restricted to use with small par$ 48. What are some of the attractive features of induction brazini, 49. Why is flux removal anecessary part of many brazing opaatiod 50. What benefits can be achieved through fluxless brazing? 51. How does braze welding differ from traditional brazing? 1, 52. What is the primary differencebetween brazing and solder4 53. Why is solderingunattractive ifa high-strength joint is deii 4 , For many y e a , the most commonsolders were all, ofw$ ,' , two base metals? 55. What is driving the conversion to lead-free solders? 1: 1 56. What aresome of thedifficulties encountered when attemp ing a conversion to lead-free solder? 57. What are the two basic families of soldering flux? ! 58. What are some of the more commonheat sourcesforprodu$ a soldered joint? I

II Problems
1 A conirnon problem w ~ t h brazed or soldered Jolnts IS galvanIC corrosion, slnce the jolnt usually mvolves d~ssim~lar metals m d~rect metal-to-metal electrical contact a. For each of the varrous solder or braze joinfs descnbed below, determine wh~ch matenal w~ll act as the corroding anode (1) Two pleces of low-carbon steel bemg brazed with a copper-base braz~ng alloy (2) A copper wlre b a n g soldered to a steel sheet using lead-tin solder
(3) Pieces of tungsten carbide being brazed into rece in a carbon-steel plate b. How do the various lead-free solders compare to the ventional lead-tin solders with regard to their pote for galvanic corrosion? c. If galvanic corrosion becomes a significant and chr problem in a brazed asscmbly, what changes might suggest that could possihly reduce or eliminate problem?

ADHESIVE BONDING, MECHANICAL FASTENING,AND JOINING OF NONMETALS

The ideal adhesive bonds to any material, needs no surface preparation, cures rapidly, and maintains a high bond strength under all operating conditions. It also doesn't exist. However, tremendous advances have been made in the development of adhesives that are stronger, easier to use, less costly, and more reliable than many of the alternative methods of joining. From early applications, such as plywood, the use of structural adhesives has grown rapidly. Adhesives are everywhere-in construction, packaging, fnrniture, appliances, electronics, bookbinding, product assembly, and even medical and dental applications.They are used to bond metals, ceramics, glass, plastics, rubbers, composite materials, woods, and even a variety of roofing materials. Even such quality- and durability-consciousfields as the automotive and aircraft industries now make extensive use of adhesive bonding.Adhesives in the automotive industry have advanced from the attaching of interior and exterior trim to the joining of major components, such as door, hood, and trunk assemblies, and the installation of the nonmoving front and rear windows. Adhesive bonding has become the preferred means of assembly for polymeric body panels made from sheet-molding compounds and reaction-injection-molded (RIM) materials. Moreover, since adhesive bonding has the ability to bond such a wide variety of materials,its use has grown significantly with the ever-expanding applications of plastics and composites.

ADHESIVE MATERIALS AND THEIR PROPERTIES In adhesive bonding, a nonmetallic material (the adhesive) is used to fill the gap and create a joint between two surfaces.The actual adhesives span a wide range of material types and forms, including thermoplastic resins, thermosetting resins, artificial elastomers, and even some ceramics.They can be applied as drops, beads, pellets, tapes, or coatings (films) and are available in the form of liquids, pastes, gels, and solids. Curing can be induced by the use of heat, radiation or light (photoinitiation), moisture, a o tivators, catalysts,multiple-component reactions, or combinations thereof. Applications can be full load bearing (structural adhesives), light-duty holding or fixturing, or simply sealing (the forming of liquid- or gas-tight joints). With such a wide range of possibilities, the selection of the best adhesive for the task at hand can often be quite challenging. The structural adkesives are selected for their ability to effectively transmit load across the joint; they include epoxies, cyanoacrylates, anaerobics, acrylics, urethanes,

91 5

Bonding, Mechanical Fasten~ng, and Jo~n~ng of Nonmetals

sihcones, high-temperature adhesives, and hot melts. Both strength and rigldity may important, and the bond must be able to be stressed to a high percentage of its maximu load for extended periods of t m e without failure. 1. Epoxies The thermosetting epoxles are the oldest, most common, and most diver of the adhesive systems; they can be used to join most engineering materialsinclu ing metal,glass, and ceramic.They are strong, versatile adhesives that can be designe to offer hlgh adhes~on,good tenslle andshear strength, toughness, hlgh r~gid~t)! Cree reslstance, easy curlng w ~ t h little shrinkage, good chemlcal resistance, and toleranc 10 ~ . ~ ~ Y : I tc.rnpr.r:~ti~~c\. I~LI \',~riour cpo~ic.; c.ln 112 u5r.l o\.cr Icmpcrcllurc r.lngc fro I ( filC 1,) . 500 k)..\itcr curm,: 1 1 10,1111 ~ c m [ ~ c r ; . t i ~ ~ . ~ ~ . \ h c ~ r 0 I can be as high as35 to 70 MPa (5000 to 10,006psi). Single-component epoxles use heat as the curing agent. Most epoxies, however, ar two-component blends involving a resin and a curing agent, plus posslble additlv such as accelerators, plastic~zers, and fillers that serve to enhance cure rate,flexlb ~ty, peel reslstance, Impact resistance, or other characterlstlcs. Heat may again be re qulrcd to dnve or accelerate the cure. Low peel strength and poor flexibility limit epoxy adheslves, and the bond strengt can be sens~tlve to molsture and surface contamlnatlon. Epoxies are often bnttle low temperatures, and the rate of curing is comparatively slow. Sufficient stren for structural dpphcations is generally achieved in 8 lo 12 hours, w t h full streng often requ~ring two to seven days.

2 . Cyonoacrylate~.These are liquid monomers that polymerize when spread into a th

film between two snrfaces.Tracc amounts of molsture on the surfaces pmmotecu m~ a t amazing speeds, oftcn in as llttle as two seconds Thus, the cyanoacrylatesoff a one-component adheslve system that cures at room temperature with no exte Impetus. Commonly known as suprrglues, thls family of adheslves is now availabl the form of hquids and gels of varymg vlscoslty, toughened versions deslgned to ov come brittleness, and even nonfrosting varieties. The cyanoacrylates provide excellent tensile strenglh, fast curing, and good shelf1 and they adhere well to most commercral plastlcs, metals, and rubbers. They a limited by their high cost, poor peel strength, and bnttleness. Bond properties ar poor at elevated temperatures, and effective curlng requires good component f (gaps must be smaller than 0.25 mm, or 0.010 m.).
3. Anaerohfcs.These one-component, thermosetting, polyester acrylics remain liqui when exposed to air When confined to small spaces and shut off from oxygen,as a jolnt to be bonded or along the threads of an inserted fastener, the polymer b comes unstable In the presence of Iron or copper, it polymenzes into a bonding-t resm, without the need for elevated temperature. Curing can occur across gap large as 1 mm, or 0.04 In. Addimes can ]educe odor, flammabiltt~ and toxlcity an can speed the curing operation. Slow-curing anaerobics require 6 to 24 hours toa tain useful strength. W ~ t h selected addit~ves and heat, however, curing can be r duced to as little as f ~ v e minutea The anaeroblcs are extremely versatile and can bond almost anythmg, including surfaces The joints resist vibrations and offer good sealing to molsture and 0th env~ronmental influences Unfortunately, they are somewhat hnttle and are llmite to service temperatures below 150C (300F).

4. Acryhcs.The acrylic-based adhesivec offer good strength, toughness, and versatih

and they are able to bond a variety of materials, including plastlcs. metals, cera and composites, even oily or drrty surfaces Most involve application systems w a catalyst primer (curing agent) IS applled to one of the surfaces to be joined and adhesive is applied to the other. The pretreated parts can be stored separately weeks wlthout damage. Upon assembly, the components react to produce a st bond at room temperature.Heat can often accelerate the curing, and at least on

r~ety cures with ultraviolet (UV) light. In comparison to other varieties of adhesives, the acrylics offer strengths comparable to the epoxies, flexlble bonds. good res~stance to water and humidity, and the added advantdges of room-temperature curing and a no-mix applrcation system. Major limitations include poor strength at high temperatures, flammab~lity, and an unpleasant odor when st111uncured.

5. urethane^. Urethane adhesives are a large and diverse tamily of polymers that are
generally targeted for applications that involve tempcratures below 65C (150F) and components that require great flexibility. Both one-part thermoplastic and two-part thermosettrng systems are available. Urethanes cure quickly t o handhng strength but are slow to reach the full-cure condition. Two mtnutes to handling with 24 hours to complete cure is common at room tempcrature Compared to other structural adhesives, the urethanes offer good flexibility and toughness, even at low operating temperatures. Thcy are somewhat sensitive to moisture, degrade in many chemical envnonments, and can involve toxic components or cunng products

6. Sclccone~. The silicone thermoseta cure from the moisture in the air or adsorbed mo~sture from the surfaces helng joined. They f o ~ m low-strength structural joints and are usually selected when considerable expansion and contraction are expected in the joint, flexibility 1s required (as in sheet melal parts); or good gasket, gap-filhng, or sealrng properties are necessary. Metals, glass, paper, plastics, and rubbers can all be joined. The adhesrves are relatively expensive, and curlng ir slow, but the bonds that ale produced can reslst moisture, hot water, oxidation, and weathering, and they retain therr flexibility at low temperature
7. High-temperature adhesives. Whcn strength must he retalned at temperatures in excess of 3U0C (500F), high-temperatu~e structural adhesives should be specified These include epoxy phenolics, modificd silicones or phenolic\ polyamides, and some ceramlcs High cost and long cure times are the major hm~tations for these adhesives, which see primary application in the aerospace industry.

8. Hot melh.. Hot-melt adhesives can be ubed to bond dissimilar substrates, such as plastics, rubber, metals, ceramics, glass, wood, and fibrous materials like paper, fabric, and leather. They can produce permanent or temporaly bonds, seal gaps, and plug holes While generally not comidered to be true structural adhesives, the hot melts are being used increasingly to transmit loads, especially in compositematerial assemblies. The joints can withstand exposure to vibration, shock, humidity, and numerous chemicals, and they offer the added features of sound deadening and vibxatlon damping. Most hot-melt adhesives are thermoplastic ieslns that are s o l ~ dat room temperature but melt abruptly when heated ~ n t o the range ol100" to 15O0C(200' to 300F). They are usually applied as heated liquids (between 160" and 180C) and form a bond as the molten adhesive cools and resohdifies.Another method of application is to position the adhes~ve in the joint prior to operatlonc, such as the paint-bake process in automobile manufacture. During the baking, the adhesive melts, flows into seam? and crevices, and seals against the entry of corrosive moisture. These adhesives contain no solvents and do not need time to cure or dry. Hot melts achieve over 80% of their bond strength withln seconds of solidification, but they do soften and creep when subsequently exposed to elevated temperatures and can become brittle when cold.

Additives also play alarge part in the success of industrial adhesives.They can Impart or enhance properties like toughness, joint durability, moisture resistance, adhesion, and flame retardance. Rheological add~tives and plast~cizers control viscos~t)~ and flow.Adhesives must penetrate the surfaces to be bonded but should not flow in an uncontrollable fashion. F~llers and extenders provide bulk and reduce cost.

Adhesive Bonding, Mechanical Fastening, and Joiningof Nonmetals

U.S. Consumption of adhesives and sealants by product, 2003

FIGURE 34-1 Distr~butlon among the common types of adhes~ves and sealants (Reprinted with perm~ssion from The Rauch Guide to the US Adhesives & Sealants Industry, Fifth Editton, 2006, Grey House NY). Publnhlng, M~flerton,

Radiation

A '

'- Dentallmedical
<1%

cured
1%

Conductive
4%

U.S. Consumption of adhesives and sealants by end-use market, 2003 (percentages by dollar value)

FIGURE 34-2 Distrlbutlon of adhesives and sealants by endwrth use areas. (Repr~nted permissran from The Rauch Gu~de to the U S Adhesives & Sealants Industry, Fifth Edrtion, 2006, Grey House Publrshing, Millerton, NY)

4%

Dentallmedical 4%

Figure 34.1 shows the distribution of various types of adhesive and sealant products for a recent year, and Figure 34-2 classilies adhesives by end-use markets.Table 34-1lists some popular structural adhesives. along with their service and curing temperatures and expected strengths. Table 34-2 presents the advantages and disadvantages of various curing processes.

NONSTRUCTURAL A N D SPECIAL ADHESIVES

There are a number of other types of adhesives whose limited load-bearing capabilities place them in a nonstructural classification. Nevertheless, they still play roles in manufacturing through a variety of uses, such as labeling and packaging.The hot-melt adhesives are often placed in this category but can be used for applications in both classifications. Evuporurive adhesives use an organic solvent 01.water base, coupled with vinyls, acrylics, phenolics. polyurethanes, or various types of rubbers. Some common evaporative adhesives are rubber cements and floor waxes. Pressure-sensitive adhesives are usually based on various rubbers, compounded with additives to bond at room

Adhesive Type
Butyral phenolic EPOXY Room-temperature cure Elevated-temperature cure Epoxy-nylon Epow-phenohc Neoprene-phenolic
NltnLe phenohc

Cure Temperature

Service Temperature
(OF)

Peel Strength at
Lap Shear Strength

P)
275 to 350

(ps~ at "FJ"
lOOOat 175 2500 at RT 1500 at 180 2500 at RT 1500at 350 2500 at RT 2000 at 180 6000 at RT 1000 at 175 2500 at RT 1000 a l l 8 0 2000 at RT 2000 at 250 4000 at RT 1000 at 1000 2500 at RT 1000 at 175 2500 at RT

Room Temperature (Ibitn )

10

-60 to 175

60 t o 90 200 to 350 250 to 350 250 to 350

-60 to 180 -60 to 350 -420 to 180

4 2 0 t a 500

275 L o 350
275 to 350 550 to 650 75 to 250

6 0 to 180
-60 to 250 -420 to 1000

Polylm~de Urethane *RT,room temperature

4 2 0 to 175

...
'~d?wsjve.~wing Processes
Curine Process
Mtx~ng reactwe components Anaerob~c cure Heat cure Molsture cure Light cure Surface-m~tlated cure Raprd cure, cure on demand Rapid cure Expenses for UV hght source, limted depth oi cure, most allow light to reach bond Poor depth of cure

Advantaees
Good shelf hfe, unltmlted depth of cure, accelerated w ~ t h heat Smgle-component adhesive, good shelf l ~ f e Unlunlted depth of cure, hzat csn ald adheston

Disadvantaees
Hlgh processmg costs, mlxral>ocnt~cal to perfomance Poor depth of cure,requtre pr~mer on many surfaces, senslt~ve ta surface contaminants Expenses for oven energy cosl, heat can adversely affect some substrates

temperature with a brief application of pressure. No cure is involved, and the tacky adhesive-coated surfaces require no activation by water, solvents, or heat. Peel-and-stick labels, cellophane tape, and Post-it notes are examples of this group of adhesives. Delayed-tack adhesives are similar to the pressure-sensitive systems but are nontacky until activated by exposure to heat. Once heated, they remain tacky for several minutes to a few days to permit use or assembly. While most adhesives are electrical and thermal insulators, conductive adhesives can be produced by incorporating selected fillers, such as silver, copper, or aluminum, in the form of flakes or powder. Certain ceramic oxide fillers can be used to provide thermal conductivity coupled with electrical insulation. Still another group of commercial adhesives are those designed to cure by exposure to radiation, such as visible, infrared, or ultraviolet light; microwaves; or electron

beams. These radmtron-cunng adhesrves offer rapid conversion from liquid to so11 room temperature and a curlng mechanism that occurs throu gressing from exposed surfaces (as with the competing low-tempera cures) Current applications mclude a wide varlety of dental amalgams that ca itles or seal surfaces while matching the color of the remaming tooth. In the manu turing realm, heat-sensitive materials can be effectively bonded, and the rapid cure t signhcantly reduces the need for fixtunng.

DESIGN CONSIDERATIONS The structural adhesives have been used for a wide range of applications in field diverse as automotive, aerospace, appliances, biomedical, electronics, construct machmery, and sportlng goods. Proper selection and use, however, requlres consldera of a number of factors, mcluding the following:

1. What materials are being joined? What are their surface finishes, hardnesses, poros~ties? Will the thermal expansions or contractions be different? 2. How will the joined assembly be used? What type of joint is proposed,what the bond area, and what will be the appl~ed stresses? How much strength is re WIN there be mechanical vibrahon, acoustical vibration, or impacts? 3. What temperatures might be required to affect the cure, and mght be encountered during service? Cons~deration should be given to thehi temperature, lowest temperature, rates of temperalre change, frequency of ch duration of exposure to extremes, properties required at the various conditions, differential expansions or contrachons 4. Will there be subsequent exposure to solvents, water or humidity, fuels or oils,I ultraviolet radiation, acid solutions, or general weathering? 5. What is the desired level of flexlbil~ty or stiffness? How much toughness is req 6. Over what length of time is stability desired? What portion of this time wil under load? 7. Is appearance important? 8. How w11 the adhesive be apphed? What equipment, labor, and sk~ll are reqw 9. Are there restrictions relating to storage or shelf life? Cure time? Diapos recyclability? 10. What will it cost?
Because there is such a large difference in bond are often classified as either continuous surface or co bonds, both of the adhering surfaces are relauvely large and are of the same shape. Core-to-face bonds have one adherend area that is very small compare other, like when the edges of lightweight honeycomb core stm face sheets (see Figure 15-19). A major design consideration for both types is the nature of joint will experience. As shown in Figure 34-3, applied stresse tension, compressron, shear, cleavage, and peel. Most of the stru significantly weaker in peel and cleavage than Lhey are in shear hesively bonded joints should be designed so as much of the stress tension,or compression,where all of the bonded area shares equal The shear strengths of structural adhesivesrange from 14 to 40 room temperature, wh~le the tensile strengths are only 4 to 8 best adhesive-bonded jomts, therefore, will be those that are rior shear strengths. Creep, vibration and associated fatigue, therm ical shocks can all induce additional stresseswhen vibration the elastomeriric adhesives are quIte attractive, since they ca Figure 3 4 4 shows some commonly used joint designs and i effectiveness.The butt joint is unsatisfactory because it offers only a surface area and little resistance to cleavage. Useful strength 1s gen

Peel

t
Cleavage

Types of stresses in adhesive-bonded loints.

FIGURE 3 4 3

increasing the bond area through the addition of straps or the conversion to some form of lap design. The scarf joint, shown previously in figure 34-14, is also used when uniform thickness is required. Figure 34-5 shows some recommended designs for corner and angle joints. Adhesives can also be used in combination with welding, brazing, or mechanical fasteners. Spot welds or rivets can provide additional strength or simply prevent movement of the components when the adhesive is not fnlly cured or is softened by exposure to elevated temperature. To obtain satisfactory and consistent quality in adhesive-bonded joints, it is essential that the surfaces be properly prepared. Procedures vary widely but frequently iuclude cleaning of the surfaces to be joined. Contaminants, such as oil,grease, rust, scale, or even mold-release agents, must be removed to ensure adequate wetting of the surfaces by the adhesive. Solvent or vapor cleaning is usually adequate. Chemical alteration of the surface to form a new intermediate layer, chemical etching,steam cleaning, or abrasive techniques may also be employed to further enhance wetling and bonding. While thick or loose oxide films are detrimental to adhesive bonding, a thin porous oxide or surface primer can often provide surface roughness and enhance adhesion.

W T H A V T 9 4 unsattsfactory
a

Butt-

Half lap-excellent, but requires machtn~ng

Plain lap-good

Strapfair

Double lap-good,
but difficult

Double strap--good

to balance load

FIGURE 344 Possible des~gns

of adhesive-bonded jolnts and a

rating of thew performance in rervlce.

Joggle lap-very good

922

CHAITER 34

Adhes~ve Bonding, Mechanical Fastening, and Joining of Nonmetals

Bond line

L
Bond lhne
Bond line

FIGURE 34-5 Adhes~vely bonded corner and angle jofnt designs.

The destructive testingof adhesivejoints, or the examination ofjoint failures,eanr veal much about the effectivenessof an adhesive system. Ifkiilure occurs by separation the adhesive-substrate mterface, as shown ID 13gnre 34-6a, it is indicative of a bonding I adhesion problem. If the failure lies entirely within the adhesive, as tn Flgugu~e 3% thf the bonding with the substrate is adequate, but thestrength of the adhesive may need1 be enhanced.Emally,if failure occurs within the substratematerials,asinFigure346c,fi jo~nt 1 s good, and failure is unrelated to the adhesive bonding operation.

Adncsve Subsrrate

FIGURE 346 Failure m6des of adhesive joint: (a) Adhesive failure, (b) cohesive failure within theadheive, and (c)cohesive failure within the
substrate.

ADVANTAGES AND LIMITATIONS Adhesive bonding has many obvious advantages. Almost any material or combination of materials can be joined in a wide variety of sizes, shapes, and thicknesses. For most adhesives, the curing temperatures are low, seldom exceeding 180C (350F).A substantial number cure at room temperature or slightly above and can provide adequate strength for many applications. As a result, very thin or delicate materials, such as foils, can be joined to each other or to heavier sections. Heat-sensitive materials can be joined without damage, and heat-affected zones are not present in the product. When joining dissimilar materials that experience changes in temperature, the adhesive often provides a bond that can tolerate the stresses of differential expansion and contraction. Because adhesives bond the entire joint area, good load distribution and fatigue resistance are obtained, and stress concentrations (such as those observed with screws,rivets, and spot welds) are avoided. Because of the high extension and recovery properties of flexible adhesives, the fatigue resistance can be up to 20 times that of riveted or spot-welded assemblies.The large contact areas that are usually employed provide a total joint strength that compares favorably with alternative methods of joining or attachment. Shear strengths of industrial adhesives can exceed 20 MPa, or 3000 psi, and additives can be incorporated to enhance strength, increase flexibility,or provide resistance to various environments. Adhesives are generally inexpensive and frequently weigh less than the fasteners needed to produce a comparable-strength joint. In addition, an adhesive can also provide thermal and electrical insulation; act as a damper to noise, shock, and vibration: stop a propagating crack and provide protection against galvanic corrosion when dissimilar metals are joined. By providing both a joint and a seal against moisture, gases, ,and fluids, adhesive-bonded assemblies often offer improved corrosion resistance throughout their useful lifetime.When used to bond polymers or polymer-matrix composites, the adhesive can be selected from the same family of materials to ensure good compatibility. From a manufacturing viewpoint, the formation of a joint does not require the capillary-induced flow of material, as in brazing and soldering. The bonding adhesive is applied directly to the surfaces and the joint is then formed by the application of heat andlor pressure. Most adhesives can be applied quickly, and useful strengths are achieved in a short period of time. Some curing mechanisms take as little as two to three seconds! Surface preparation may be reduced since bonding can occur with an oxide film in place, and rough surfaces are actually beneficial because of the increased contact area.Tolerances are less critical since the adhesives are more forgiving than alternative methods of bonding.The adhesives are often invisible; exposed surfaces are not defaced; smooth contours are not disturbed; and holes do not have to be made, as with rivets or bolts. These factors contribute to reduced manufacturing costs, which can be further reduced through the elimination of the mechanical fasteners and the absence of highly skilled labor. Bonding can often be achieved at locations that would prevent the access of many types of welding apparatus. Robotic dispensing systems can often be utilized. The major disadvantages of adhesive bonding are the following:
1 . There is no universal adhesive. Selection of the proper adhesive is often complicated by the wide variety of available options. 2. Most industrial adhesives are not stable above 180C (350F). Oxidation reactions are accelerated, thermoplastics can soften and melt, and thermosets decompose. While some adhesives can be used up to 260C (500F),elevated temperatures are usually a cause for concern. 3. Some adhesives shrink significantly during curing. 4. High-strength adhesives are often brittle (poor impact properties). Resilient ones often creep. Some become brittle when exposed to low temperatures. 5. Surface preparation and cleanliness, adhesive preparation, and curing can be critical if good and consistent results are to be obtained. Some adhesives are quite sensitive to the presence of grease, oil, or moisture on the surfaces to be joined. Surface roughness and wetting characteristics must be controlled. 6. Assembly times may be greater than for alternative methods, depending upon the curing mechanism. Elevated temperatures may be required, as well as specialized fixtures

e Bonding, Mechanical Fastening, and Jo~ning of Nonmetals

7. It 1s difficult to determ~ne the quality of an adhesive-bonded joint by tradmtmonal destruct~ve techniques, although some inspection methods have been develope give good results for certain types of lolnts 8. Some adhesives contain objectionable chemicals or solvents, or produce the upon curing 9. Many structural adhesmves deteriorate under certain operatmng condmtions. Enm ments that may be part~cularly hostile mnclude heat, ultravlolct light, ozone, acid (low pH), water and hum~dmty, salt, and numerous solvents Thus, long-term dura ity and reliabihty may be questioned, and l ~ f expectancy e is hard to pred~ct. 10. Adhesively bonded jo~nts cannot be readmly disassembled.

Nevertheless, the extensive and successful use of adhes~ve bondmng provldes amp evtdence that these lmmitatmons can be overcome if adequate quahty control procedur are adopted and foIlowed.

FASTENING
INTRODUCTION AND METHODS Me~hanicalfast~~zmy is a classthcatlon that ~ncludes a wide variety of techniques and f teners designed to s u t the indmvidual requirements of a multmtude of jo~ntsand asse bhes Included withln this family are Integral fasteners,threaded d~screte fasteners (wh mcludes screws,bults,studs, and inserts),nonthreaded &mete fasteners (such as rrvetqp retaintng nngs, natls, staples, and wire stitches), special-purpose fasteners (such as qumck-release and tamper-resistant types), shrmnk and expansmon fits,press fits,sedms, others. Select~on of the specific fastener or fastening method depends primarily on materials to b e joined, the function of the joint, strength and reliability requiremen me~ght Iimitations, dimensions of the components, and enmronmental condthons conaiderations include cost, tnstallation equipment and access~bil~ty, appearance, an need or desue for di~assembly.When disassembly and reassembly are desired (as for replacement. mamtenance, or repair), threaded fasteners,snap-fmts,or other fastener can be removed qu~ckly and easily should be .~pec~fied. Such fasteners should not ha tendcncy to loosen after installation,however If disassembly is not necessary, perman fasteners are ohen preferred, or threaded fastenerscan be coupled with anaeroh~c ad sives that cure to full strength at room temperature and "lock" the fastener in place A mechanical jolnt acquires its strength through e~ther mechamcal interlochg interference of the surfaces as a result of a clamping force. No fusion or adhesion oft surfaces is rcqu~redThe fasteners and fastening processes should be selected to vide the required strength and properties in view of the nature and magnitude of sequent loading. Connder the possmbility of vibrations andlor cycl~c stresses that m' promote loosening over a period of time. Added wetght may be a significant fact certain applications, such as aerospace and automotmve.The need to w~thstand c hive envrronments, operate at high or low temperatures, or face other severe condi may be addmtional constraints to the selectman of fasteners or fastening processes. The cffect~veness of a mechanmcal fastener often depends upon (1)the maten the load-bearing area of the head), (3) the fastener, (2) the fastener destgn (includ~ng preparation, and (4) the installation procedure.The general desire IS to ach~eve a form load transfer, a minimum of stress concentratmon, and uniformity of installat torque or interference fmt. Various means are ava~lable for achieving these goals descr~bed in the fallowing paragraphs. Integralfasteners are formed areas of a component that ~nterfere o r mterlock other components of the assembly,they are most commonly found in sheet metalp FIGURE 34-7 Several types of ucts Examples include lanced or shear-formed tabs, extruded hole flanges, ernbos integral fasteners: (a) lanced tah to fasten wires or cables to sheet protrusmons, edge seams, and cnmps. Figure 34-7 shows some of these techniqnese or plate; (b) and (c) assembly of which mnvorve some form of metal shearing andlor form~ng The common beve through folded tabs and slots for can mncludes several of these jomts-an edge seam to join the top of the can to the b different types of load~ng, (as in Figure34-7e) and an embossed protrusion that is subsequently flattened to at (d) use of a flattened embossed the opener-tab [as in Fmgure 34-7d). protrusion: (e) single-lock seam.

Cutaway section

Nail

Screw

Bolt and n u t

After Plates Fracture pin

Before

ARer Before One-slde rivet

Two-s~de rlvet

Locked

Unlocked (turned 90) Qu~ck-release fastener

Snap-ht

Discrete fasteners, like those illustrated in Figure 34-8, are separate pieces whose function is to join the primary components.These include bolts and nuts (with accessory washers, etc.),screws, nails, rivets, quick-release fasteners, staples, and wire slilches. Over 150 billion discrete fasteners are consumed annually in the United States, with a variety so immense that the major challenge is usually selection of an appropriate, and hopefully optimum, fastener for the task at hand. Fastener selection is further complicated by inconsistent nomenclature and identification schemes. Some fasteners are identified by their specific product or application, while others are classified by the material from which they are made, their size, their shape, their strength, or primary operational features.' The commercial availability of such a wide range of standard and special types. sizes, materials,strengths, and finishes virtually ensures that an appropriate fastener can be found lor most all joining needs. Discrete fasteners arc easy lo install, remove, and replace. In addition,most standard varieties are interchangeable. Steel is the most common material due to its high strength and low cost.Various finishes and coatings can be applied to withstand a multitude of service conditions. Shrink cmd expunsion fits form another major class of mechanical joining. Here, a dimensional change is introduced to one or both of the components by heating or cooling (heating one part only, heating one and cooling the other, or cooling one). Assembly is then performed, and a strong interference fit is established when the temperatures return to uniformity. Joint strength can be exceplionally high. A corrosion-resistant cladding or lining can also be applied to a less costly bulk material.
I

Discussion of thenrimarv tcrrns used in identif~ine discrete lasleners can he found in ANSI Standard 818.12.

CHAPTER 34

Adhesivt Bonding, Mechanical Fastening, and Joiningof Nonmetals

Press fitr are s~mllar to shrink and expansion hts, but the results are obtam through mechanical force instead of differential temperatures.

REASONS FOR SELECTION Mechanical fastening offers a number of attractwe features, among them the followin

1. They are easy to disassemble and reassemble.The threaded fasteners are notew thy for thls feature, and semipermanent fasteners (such as rivets) can be drilled o for a major disassembly. 2. They have the abil~ty to join similar or different materials in a wide variety of siz shapes, and joint designs Some joint designs, such as hinges and slides,permlt hmit motion between the components. 3. Manufacturing cost is low. The fasteners are usually small formed components th cost little compared to the components being jomed. They are readily available in vanety of mass-produced sizes. 4. Installation does not adversely affect the base materials,as 1s often the case w~th te n~ques involving the apphcation of heat andlor pressure 5. Little or no surface preparation or cleaning is required
MANUFACTURING CONCERNS Many mechanical fasteners require that the components contain aligned holes. Cas lngs, forgings, extrusions, and powder metallurgy components can be des~gned to ' clude integral holes. Holes can also be produced by such techniques as punching, drillin and electrical, chemical, or laser-beam machining. Each of these techniques produ holes with characteristic surface finish, dimensional features, and properties. Second operations, such as shaving deburring, reaming, and honmg, can be used to impr precision and surface finish. Hole making and the proper positioning and alignme the holes are major considerations in mechanical fastening. Some fasteners,such as bolts coupled with nuts, requlre access to both sides of assembly during jornmg. In contrast, screws offer one-side joining. If a bolt can he inser Into a threaded (tapped) hole, however, the nut can be eliminated, and only one-side cess is required. If the bolt or screw is sufficiently hard, the fastener can often own threads, thereby eliminating the need for a threaded receptacle.This selffeature is particularly attractive when assembling plastic products. Stapling~s a fast way of lorning thin materials and does not require pnor ing. Rivets offer good strength but produce permanent or semipermanent jo fits utihze the elasticity of one of the components, but the necessary elastic de must be possible without fracture. DESIGN AND SELECTION The design and selection of a fastening method requires numerous considerations, clud~ng the posslble means of jo~nt failure. When a product is assembled with faste joints, the fasteners are extremely vulnerable sites. Mechanical joints generally fail b cause of oversight or lack of control in one of four areas: (1)the des~gn of the fasten itself and the manufacturing techniques used to make it, (2) the mater~al from whi the fastener is made, (3)joint design,or (4) the means and details of installation.Fasten may have tnsufficient strength or corrosion resistance or may be subject to stress c rosion cracking or hydrogen embntt1ement.They may be unable to withstand the t perature extremes (both high and low) experienced by the fmal assembly.Metal faste provide electrical conductivity between the components, and an inappropriate choic fastener or component material can cause severe galvanrc corrosion. Nonmetallic teners (such as threaded nylon) can be used for low-strength applicat~ons where co slon is a concern, but creep under load 1s a concern for these matenals Since mecha fasteners only join at discrete points, gases or liquids can easily penetrate the joint a and further aggravate conditions. Many failures are the result of poor joint preparation or improper fastener stallation. A high percentage of the cracks m aircraft structures originate at fasten

SECTION 34.3 joining of Plastic\

927

holes, and tat~gue of fasteners 1s the largest slngle cause of fastener failure. Installation frequently imparts too much or too little preload (too tight or too loose).The joint surfaces may not be flat or paraIlel, and the area under the fastener head may be msuffinent to bear the load. Vibrational loosening enhances fastener fatigue.The details of joint design should further cons~der stress diitnbution, since much of the load will be w n centrated on the fasteners (in contrast to the previously discussed adhesive joints that d~stribute the load uniformly over the entlre joint area) Nearly all fastener failures can he avoided by proper design and fastener selection Consideration should be given to the operatrng environment,required strength, and magnltude and frequency of vibration Fastener design should Incorporate a shank-to-head fillet whenever possible. Rolled threads can be specified for their supenor strength and fracture resistance. Corrosion-resistant coatlngs can be employed for enhanced performance. Jolnt design should seek to avoid such features as offset or oversized holes.Proper installation and tightening are cntical to good performance. Standard size$ shapes, and grades should be used whenever possible, with as little variety as is absolutely necessary.

Mechanrcal fasteners, adheslves, and welding proceqses can all be employed to form joints between engineering plastics. Fasteners are quick and are sultable for most materials, but they may be expenslve to use, they generally do not provrde leak-tight joints,and the localized stresses may cause them to pull free of the polymeric mater~al.Threaded metal Inserts may have to be rncorporated into the plastic components to receive the fasteners. further increasing the product cost. Adhesives can provide excellent properties and fully sound joints, but they are ottcn difficult to handle and relatively slow to cure. In addrtion,cousiderable attention is requ~red in the areas of ~ o i npreparation t and surface cleanhness. In a modification of adhesive bonding, solvents may he used to soften surfaces, whlch are then pressed together to form a bond. Welding can be used to produce bonded jolnts with mechanical properties that approach those of the parent mater~al. Unfortunately, only the tlzermoplrrst~cpolymer~ can be welded, since these materials can be melted or softened by hzat without degradatlon and good bonds can be formed w ~ t h the subsequent application of pressure. The thermorettrngpolymer~do not soften w ~ t h heat, tending only to char or butn, and must b e loined by alternat~ve methods, such as mechanical fasteners, adheslves, snap-bts, or poss~ble co-curing (placing the components together and cunng whlle rn contact). Because the thermoplastics soften and melt at such low temperatures, the heat requrred to weld these matenals is s~gnificantly less than that required in the welding of metals. The processes used to weld plastics can be divided into two groups: (1) those that utilize mechanical movement and friction to generate heat,such as ultrasonic welding, spin welding, and vibration weldmg. and ( 2 )those that involve external heat sources, such as hot-plate welding, hot-gas welding, and res~stive and Inductive ~mplant welding. In both groups, it is important to control the rate of heating Plastlcs have low thermal conductlvlty, and it is easy to induce burning, charring, or other material degradation before softening has occurred to the desired depth. Ulfrason~c welding of plastics uses high-frequency mechameal vibrations to create the bond. Parts are held together and aresub~ected to ultrasonicv~brations(20 to 40 k I 3 frequency and 10 to 1M) micron amplitude) perpendicular to the area of contact The high-frequency stresses generate heat at the jourt interface su&c~ent to produce a hlgh-quality weld 1n a perlod of ' 1, to l1I2seconds.The process can be readily automated, but the tools are expenswe and large production suns are generally required. Ultrasonic welding is usually restricted to small components where relatlve movement 1s restricted and weld lengths do not to exceed a iew centuneters. In vibration weldlng, or linear friction welding, relative movement between the two parts is again used to generate the heat, hut the direction of n~ovement is now parallel to the interface and aligned with the longest d~mension of the joint.7le vibration

928

CHAPTER 34

Adhesive* Bondlng, Mechanical Fasten~ng, and Joining of Nonmetals amplitudes are 10 tlmes larger than m ultrasonic welding, and the frequencies are co siderably less (on the order of 100 to 240 Hz).When molten material is produced vlbrat~on is stopped,parts are aligned,and the weld reglon is allowed to cool and soli The entue process takes about one to five seconds. Long-length, complex joints can produced at rather high production rates Nearly all thermoplastics can be joined, dependent of whether their prior processing was by lnjection moldtng, extrusion, blo molding, thermoforming, foaming, or stamping. The frrcfion weldrng of plastlcs (also called spin wrldrng) is smilar to vlbratio welding, except the relahve motlon is now continuous and rotatlonal.The process is sentially the same as the friction welding of metal&but melting now occurs at the interface High-quality welds are produced with good reproducibility, and little en preparation is required.The maJor hrnitat~on e that at least one of the componentsm exhibit circular symmetry, and the axis of rotation must be perpendicular to the surface. Weld strengths vary from 50 to 95% of the parent material in bonds of plast~c. Joints between dissimilar materials generally have poorer stren~ths Butt we1 can he made between plates of thermopl~stic using the friction-sttr welding described in C h a ~ t e 32. r Hot-plate welding uses an external heat source and is probably the simplest of mass-product~on tech~uques used to jom plastics.The parts to be joined are held I tures and pressed against the opposite sides of an electrically heated tool. Con maintained until the surfaces have melted and the adjacent material has softene spec~fied distance Gom the interface.The parts separate, the tool is removed, and the prepared surfaces are pressed together and allowed to cool. Contammated surface terlal is usually displaced into a flash region. Weld times are comparatively slow, mg from 10 seconds to several minutes, The joint strength can be equal to that parhnt material, but the joint des~gn is usually limited to a square-butt configuratio that encountered when joinmg sectioas of plastic plpe. If the bond interface has a n flat profile, shaped heatrng tools can be employed. Heated-tool welding can also to produce lap seams between flexible plastic sheets Rollers apply pressure after terial has passed over a healer. The hot-gas welding of plastics is similar to the oxyacetylene welding of me and V-groove or fillet welds are the most common jolnt configuration. A gas (c pressed air, nitrogen, hydrogen, oxygen, or carbon d~oxide) is heated by an electric as it passes through a welding gm,like the one shown m F'igure 34-9.The hot-gasst emerges from the gun at 200" to 300C (400" to 570F) and Impinges on the joint Thin rods of plastic material are heated along w ~ t h the workpiece and are then f Into the softened jolnt area,providing both the filler materlal and the pressur to produce coalescence. Because this process 1s usually slow and the results a ally dependent on operator s k i , rt is seldom used in production applications.It IS, ever, a popular process for the repait of thermoplast~c materials. Extruszon weldrng is similar to the hot-gas process, except that the external f material rod 1s replaced by a stream of fully molten polymer that emerges from weld tool as it moves along the jomt.This is simlar lo the wire feed in the metaling proceyses. In the implant welding of plastics, metal inserts are placed between the p be joined and are then heated by means of induction (rad~o frequency) or res' heatlng. The resistance method requires that a current-carrying path exist to c the current to the implants, a feature that is not required for induction beating. thermoplastic materlal melts around the heated implants and tlows to form a j Since a weld forms only in the vicin~tyof the implants, the process resembles weldlng and produces joints that are cons~derably weaker than those forme processes that bond the entlre contact area. When bonding is desired over larger tapes, rods, or gaskets of thermoplast~e matenal laced w ~ t h iron oxide or metal cles can be used to concentrate the heat at the interface and provide filler mater In all cases o f implant weldlng, the metal implant material remains as an integralp of the final assembly.

FIGURE 34-9 Using a hot-gas torch to make a weld in olastic plpe.

SF.CTION 34.5 join~ngof Composites

929

Still other processes to weld thermoplast~cs are based on rnfrared radiation or microwave heatzng. Laser welding has also been performed on plastics The most common method of joining plastic> to one another and joining metal to plastic is through mechanzcal fasteners. It is important that the plastic be able to withstand the straln of fastener insertlon and the localized stresses around the fastener. Conventional machine screws are rarely used, except with extremely strong plastics. Instead, there are a number of fasteners designed specifically for use with plastics.Threaded fasteners work best w ~ t h thlck sections. Self-tapplng, thread-cnttlng screws are used on hard plastics, and thread-forming screws are used w t h softer materials. If the joint is to undergo disassembly and reassembly, threaded metal inserts may be ~ncorporated Into the part to receive the fasteners Because of the low elastic modulus of plastic materlals, snap-fit assemblies are often an attractwe alternatrve to the use of fasteners. As descnbed previously in this chapter, adhesives provide an attractive means of joining plastics. Slnce adhesives are polymeric materials, the joint material can be selected for compatlbllity with the mater~al belng joined. Probably the most common application 1s the adhesive jolnlng of PVC (polyvinyl chloride) pipes and fittings in household plumbing.

34.4

JOINING OF

CERAMICS AND GLASS

The properties of ceramic materials are s~gn~ficantly different from the engineering metals, and these differences restrict or lim~t the processes that can be used for joining. High melting temperatures can be a s~gnificant deterrent to fusion weldmg. More sign~ficant, however, are the effects of low thermal conductivity and brittleness Heating and cooling will l~kely result In nonuniform temperatures, and the thermally Induced stresses are likely to result in cracking oc fracture. The lack of useful ductility virtually eliminates any form of deformatlou bondlng. Mechanical fasteners, and their assoc~ated threads and holes, create high concentratedstresses, and these stresses often lead to material fracture.As aresult,most ceramic materlals are jo~ned by some form of adheslve bonkng, brazing, hffusion or slnter bondmg, or ceramic cements. Adhesives and cements are probably the most common melhods of jolning ceramics to ceramlcs, ceramics to glasses,and ceramlcs to metals and other materials. The inserted materlal (polymer adheslve, glass or glass-ceram~c fnt, or ceramic cement or mortar) w~ll bond to the surfaces and bridge what are often radically different composltio~~s and structures. Brazing and soldering use a low-melting-point metal or lower-melting ceramic as the mtermed~ate material. Some materials, such as indium solders, dlrectly wet ceramlc sulfaces.To promote adhesion and bonding with other jolnt materials, it may be necessary to first coat the ceramic with some form of metallized or deposited layer. These coatings bond to the ceramic and the braze or solder material bonds to the coating. Slnter bonding is a means of joining ceramic materlals during the11inltial production. As the component pieces are held together and co-Elred, diffusion bonds form across the interface while similar bonds form w~thln the components. Thib process is best performed when the components are of ldentlcal material or materials wlth simllar composition and structure. Various Intermediate materials have been used to assist the joining of diss~milar ceramlcs. The joining of glass is a much easier operation. Heating Softensthe two materials, wh~ch are then pressed together and cooled. A wide variety of heating methods are used, depending on the size,shape, and quantity of components to be joined.

34.5

JOINING OF

COMPOSITES
The joimng of composite materials can be an extremely complex subject, especially when one considers the variety of composites and the fact that the joint mterface 1s likely to be a distinct disrupt~on to the contlnulty o f structure and properties.

930

CHAmkn 34

Adhesive Bonding, Mechan~cal Fasten~ng, and Joining of Nonmetals Particulate composites may have the least structural difference a t an interface. Laminar composites will certainly behave differently it the j o ~ n surface t 1s a core (or multilayer) surface o r a single-material exposed face, Fiber-reinforced composites, regardless of the type of fiber and fiber configuration, will certainly lack fiber continuity across the joint. The usual joimng methods tend to be those w e d with t h e matrix of the composite. Metal-matrix composites can be welded, brazed, or soldered, or joined with screws or bolts o r any of the other techmques applled to metals.The techmqnes for polymermatrix and ceramic-matrix follow those for plastics and ceramics, as discussed previously in this chapter. Theoretically, all composites can be adhesively bonded, but there may be limits set by the applied stresses, operating temperatures, o r size of the workpiece (Since many adhesives require a thermal cure). When workmg with polymer-matr~xcomposites, the adhesive is often selected to match or be compatible with the matrix polymer I n many cases, however, the adhesive is being asked to bond t o already-cured polymer surfaces.

Key Words
adherend adhesive adhesive bonding anaeroh~c cleavage continuous-surfacebonds core-to-face bonds curlng d~screte fasteners elastomer epoxles evaporative adhesives extrusion welding fnction weldlng hot-gas weldlng hot-melt adhes~ves hot-plate welding implant weldmg ~ntegral fasteners mechanical fastening peel press fit pressureaensrtive adhesives rad~at~on-cur~ng adhesive shear shr~nk fit srnter bonding s p ~ welding n siructural adhesive superglue thermoplastrc thermoyetting ultrason~c weld~ng nhratiou weldmg

Review Questions
I. What would he some of the charactenstlcs of an ideal adhesive?
2. What are some of the newer applications that have helped

promote mcreased use of adhesive bonding? 3 What are some of the types of mater~als that have heenused as industrial adhesives? 4 What are some of the ways in which adhesives can be cured? 5. What IS a structural adhesive? 6 Characterue the temperature range over wh~ch epoxles nught he used, typical values of shear strength, and commonly observed cunng times 7. What promotes the curing of cyanoacrylates?OI anaerohics? 8 What features or characteristicsmight favor the select~on of a sil~cone adhesive') 9 What are some common apphcations of hot-melt adhesives? 10. What features or properties are provlded or enhanced by adhesive addruves? 11. What are some types of nonstructural or spec~aJ adhesives? 12. How can polymeric adhesivesbe made electrically or thermally conductive? 1 3 . What are some of the temperature cons~derations that apply when selecting an adhes~vev 14 What are some of the environmental conditrons that might reduce the performance or lifet~me of a structural adheslve? 15. Why is it desirable for adhesive jo~nts to be des~gned so the adheslve 1s loaded in shear, tension, or compression? 16. Why are butt lomts unattractive for adhesive bonding? 17. What types of joints provide large bondlug areas?

18. What are some wrnmon techniques by which surfaces are prepared for adheave hond1ng7 19. Why are the structural adhes~ves an attractive means of loinmg d~sslmilar metals or materials7 D~Llereui sues or th~cknessesq 20 What are some of the other attractive properties of structural adhesives? 21. In what ways might a structural adhes~ve offer manufacturing ease or reduced manufactunng mst? 22 In view of the relatively L o w strengths of the structural adheslves, how can adhesively bonded lolnts attan strengths comparable to other methods of loinmg? 23 Why are adhesive lomnts unattractive for applicatlons that involve exposure to elevated temperature? Is low temperature a concern? 24. What factors would influence the selection of a specfic ttype of mechamcal fastener or fastening method? 25 What types of fasteners are attracuveif the apphcat~onrequues the abhty to d~yassemble and reassemble the product7 26 What factors determine the overall effect~veness of a mechanical fastener? 27. What is an integral fastener? Provide an example 28 What are some of the primary types of dlscrete fasteners? 29 How are press fits s~m~lar to shrink or expans~on fits? HOW do! they differ? 30. What are some of the major assets of rnechmcal fasteners? 31. What are some of the ways that fastener holes can be made i n manufactured products7

Problems 32. What are some of the common causes for failure of mechanically fastened joints? 33. From a manufacturing viewpoint, why is it desirable to use standard fasteners and minimize thevariety of fasteners within a given product? 34. Why can the thermoplastic polymers be welded, but not the thermosettina - varieties? 35 Descnbe several of the plastlc joinang processes that use mechan~cal movement or fr~ction to generate the requred heat.

931

36. What are some of the external heat sources that can be used in the welding of plastic materials? 37. Why are the crystalline ceramic materials particularly difficult to join? 38. What is sinter bonding, and how does it join ceramic materials? 39. When materials are joined, we create interfaces between the various components. Describe the structural features that might result at the interface when we jom: (a) particulate composites, (b) lamrnar composites,and (c) fiber-reinforced composites.

II Problems
1. Some automakers are using adhesives and sealants that cure under the same conditions usedfor thepaint-bake operation. Determine the conditions used for paint-bake, and identify some adhesives and sealants that could be used. What are some of the pros and cons of such an integration? 2. A contractor has installed aluminum siding on a house with steel nails. Use the galvanic series to evaluate the corrosion properties of this assembly. (Note: The aluminum is exposed to air,so it should be considered to be in its passive condition.) What do you expect will be the outcome of this fastener selection? Can you recommend a better alternative? 3. Mechanical fasteners are an attractive means of joining composite marerials because they avoid exposing the composite to heat andlor high pressure. Assume that the composite is a polymer-based fiber-reinforced material with either uniaxial or woven fibers. For this particular system, what are some possible fastener-related problems? Considerjoint preparation, assembly, and possible service failure. 4. The heat-resisting tiles on the U.S.space shuttle are made from heat-resisting ceramics. Determine the method or methods used to attach them to the stmcture.What difficulties or problems have been encountered relating to this bonding? 5. The bicycle frames used by riders in recent Tour de France races have been single-piece fiber-reinforced composite% What difficultiesor property compromisesmight be associated with a fabrication method that uses joints? What methods might be available to join the carbon fiber-epoxy composite materials commonly used in these bicycles? 6. The processes describedfor the joining of plastics focused almost exclusively on the thermoplastic polymers. What types of joining techniques could be applied to thermosetting polymers? To elastomeric polymers? To ceramic materials?

932

CHAFTER 34 Adhesive Bonding, Mechanical Fastening, and Jorningof Nonmetals

ou are employed by a small manufactufer of sportrng goods equipment, and yourdes~gn team has recently proposed a new line of hlgh-pertprmance golf clubs. While the dub head of the imns rs to be a "standarrf" AlSl431 martensitrc stamless steel investwent oasttng, the stnklng face w~ll mco?porate a metal insert, as shown In the figure. This tnsert will be prgduced by p~wder metallurgy and will consist of a copper-based alloy laced wtth pre~~zed particles of tungsten oarbtde. After a mild and etching of the copper matnx, the carbide particles will protrude suffr clently to better grtp the wsurface of the ball, irnparttng an enhanced amount of backspin to better control the "bite"of the bdl upon landing. Since ~ t s purpose IS to modify the strlkng face, the Insert is rather thin, about 1.5 fo 3 mm ('I,, to 'I, m.). It is rmportant that the insert be incorporated into the club face in a manner that does not dampen the impact or campfmfse tha ?eel" of the club. 1. You must devise a means of incorporatlngthe praposed insert Into the face of the club. What are some possible maans of lointng or bonding the dtssimtlaf matertak? What are the advantaws and limltabons of mch of your alternatives? What would you recommendQ 2. An additional joint oceurs where the club head rs attached to the shaft. I 7 a graphtteftber-retnforced epoxy

IS being considered fcr the shaft, how might d be ailached to the starnless steel head? If the mmpoxe shaft were seleerad, which joining method would you recomrwnd7 if the shafts were metal, what other methods mtght be posslble7 3. For production simplicity: t t might be preferableto use me same lointng procedure at all locations. In Vtew o f you( answers to Questions 1 and 2, does this appear to be a posstb~ltty for thrs product? 4\ If the bonding processand resuitlng interfaceprove to be pcohlematic, there may be other ways to producea ratsed-carblde surface on a SYainless steel golf dub face. Gons~der processes such as thsrmal spray, frict~on-stir to embed particles, and others. What do you see as the acimfltages and ltmltations for each of these alternatives, oonsidenng both manufacturing and parformanoe?Would you expect them to be OheaperOr than the proposed . ' Ifqua'lty and pelformance were the Rrrmary objmlve, which O f The opttonsin Question 4 would yw recommend?Why? 6. If ccnf mtnim~sat~on were to became Importantas you seek an edge over competitors, what would be your recommendat~on and why?

35.1 I~~TRODULTION 35 ? . MECHANICAL CLEANING AND FINISHING BLAST CLEANING Barrel Fmishlng or7llmbllng Vibratory 13nishing Media Compounds Summary of Mass-Fmishing Methods Belt Sanding Wire Brushing Buffing Electropolrshing 35.3 CHEMICAL CLEANING Alkaline Cleaning Solvent Clcanmg

Vapor Degreasmg Ultrasonic Cleanmg Acid Pickling 35.4 COATINGS Parnting, Wet or Liquid Palnt Application Methods Drying Powder Coatmg Hot-DIDCoatln~s Chemlcdl Conversion Coatmm - eBkackenin~ or Colorrng Metals Electroplabng Anodizing Blectroless Plating

Electfofess Composite Plating Mechanical Plating Porcelain Enameling 35 5 VAPORIZED METAL COATINGS 35.6 CLAD MATERIALS 35.7 TEXTURED SURFACES 35.8 COIL-COATED SHEETS 35.9 EDGEFINISHING AND BURRS Design to Fadlitate or Elimmatc BurrRemoval 35 10 SURFACE MEGRIT%' Illfluence of Surface Finish on Case Study: DANA LYNN'S FATIGUE: LESSON

35.1

INTRODUCTION
Surface englneerlng is a multidisc~plinary activity intended to tailor the properties of the surfaces of manufactured components so that their function and servkceabllity can be improved. Processes include solld~f~cation treatments such a? hot-drp coatings, weld-overlay coatings,and thermal spray surfaces;deposition surface treatments such as electrodepos~tion, chem~cal vapor depos~tlon, and physical vapor depositlon: and heat treatment coatings such au diffusion coatings and surface hardenmg. Electroplating means the electrodepos~tion of an ddherent metallic coating onto an object that serves as the cathode in an electrochemical reactlon The resulting surface provides wear reststance, corrosion resistance, high-temperature resistance, or electrical properties dlfferenl from those In the bulk mater~al Many manufacturing processes influence surface properties, whrch in turn may s~gniticantly affect the way the component functions in servlce.The demands for greater strength and longer hfe in components often depend on changes in the surface properties rather than the bulk propertles.These changes may be mechanical, thermal, chemical, andiw physical and therefore are difficult to describe ln general terms. For example, two different surface finlshes on Inconel 718 can have a marked effect on the fatigue life, changrng the fat~gue limit from 60 ksi after gentle grind~ng to as low as 22 ksi using electrical discharge mach~nmg (F~gure 35-1). Many metalcutting processes specified by the manufacturing engmeer to produce a specific geometry can often have the effect of producing alterations in the surface mater~al of the component, wh~ch, in turn, produces changes in performance. The term surface zntegrlty was coined by Field and Kahles in 1964 in reference to the nature of the surface condition that is produced by the manufacturing process. If we vlew the process as having five main components (workpiece, tool, machine tool, envlronment, and process variables), we see that surface properties can be altered by all of these parameters (see Table 35-1) by producing the following: High temperatures involved in the machining process Plastic deformation of the work materra! (residual stress) Surface geometry (roughnesr, cracks, distortion) Chemical reactions,particularly between the tool and the workpiece

934

C H A ~ 35 ~ R Surface Eng~neenng
0 Gentle gr~nding

FIGURE 35-1 Fatigue strength of lnconel 71 8 components after surface f~nishing by grind~ng or D M . (Field and Kohles, 1 977).

Cycles to failure

condition of a compwnent or specimen that influences its performance in se Surface integrity has two aspects: topography characteristics and surface-layer acterisgirs. Topography is made up of surface roughness, waviness, errors o f form, an ftaws (Figure 3 5-2). Waviness height

Roughness he~ght

Roughness Profile

Max~mum roughness depth (peak tovalley) along I, RA=Arlthmetic roughness average processe3 produce surfaceflaws, waviness, and roughness that can influence the FIGURE 35-2 Mach~ning performance of the cornportent,
R,
=

Workpiece chntacteristim Geometry Shape Dunemlons Material Type Route of manuldcturc Mechamcat propert~es Elastic mnstants Plastle constants Physlcal propettles Meltmg pornt Thermal d~ffusivlty, condueov~ty, capanty Coefficient of ~ h e m aenpansvon l Phase LrandornraOons Chem~cal propcrt~os Chem~cal composition Chsmt~al aPflnlty to tool material and cnvrronmsnt Metallurg~~al properiies Structure Gram size Hardness

To01 chataeteristies Tool body Type of tool Size Shape Number orcutlma edges Cuttmg edge Shapc jangles) Nose gamttlryitopography Mlcmgeametry Wear Material Type Coating
QPV

Th~ckness
Number and kind of layers Mechanical properties Elas116cunnanh Pkd~tlc properties Physical properties Thermal dlffusinty, mductluity, mpacrty f thermal expaasion Caeffidcnr o Fhem~cill prnpertle3 Chem~oal compowtton Chcmrcal a&fy to tool material Metdllurgtcal pmpentas Structure Oram size
Procesa variable8 Speed Fecd Depth Df cut

Envirvnmenf chsracteristia p p c of medium (gos,nuid, rn~sl) Lubr~clty Coolmg abdity flow rate Temperaturn Chcm~eal camposit~on

Madtine tool chmrteriaiies Error mot~ons

Sourer Advanced Manufacrunna Engcncenng Vol. 1.July 1989

A typlcal roughness profile includes the peaks and valleys that are considered separately from waviness Flaws also add to texture but should be measured independent of it. Changes m the surface layer, as a result of processing,include plastic deformation, residual stresses, cracks, and other metallurgical changes (hardness, overaging, phase changes, recrystalluation,intergranular attack,and hydrogen emhrittlement).SeeFigure30-2.The surface layer will always contaln local surface detosmabon due to any machining passes The material removal process= generate a w~de variety ot surfaces textures, generally referred to as surfacefinish.The cutting processes leave avnde variety of surface patterns the drection of the predommant surface on the matenals. Luy 15 the tern used to des~gnate pattern produced by the machining process In add~t~on, certain other terms and symbols have been developed and standardued for specifying the surface quahty.The most important terms are surfuce roughnes~, wavmess, and lay (Figure 35-3). Roughness refers to the finely spaced surface irregularities. It results from machining Operations in the case of machined surfaces. Wavrness is surface irregularity of greater spacing than in roughness. It may be the result of warpmg,vihrat~on, or the work being deflected dunng machining.

936

CHAPTER 35

Surface Engineering Lay parallel to the surface where symbol is applted

Lay

d z
Lay perpend~cular to the surface where symbol 1s awlled

Roughness he~ght
(a)

Roughness-wldth ~utoff

D
Waviness height
Max, roughness72 Min. roughness 42

{Waviness
\

width

Surface of part
ibl

ic)

Lay symbols Parallel to the boundary line of the nominal surface I Perpendicular to the boundary line of the nominal surface X Angular in both directions tothe f the nominal surface boundary line o M Multidirectional C Approximately circular relative to the center R Approximately radial relative to the center of the nominal surface
=

Lay angular in

both d~rections

l ! ! z
Lay multid~rect~onal

Lay concentric wlth

the center of the

Lay radlal from the

center of the surfac

id1

FIGURE 35-3 (a) Terminology used in speclfylng and measurtng surface quallty, (b) symbols used on drawing

by part des~gners, wdh deflnlt~ons of symbols; (c) lay symbols; (d) lay symbols applied on drawings.

A variety of instruments are available for measuring surface roughness and surface profiles. The majority of these devlces use a dlamond stylus that is moved at a constant rate across the surface, perpendicular to the lay paltern.The rise and fall of the stylus is derected eleckonica~y [often by a Linear Variable Differential Transformer Device (LVTD)], is amplified and recorded on a strip-chart, or 1s processed electronically to produce average or root-mean-square readings for a meter (F~gure 35-4).The unit containing the stylus and the driving motor may be handheld or supported by shds that ride on the workpiece or some other supportmg surface Roughness 1 s measured by the helght of the irregularities with respect to an average line. These measurements are usually expressed in micrometers or microinches.

S-ON

35.1 Introduction

937

Snip chart

recorder pmduces a ptohle

Center line

Profile of surface produced by stylus device

Elanchard ground
X&Y200X Z 5UX

Milled
X & Y 50X Z200X

Ground

lrustspotsl

X & Y 50X

Bead blasted X & Y 50X Z200X

EDM machined Zm x
X &Y240X

FIGURE 354 (a) Schematic of stylus profile dwice far measuring surface roughness and surface profilewith two readout dwiees shown: a meter for AA or rms valuer and a strip chart recorder for f surface profiles. surface profile. (b) Profile enlarged. (c) Examples o

In most cases, the arithmetical average (AA) is used. In term of the measurements,the AA would be as follows:

Curofj'refers to the sampling length used for the calculation of the r o u e height When it is notspecifred, a value of 0.030 in (0.8mm)is assumed.In the previous equation, y, is avertical distance from the centerline andn is the total number of vertical mewwemeats taken within a specifled cutoff dirtance.Tbiis average roughness value IS also called R,occasionaUy used is the raot-memr-sqwre(rrns) value, which is defrned as tms = The resolution of stylus profile devices is determined l%ythe radius or the diame-

ter of the tip of the stylus. When the magnitude of the geometric features begins to

938

CHAITER 35

Surface Engineering

FIGURE 35-6 (a) SEM micrograph of a U.S. dime, showing the S in the word TRUSTafter the region has been traced by a stylus-type machine. (b) Topographical map of the S region of the word TRUSTfrom a 1J.S. dime [compare to part (a)].

SECTION 35.1

Introduction

939

The range of surface roughnesses that are typically produced by various manufacturing processes is indicated in Figure 35-7, which is a very general picture of typical ranges associated with these processes. However, one can usually count on its being more expensive to generate a fine finish (low roughness). To aid designers, metal sarnples with various levels of surface roughness are available. All of the processes used to manufacture components are important if their effects are present in the finished part. It is convenient to divide processes that are used to manufacture parts into three categories: traditional, nontraditional, and finishing treatments. In traditionalprocessesthe tool contacts the workpiece. Examples are grinding,milling,and turning. These material removal processes will inflict damage to the surface if improper parameters are used. Examples of improper parameters are dull tools, excessive infeed,inadequate coolant, and improper grinding wheel hardness. The nontraditional processes have intrinsic characteristics that, even if well controlled, will change the surface. In these processes the workpiece does not touch the tool. Electrochemical machining (ECM),electrical discharge machining (EDM),laser machining, and chemical milling are examples of

(2000) (1000) 1500) (250) (125) 163) (321 (16)

(81

(41

(2)

(11

(0.5)

Elect. discharge rnach

Ream~ng Electron beam Laser Electrochem~cal Bor~ng, turning

Roller burnishing Gr~nd~ng Hon~nq Electropol~sh Polishing Lapping Superf~n~sh~ng

Investment casting Cold rolling, drawing The ranges shown above are typical of the processes listed. Key Higher or lower values may be obtained under special conditions. Extractedfrom General Motors Drafting Standards, June 1973 revision Average application Less frequent applicatio

FIGURE 35-7 Comparison o f surface roughness produced by c o m m o n production processes. (Courtesy ofAmerican Machinist.)

940

CHAP~ER 35 Surface Engineer~ng

nontraditional methods. Such methods canleave stress-free surPdce$ remelted layers, excessive surface roughness. Fznzshzng treatmens can be used to negate or remove the part of both traditional and nontraditional processes as well as providmg good surfa fintsh For example,residual tensile stresses can be removcd by shot peening or roller b mshing. Chemical milhng can remove the recast layer left by EDM. The objectives of the surface-modificat~onprocesses can be q u ~ t c vaned Som are designed to clean surfaces and remove the kinds of defects that occur during pro Others furthe cessing or handling (such as matches, pores, burrs, fins, and blem~shes). improve or modify the products' appearance, providing features such as smoothnes texture, or color. Numerous techniques are ava~lable to Improve resistance to wear corrosion, or to reduce fnct~on or adhes~on to other materials. Scarce or costly maten can be conserved by makmg the interior of a product from a cheaper, more commo material and then coating or plating the product surface. As wrth all other processes,surface treatment requires tlme, labor, equ~pment,an mater~al handling, and all of these have an assocrated cost. Efficiencies can be realm through process optim~zationand the integration of surface treatment Into the enh manufacturing system. Design modifications can often facilitate automated or bul finishing, elim~nat~ng the need for labor-intenswe or single-part operahons. Proces select~on should further conslder the size of the part, the shape of the part, the quanti ty t o be processed, the temperalures requ~redfor processing, the temperature encountered durmg subsequent use, and any d~mensional changes that mlght occu due to the surface treatmenLThrough knowledge of the available processes and th relative advantages and limitations, finishing costs can oftcn be reduced or elimma while maintaining or unproving the qualrty of the product. In addition to the above, the field of surface finishmg has recently undergon another significant change. Many chemicals that were once "standard" to the field,suc as cyantde, cadmium, chromium, and chlormated solvents, have now come undet stn government regulation Wastewater treatment and waste d~sposal have also be significant concerns. As a result, processes may have to be modified or replace processes may have to be used Because of their similarity to other processes, many surface finishing techniq have been presented elsewhere in the book.The case hardening tech~ques, both se tive heating (flame, mduction, and laser hardening) and altered surface chemistry ( fusion methods such as carbunzing, n~tr~ding, and cd~bonitr~ding), are presented Chapter 5 asvanations of heat treatmg. Shutpeening and roller burnzshshm~ are prese ed in Chapter 17 as cold-workingprocesses.Rol1 bundrng andexplos~ve bondmg are dls cussed in Chapter 15 as means of producmg lammar composites. Hurd facing and me sprayingare included in Chapter 33 as adaptatlous of weld~ng techmques. Chemical vap deposition and phyncal vapor deposrtlon are d~scussed In Chapters I9 and 21 Sputte mg and Ion ~mplantatlonare also discussed in Chapter 19 as processes needed in ele tronics manufacturing. In t h ~ s chapter we focus on techn~ques for cleanmg and surfa preparation as well as the remainmg methods of surface fin~shing or surface modrficatio

( 1 35.2

MECHANICAL CLEANING AND FINISHING BLAST CLEANING

It is not uncommon for the various manufactnrlng processes to produce certain types surface contamination. Sand from the molds and cores used in casting often adheres product surfaces. Scale (metal ox~de) can be produced whenever metal 1s processed elevated temperatures. Oxtdes such as rust can t o m if mater~al e stored betwe operation$. These and other contaminants must be removed before decorative protective surfaces can be produced While v~bratory shaking can be useful, some fo of blast cleanrng is usually requ~red to remove the fore~gn material. Blast clean~ng u a me&a (abras~ve)propelled into the surface using alr, water. or even a wheel (wh blasting uses a high-rpm blocked wheel to deliver the medla). The bulk of the wor is dane by kmetic energy of the impacting media, KE = 'I2 MV2 where m = mass of median and v = the velocity. Abrasives, steel grit, metal shot, fine glass shot, plas

SEcnoN 35.2

Mechan~cal Cleaning and

F~n~shtng Blast Cleamng

941

beads, and even CO, are mechanically ~u~pelled against the surface to he cleaned. When sand IS used, 1t should be clean, sharp-edged 31lica sand Steel grit tends to clean more rapidly and generates much less dust, but d is more expensive and less flex~ble When the parts are large,it may be easier to bring the cleaner to the part rather than the part to the cleaner. A common technique for such applications 1s ~ u n d blasting or shot blasting, where the abrasive particles are c a n e d by a hlgh-veloctty blast of alr emergmg Gom a nozzle wlth about a '/,in. operung.Air pressures between 60 and 100 pst. producmg partlcle speeds of 400 mph, are common when cleamng ferrous metals, and 10 to 60 psi 1s common for nonferrous metalsThe abrasive may be sand or shot, or materials such as walnut shells, dry-~ce pellets, or even baking soda. Pressurized water can also be used as a carrier medium. When production quant~ties are large or thc parts are small, the operation can be conducted In an enclosed hood, with the parts travelingpast stationary nozzles.For large parts or small quanntles, the blast may be delivered manually Protective clothlng and breathing apparatus must be provided and precautions taken to control the spread of the resulting dust. The process may even requlre a ded~cated room or booth that is equipped with integrated air pollution control devices. From a mnnufacturmg perspective, these processes are limited to surfaces that can be reached by the movlng abrasive (he-of-s-slght)and cannol be used when sharp edges or corners must be matntained (since the abrasive tends lo round the edges).

BARREL FINISHING OR TUMBLING Barrel fimihzng or bmbl~ng ir an effective means of finishing large numbers of small parts. In the Middle Ages, wooden casks were iilled with abrasive stones andmetal parts and were rolled about until the desired finlsh was obtamcd.Today, rnodificatlons of thrb technique can be used to deburr, radlus, descale, remove rust, pollsh, brighten, surfaceharden, 01 prepare parts for further finishing or assembly.The amount of stock removal can vary from as little as 0.0001 to as much as 0.005 in. In the typical operation, the parts are loaded into a special barrel or drum until a predetermined level is reached. Occasionally,no Other additions are made, and the parts are simply tumbled against one another. In most cases, however, additional rnedla of metal slugs or abrasives (such as sand, granite chips, slag, or ceramicpellets) are added. Rotation of the barrel causes the material to rise until gravlty causes the uppermost layer to cascade downward in a .'landsliden movement, as deplcted m Figure 35-8. The sltding produces abraslve cutting that can effectively remove fins tlash,scale,and adhered sand. Since only a small portion of the load 1s exposed to the abrasive acttan, long times may be required to process the entire contents.
End vtew of barrel

Media and parts

Sliding layel

Fast speed

Media and parts

and

Dlrect~on of barrel rotatlon

Slow speed

C H A ~ E 35 R

Surface Engineering Increasing the speed of rotation adds centrifugal forces that cause the material t rise higher m the barrel.The enhanced action can often accelerate the process, provid that the speed is not so great as to destroy the cascading achon and that the addition4 adion does not damage the workpiece. By a suitable selection of abras~ves, filler, barred size,ratio of workpieces to abrasive,fill level, and speed, a w~de range of parts can be tumi bled successfuIly.Delicate parts may have to be attached to racks within the barrel to; reduce their movement while permitting the media to flow around them. Natural and synthetic abrasives are available in a wide range of slzes and shapeq, including those depicted in Figure 35-9, that enable the fimshing of complex parts wiq Irregular openings.The various media are often mixed in a given load, so that some will? reach into all sections and corners to be cleaned. 1 Tbmbling is usually done dry, but it can also be performed with an aqueous solution in the barrel. Chemical compounds can be added to the media to assist in cleaning) or descalimg, or to provide features such as rust inhibition. Support equipment usual@ assists with loading and unloading the barrels as well as, with the separation of the work-! pleces from the abrastve media. The latter operation often uses mesh screens with; selected size openmgs. Barrel tumbling can be a very inexpensive way to finish large quantit~es of smab parts and produce rounded edges and corners. Unfortunately, the abrasive action od curs on all surfaces and cannot be limited to selected areas.The cycle time is often longi and the process can be qmte noisy. In the barrel burnrshingprocess, no cutting action is desired. Instead, the parts arc? tumbled against themselves or with media such as steel balls, shot, rounded-end pins, oi ballcones. If the orlginal material is free of visible scratches and pits, the combination of peening and rubbing will reduce minute irregularities and produce a smooth, u-, form surface. I Barrel burnishing is normally done wet, using a solution of water and lubrlcatin$ or cleaning agents, such as soap or creamof tartar. Because the rubbing achon between! the work and the media is vesy important, the barrel should not be loaded more thad half full, and the volume ratio of media to work should be about 2:l so the workpiece! rub against the media, not each other.The speed of rotation should be set to maintq the cascading action and not fling the workp~eces free of the tumbling mass. I Catnfugal barrel tumblmgplaces the tumbhng barrel at the end of a rotating a r m . I This adds centrifugal force to the weight of the parts in the barrel and can accelerate the, process by as much as 25 to 50 times

Angle cut cyl~nder Angle-cut trtangle

\B
Sphere
Cone

i
I
i

Diamond

Star

0&
FIGURE 35-9 Synthetic abrasive media are available in a wide variety of sizes and shapes. Through proper selection, the media can be tailored to the product being cleaned.
Ball
Ballcone

o@&@@
Angle-cut prism Angle-cut star Shapes o f medla used for f~n~shjng
Pyram~d
'

dvi
Ovalball

Arrowhead

II'

Cone

D~agonal

Steel medla shapes usedfor burnishing

V~bratory action pattern in single-shaftv~brator

FIGURE 35-10 Schematic diagram of a vibratory-finishing

tub loaded with parts and media. The single eccentric shaft drive provides maximum motion at the bottom, which decreases as one moves upward. The duaishaft design produces more uniform motion of the tub and reduces processing time.

Vibratory actton pattern in dual-shaft vibrator

t'Eccentrics W )
In spindle finishing the workpieces are attached to rotating shafts, and the assembly is immersed in media moving in a direction opposite to part rotation. This process is commonly applied to cylindrical parts and avoids the impingement of workpieces on one another.The abrasive action is accelerated, but time is required for fixturing and removal of the parts.

VIBRATORY FINISHING Vibratory finishing is a versatile process widely used for deburring,radiusing, descaling, burnishing, cleaning, brightening, and fine finishing.In contrast to the barrel processing, vibratory finishing is performed in open containers. As illustrated in Figure 35-10, tubs or bowls are loaded with workpieces and media and are vibrated at frequencies between 900 and 3600 cycles per minute.The specific frequency and amplitude are determined by the size, shape, weight, and material of the pan, as well as the media and compound. Because the entire load is under constant agitation, cycle times are less than with barrel operations.The process is less noisy and is easily controlled and automated. In addition, the open tubs allow for direct observation during the process, which can also deburr or smooth internal recesses or holes. MEDIA The success of any of the mass-finishing processes depends greatly on media selection and the ratio of media to parts, as presented in Table 35-2. The media may prevent the parts from impinging upon one another as it simultaneously cleans and finishes. Fillers, such as scrap punchings, minerals, leather scraps, and sawdust, are often added to provide additional bulk and cushioning. Natural abrasives include slag, cinders, sand, corundum, granite chips, limestone, and hardwood shapes, such as pegs, cylinders, and cubes. Synthetic media typically contain 50 to 70 wt% of abrasives, such as alumina (AI,O,), emery, flint, and silicon carbide.This material is embedded in a matrix of ceramic,polyester, or resin plastic, which is softer than the abrasive and erodes, allowing the exposed abrasive to perform the work.Tne synthetics are generally produced by some form of casting operation, so their sizes and shapes are consistent and reproducible (as opposed to the random sizes and shapes of the natural media). Steel media with no added abrasive are frequently specified for burnishing and light deburring.

Surface Engrneenng

MediaiPart Ratlo by Volume

01 11

Typ~cal Apphcation
Part-onpart process~ng or burr removal without m e d ~ a Produces very rough surfaces and IS 3u1tahle for parts in wh~ch part-on-part damage is nOta problem Somewhat less severe p a n u n part damage, but more actmn from less m & May be acceptable for very small parts and very small medra Part-on-part contact is llkely on larger and heav~er parts In gcneral, a good average ratio for many parts, a good ratlo for evaluatmg a new deburnng process Better for nonferrous parts subject to pan-on-part damage Su~table for nonferrous parts especially preplate surfaces on zinc parts wrth rean-bonded medta For improved preplate surfaces w ~ t h resm-bonded medla Produces very ime fm~shes

21 31 41 51 61

81
10 1

Source Amerrcon Maehcnisr, Augu~t 1983

Media selection should also be correlated with part geometry, since the abrasives should be able to wntact all critical surfaces without becoming lodged in recesses or holes. T h ~requirement s has resulted in a w ~ d variety e of sizes and shapes, including those presented in Figure 35-9. The different abrasives, sizes, and shapes can he selected or combined to perform tasks rangmg from hght deburring with a very fine finish to heavy cuttrng w ~ t h a rough sulface.

COMPOUNDS
acid, neutral, or alkaline. They are often designed to assist in deburring, burnishing, an abrasive cuttmg, as well as to provide cleaning, descalmg, or corrosion inhibition. In deburring and finishmg,many small particles are abraded from both the med and the w~rkpiece$ and these must be suspended in the compound solution to preven them from adhering to the parts. Deburrmg compounds also act to keep the parts an med~a clean and to mhilnt corrosion. Burnishing compounds are often selected for the

tenals contain heavy oil or grease. Corrosion mh~bitors can be selected for both ferro and nonferrous metals and are particularly important when steel media are being use Another funchon of the compounds may be to condition the water when aqueou solutions are being used. Consistent water quality, in terms of "hardness" and metalio content, is important to ensure uniform and repeatable finishing results. Liquid GO pounds may also provide cooling to both the workpieces and the media.

SUMMARY OF MASS-FINISHING METHODS The barrel andvibratory finishingprocesses are really quite simple and economical can process large numbers of parts in a batch procedure. Soft, nonferrous parts can finished in as little as 10 minutes, while the harder steels may require 2 hours or mo Sometimes the operations are sequenced,using progressively finer abrasives Figure 35shows a variety of parts before and after the mass-finish~ng operation, using the tn gular abrasive shown with each component. Despite the high volume and apparent success, these processes may still be much art as science. The key factors of workpiece, equipment, media, and compo

may still be difficult.

RGURE 35-11 A variety of

parts before and after barrel i t h trtangufm-shaped finishing w media. (Courtesy of Norton Company]

BELT SANDING In the belt sandzng operation, the workpieces are held against a moving abrasive belt until the desired degree of finish 1s obtained. Because of the movement of the belt, the resulting surface contains a series of parallel scratches with a texture set by the gnt of the belt. When smooth surfaces are desired, a senes of belts may be employed, with progressively finer grits. The ideal geometry for belt sanding is a flat surface, for the belt can be passed over a flat table where the workpiece can be held firmly against it. Belt sanding is frequently a hand operation and 1s therefore q u ~ t e labor intensive Furthermore, it 1s difficult to apply when the geometry includes recesses or interior corners. As a resdt, belt sanding is usually employed when the number of parts is smaU and thegeometry is relatively slmple. See Chapter 28.
WIRE BRUSHING
High-speed rotary wtrc brushing is sometimes used to clean surfaces and can also impart some small degree of material removal or smooth~ng.The resulting surface consists of a series of umform curved scratches. For many apphcations, this may be an acceptable f not, the scratches can easily be removed by barrel finishing or buffing. frnal finish. I Wire brushing is often performed by hand application of a small workp~ece to the brush or the brush to a larger workpiece. Automatic mach~nes can also be used where the parts are moved past a series of rotating brushes. In another modification the brushes are replaced wlth plastic or fiber wheels that are loaded with abrasive

BUFFING
Bufing is a polishing operation m which the workplece is brought into contact with a revolving cloth wheel that has been charged w t h a fine abrasive, such as pohshing rouge.

The "wheel3", which are made of disks of linen, cotton, broadcloth, or canvas, achieve the desired degree of firmness through the amount of stitching used to fasten the layers of cloth together. When the operation calls for very soft polishing or polishing into interior corners, the stitching may be totally omitted, the centnfugal force of the wheel types of polrotation being sufficient to keep the layers in the proper posltlon. Var~ous lshing compounds are also available, wzth many consisting of ferric oxlde particles in some form of hinder or carnel The buffing operation is very similar to the lapp~ng process that was discussed in Chapter 27. In buffing, however, the abrasive removes only minute amounts of metal from the workpiece.Fme scratch marks can be eliminated and oxide tarnish can be rem0ved.A smooth, reflective surface is produced. When soft metals are buffed, a small amount of metal flow may occur,which further helps to reduce high spots and produce a high polish In manual buff'mg, the workpiece 1s held agamst the rotating wheel and manipulated to provide contact w t h allcritical surfaces.Once agam, the labor costs can be quite extenive. If the workpieces are not too complex,semiautomatic machines can be used, where the workpieces are held ~nfixtures and move past a serles of mdvidual buflimg wheels. By des~gning the part wth buffingmmind,good results can be obtamed quite economically

946

CHAP~ER 35 Surface Engineer~ng

ELECTROPOLISHINC Electropolishtng is the reverse of electroplating (discussed later in this chapter) since tenal IS removed from the surface rather than being dep0sited.A DC electrolyticur is constructed with the workpiece as the anode.As current is apphed,matenal~s strip from the surface, w t h material removal occurring preferentially from any raised 1 tion. Unfortunately,it is not economical to remove more than about 0.001 in of m ial from any surface. However, if the initial surface is suffic~ently smooth (less than rms), and the grain size is small, the result will be a smooth pohsh with irregulanh less than 2 pin.-a mirrorlike finlsh. Electropolishing was originally used to prepare metallurg~calspecimens for ammat~on under the microscope. It was later adopted as a means of polishing sta steel sheets and other stamless products It is particularly useful for polishing irre shapes thal would be difficult to buff.

35.3 CHEMICAL CLEANING

Chemccal cleaning operations are effective means ot removing 011, dirt, scale,or other elgn nlatenal that may adhere to the surface of a product, as a preparation for su quent painting or plat~ng. Because of environmental, health, and safety conc however, many processes that were once the lndustr~al standard have now been e nated or substantiaIly modified. While the major concern with the mechanical met has usually been airborne particles, the chemical methods often require the dispos spent or contaminated solutions, and they occasionally use hazardous, toxic, or ronmentally unfriendly matenals Chlorofluorocarbons (CFCs) and carbon tetrachlo for example, have been identified as o7one-depleting chemicals and have been ph out of commerc~al use. Process changes to comply with added regulations can si cantly shift process economics. Manufacturers must now ask themselves if a part r has to be cleaned, what soils have to be removed, how clean the surfaces have and how much they are willing to pay to accomplish that goal. Selection of thecle method will depend on cost of the equlpment,power, cleanmg materials, mamten and labor,plus the cost of recycling and diaposal of materials. Specificprocesses w pend on the quantity of parts to be processed (part per hour), part configuration, material, desired surface finish, temperature of the process, and flexlbiiity. Manufa ers want machines they can integrate with manufacturing cells so changes m pro can be quickly handled.

ALKALINE CLEANING Alkal~necleaning is basically the "soap and water" approach to parts cleanlug an commonly used method for removrng a wide variety of soils (including oils,greas fine particles of metal, and dirt) from the surfaces of metalsThe cleaners are plex solutions of a l k a h e salts, add~hves to enhance cleaning or surface mod surfactants or soaps that are selected to reduce surface tenslon and displace, em and disperse the insoluble so11s.The actual cleanmg occurs as a result of one or m the following mechanisms: (1)saponlf~cation, the chemical reaction of fats and organlc compounds with the alkaline salts; (2) displacement, where so11particle lifted from the surface; (3) dispersion or emnlsificatlon of insoluble hquids; and ( solut~on of metal oxides. Alkaline cleaners can be appl~ed by immersion or spraying, and they are heated to accelerate the cleaning action. The cleanmg 17 then followed by a wate to remove all residue of the cleaning solution, as well as flush away some small am of remalnmg soi1.A drying operabon may also be required since the aqueous cle do not evaporate quickly, and some form of corrosion inhibitor (or rust preventer be required, depending on subsequent use. Environmental issues relatmg to alkaline cleaning Include (1) reducing or nating phosphate effluent, (2) reducing toxicity and mcreasing biodegradability, a recyclmg the cleaners to extend their life and reduce the volume of discard

SECTION 35.3 Chemical Cleaning

947

SOLVENT CLEANING In solveni cleaning, oils, grease, fats, and other surface contaminants are removed by dissolving them in organic solvents derived from coal or petroleum, usually at room temperature. The common solvents include petroleum distillates (such as kerosene, naphtha, and mineral spirits), chlorinated hydrocarbons (such as methylene chloride and trichloroethylene), and liquids such as acetone, benzene. toluene, and the various alcohoIs. Small parts are generally cleaned by immersion, with or without assisting agitation, or by spraying. Products that are too large to immerse can be cleaned by spraying or wiping. The process is quite simple, and capital equipment costs are rather tow. Drying is usually accomplished by simple evaporation. Solvent cleaning is an attractive means of cleaning large parts, heat-sensitive products, materials that might react with alkaline solutions (such as aluminum, lead, and zinc), and products with organic contaminants (such as soldering flux or marking crayon). Virtually all common industrial metals can be cleaned, and the size and shape of the workpiece are rarely a limitation. Insoluble contaminants, such as metal oxides, sand, scale, and the inorganic fluxes used in welding, brazing, and soldering, cannot be removed by solvents. In addition, resoiling can occur as the solvent becomes contaminated. As a result, solvent cleaning is often used for preliminary cleaning. Many of the common solvents have been restricted because of health, safety, and environmental concerns. Fire and excessive exposure are common hazards.Adequate ventilation is critical. Workers should use respiratory devices to prevent inhalation of vapors and wear protective clothing to minimize direct contact with skin.In addition,solvent wastes are often considered to be hazardous materials and may be subject to high di& cost. VAPOR DECREASING In vapor degreasing,the vapors of a chlorinated or fluorinated solvent are used to remove oil, grease, and wax from metal products. A nonflammable solvent,such as trichloroethylene, is heated to its boiling point, and the parts to be cleaned are suspended in its vapors. Tne vapor condenses on the work and washes the soluble contaminants back into the liquid solvent. Although the bath becomes dirty, the contaminants rarely volatilize at the boiling temperature of the solvent. Therefore, vapor degreasing tends to be more effective than cold solvent cleaning, since the surfaces always come into contact with clean solvent. Since the surfaces become heated by the condensing solvent, they dry almost instantly when they are withdrawn from the vapor. Vapor degreasing is a rapid, flexible process that has almost no visible effect on the surface being cleaned. It can be applied to aU common industrial metals, but the solvents may attack rubber, plastics, and organic dyes that might be present in product assemblies. A major limitation is the inability to remove insoluble soils, forcing the process to be coupled with another technique, such as mechanical or alkaline cleaning. Since hot solvent is present in the system, the process is often accelerated by coupling the vapor cleaning with an immersion or spray using the hot liquid. Unfortunately, environmental issues have forced thz almost complete demise of the process. While the vapor degreasing solvents are chemically stable, have low toxicity, are nonflammable, evaporate quickly, and can be recovered for reuse. the CFC materials have been identified as ozone-depleting compounds and have essentially been banned from use. Solvents that can be used in the same process, or in a replacement process that offers the necessary cleaning qualities, include chlorinated solvents (methylene chloride, perchloroethylene, and trichloroethylene), most manufactarers have converted to some form of water-based process using alkaline,neutral, or acid cleaners or to a process using chlorine-free,hydrocarbon-based solvents. Sealed chamber machines use non-VOC, non~hlorinat~d solvents that are continuously recycled. ULTRASONIC CLEANING When hlgh-quality cleanmg is required for small parts, ultratonrc cleanmg may be preferred. Here the part? are suspended or placed m wire-mesh baskets that are then immersed in a liqud cleaning bath, often a water-based detergent.The bath contains an

.en 35 Surface EIngineering

ultrasonic transducer that operates at a frequency that causes cavitation in The bubbles that form and implode provide the majority of the cleaning action, if gross dirt, grease, and oil are removed prior to the immersion, excellent resuIfs usually be obtained in 60 to 200 seconds. Most systems operate at between 10 40 kHz. Because of the ability to use water-based solutions, ultrasonic cleaning replaced many of the environmentally unfriendly solvent processes.

ACID PICKLING 1 1 1 the acid-picklingprocess,metal parts are Grst cleaned to remove oils and other co taminants and then dipped into dilute acid solutions to remove oxides and dirt that a
left on the surface by the previous processing operations.The most commonsolution a 10% sulfuric acid bath at an elevated temperature between 150" and 185" EMuria acid is also used, either cold or hot. As the temperature increases, the solutions c become more dilute. After the parts are removed from the pickling bath, they should be rinsed to flu the acid residue from the surface and then dipped in an alkaline bath to prev ing. When it will not interfere with further processing, an immersion in a cold m i solution is olten used. Caution should be used to avoid overpickling, since the can result in a roughened surface.

Each of the surface finishing methods previously presented has been a matcrial remo process, designcd to clean,smooth, and otherwise reducc the size of the part. Many 0th techniques have been developed to add matcrial to the sitrlace of a part. If the material is deposited as a liquid or organic gas (or lrom a liquid or gas medium), the processis I T thc added material is a solid during deposition, the process is known called conri~zg. as cladding.

PAINTING, WET OR LIQUID P(~int.s and enunze1.s are by far the most widely used finish on manufactured
a great variety are available to meet the wide range of product requirements. Most of today's commercial paints are synthetic organic compounds that contain pigments and dry by polymerization or by a combination of polymerization and adsorption of oxygen. Water is thc most common carsying vehicle lor thc pigments. Heat can be used to ac. celeratc the drying, but many of the synthetic paints and et~amels will dry in less than an hour without the use oladditional heat.The older oil-based materials have a long drying time and require excessive environmental protection measures. For these reasons, they are seldom used in manufacturing applications. Paints are used lor a variety of reasons, usually to provide protection and decoration but also to fill or conceal surface irregularities change the surface friction, or modify the light UI- heat absorption or radiation characteristics.Table 35-3 provides a list of some ol the more commonly used organic finishes, along with their significant characteristics. Nitrocell~rIo.selacq~lers consist of thermoplastic polymers dissolved in organic solvent.: Although fast drying (by the evaporation olthc solvent) and capable of producing very beautiful finishes, they are not suIficiently durable for most commercial applications. are a general-purpose paint but are not adequate lor hard-service applican ~ alkycls e tions. Acrylic ename1.s are widely used for automotive linishcs and may require catalytic or oven curing. A.phrrlticprrints, solutions of asphalt in a solvent, arc used extensively in the electrical industry. where rcsistance to corrosion is required and appearance is, not o l prime importance. When considering a painted finish, the temptation is lo focus on the outermost coat, to the exclusion of the underlayers. In reality, painting is a complex systcm that includes the substrate material, cleaning and other pretreatments (such as anodizing, phosphating, and various conversion coatings), priming, and possible intermediate layers. The method of application is another integral feature to be considered.

Mater~al
N~%~eAlulose lacquers Epoxy esters Akyd-mine Acryl~c lacquers Aoryllcenarnels

Durab~lity (Scale of 1-10)

1
1

Relative Cosl (Scale of 1-10)

2

Charactensi~cs Fast drvrng, low durdhlhly

Coodchem~cnl reslstancc Versallle,low adheslon Good olar retention, L o w adhaston Good color retention, though, hlgh
baking tempcratm Flenble, good chern~cal reststanCe: low sollds Good gloss retsnllrm. lo*

2 4 4

fkx~b~llty

Excellent durab>l~q, difficult fo apply

PAINT APPLICATION METHODS In manufacturing,almost all painting is done by one of four methods: dipping, h n d spraymg, auaomntic sprayrag, or electrm*rtic~rayfin&hing. In most casegat leatt two coats are required. llre first (or prime) coat serves to (1)ensure adhesion, (2) provide a leveling effect by Eill~ng in minor poroslty and other surface blemlshes,and (3) improve corrosion
resistance and thus prevent later coatings from being dislodged in service These properties are less easily attainable in the more highly pigrnent~d paints that are used in the Einal coats to promote color and appearance. When using multiple coats, however, it is important that the carrylng vehicles for the final coats do nut unduly soften thc underlayers Dipping is a simple and economical means of paint applicat~on when all surfaces of the part are to b e coated.The products can be manually immersed into a paint bath or passed through the bath whde on or attached to a conveyor. Dipping is attractive for applying prlme coats m d for painting small parts where spray painting would result m asignificant waste due to overspray Conversely, the procew is unartractive where only some of the surfaces requlre painting or where a very thin,unifornl coating would be adequate, as on automobile bodies. Other d~fficulties are associated with the tendency of paint to mm,producing both a wavy surfaceand afinal drop of pamt attached to the lowest drip point. Good-quality dipp~ng requires that the paint be st~rred a t all times and be of uniform viscosity. Spray painting isprobably the most widely usedpaint application process because of its versatility and the economy in the use of pamt. In the convennonal technique, the paint is atomized and transported by the flow of compressed air. In a variation known as i?irlt?~.q spruyLr.g, mrch;~nicnl prcssurc i o r c thc ~ ~ paint through ;Irl orifice a t prcisurc, I ~ p.ci.'lhis prnvid2s sufficient \clo:ity t o produc: ;itomi7ot1<11l ;151d bctwczn .iOll ~ I I450i) also propel the particles to the workpieee. Because no air pr&sure is used for atomlzation, thereisless spray loss (paint efficiency may be as high as 99%) and Less generatlon of gaseous fumes. Hand sp~aying is probably the m o s t versatil~ means of appiicabon but can be quite costly in terms of labor and production time. When air or n~echanical means provide the atomizatxon,workers must exercise cons~derable skill to obtain the proper coverage without allowing the paxnt to "run"or '%rape." Only a very thin film can be deposited at one time,mually less than 0.001 in. As a result, several coats may he required wlth intewenina - time for drvine. . ., Onc means of npplylng thicker 1;lvcrs in a hinglc applic:,lion 15known 2 s / r , ~ l,pra)i l l s . Speci.11sol\cnt3 arc used th.at icduc,. the \,i>c.~~\ily (if thc rn~terial whcn hc;~ted. llpc~n atomization, the faster-evaporatiq solvents are removed, and the drop in temperature produces a more viscous, run-resistant material that can he deposited m thicker layers. of similar or idcntic.ll parts, s<,mciorm <)I ;iuto\Vhcn prodticine large clu:~ntilic> matic system is usually cmployed.'lhe siiuplcsr automatic ~ . q u i p m cr.~ns$ts ~~t c ~ sunle i form of parts conveyor that transports the parts past a series of stationary spray heads.

CWER 35

Surface Engineering

and repetitious pracess can be performed with consistent results. In addition, use of a robot removes the human from an unpleasant, and possibly unhealthy, environment: Nowdays, cars are painted almost exclusively with robots.

Direct-current Water & detergent power supply o Emulsion of

FIGURE 35-12 Basic steps in the electrocoating process.

Electrocoating: Time cycle 1 minute

Rinsing

Baking: Time cycle 10 minutes - 380F

Since the solvent is water, no Iire hazard exlsts (as with the use of many solvents), and alr and water pollution is reduced significantly. In addlt~on, the process can be read~ly adapted to conveyor line production.

DRYING
Most palnts and enamels used in manufacturing requue from 2 to 24 hours to dry at normal room temperature. This tlme can be reduced to between 10 minutes and 1 hour if the temperature can be rased to between 275" and 450"E As a result, elevated-temperature drying is often preferred. Parts can be batch processed in ovens or contlnuously passed through heated tunnels or under panels of infrared heat lamps Elevated-temperature drylng is rarely a problem \nth metal parts, but other mater~als can be damaged by exposure to the moderate temperatures.For example, when wood is heated, the gases, moisture, and residual sap are expanded and driven to the surface beneath the hardening paint. Small bubbles tend to form that roughen the surface, or break, producing small holes in the paint.

POWDER COATING Powder coating is yet another vanation of electrostatx spraying, but here the particles are solid rather than l~quid. Several coats, such as primer and finish, can be apphed and then followed by a single baking, in contrast to the baking after each coat that is required in the conventional spray processes. In addit~on, thc overspray powder can often be collected and reused. While volatilized solvents are no longer a concern, operators must now address the possiblltty of powder explosion, as well as the health hazards of airborne particles. Modem powder technology can produce a high-quahty finish with supenor surface properties and usually at a lower cost than l~quid painting. Powder painting 1s more efflc~ent in the use of materials (the overspray can be captured and reused) and lower energy requirements. The economic advantages must be weighed against the limitations of powder coating. Dry systems have a longer color change time than wet systems.The process is not good for large objects (massive tanks) or heat-sensitive objects. It is not easy to produce film thickness less than 1mil (0.03mm) Table 35-4provldes details on powders that are used in powder coatings Thermoplastm can also be used, but thermosetting powders aremost common.The elements

., '

w:.
A

,;,' , . .

. .. .. .. $)'& . & .. By .$ eb . yarii.of ,. Properties . and ~ ~ ~ l i c a t i o n s

.

..

Epoxyhlyester Propertieq Application thick~tess Cur5 cyek (metal temperaturr~)~ Outdorrr wealherab~l~ty Pcnctl hardness Dlrect ~mpact reqrstance, tn ihi Chemical resi~tancl Cost (relative) Applicdtions
EPOXY

Polyester

Acrylic
Urethane
0.5-10 mils
400F-7 mm, 360F-25 rmn Excellent

Hybnd
0.5-10 mils 450F-3 mm. 325F-25 mln Poor

TGIC Polyester
0.5-10 n~ils 400-F-7 min, 310F-20 mtn Very good

Urethane
0.5-10 mils 4 0 0 F 7 mm, 350F-17 mm
Very good

0.5-20 milss 45OFF-3 mm 250F-30 mm Poor

HB 5 H
80-1 60

HBdH 80-160
Very goad Least expensive 1 Water heaters, radiators, office furniture

HB-2H 8W60
Good

HB 3H $&I60
Good

HJII 20-60
Very goad Most expensive

Fxcenent

Furniture, cars, ovens, appliances

3 Architectural aluminum, outdoor furniture, farm equipment

5
Car wheelslrims, playground equipment Washing machines, refrigerators, ovens

up to 150 m i l s can be applied via multiple costs in a fluidized bed, 6Thickni.r~ Time and t~mperaturccan be reduced.hy utilizing acceierated curing rnechimismr, whilemaintaining the same general propertier Tested at a coaling thickness of211 m i l s .

952

CHAITER 35

Surface Engineering Powder application equipment

Dry cartridge collector

Powder spray gun

Source Nordson Co of a powder coating system. The wheels on the color modules permit it to be FIGURE 35-13 A schemat~c exchanged w~th a spare module to obtain the next color of a powder coating system are shown in Figure 35-13. The following aspects of process must be cons~dered: Types of guns--corona charged or tribo charged Number of guns--depends on many factors, such as parts per hour, slze of parts, speed, and powder types Color change timeifrequency Safety Curing oven-coated parts put m ovens to melt, flow, and cure the powder

HOT-DIP COATINGS

ping into a solutlon of zinc ammomum chlonde and dned. Next, the art~cle is wrn immersed in a bath of molten zinc.The zinc and iron react metallwgcally to produce mg that consists of a series of zinc-iron compounds and a surface layer of nearly pu The coating thickness is usually specified in terms of weight per unit area between 0 5 and 3.0 oz/ft2are typ~cal, w t h the specific value depending on the

SECTION 35.4 Coatings

953

coatings are properly applied, bending or forming can often follow galvanizing without damage to the integrity of the coating. Zinc-galvanized sheet can be heat treated with a zinc-iron alloy coating. The 10% iron content adds strength and makes for good corrosion and pittingichipping resistance. In auto applications, galvannealing beats out pure zinc on several counts: spot weldability, pretreatability, and ease of painting. Electrogalvanized zinc-nickel coatings that contain 10 to 15% Ni can be used in thinner layers (56 microns) and are easier to form and spot weld. The primary limitations to hot-dip galvanizing are the size of the product (which is limited to the size of the tank holding the molten zinc) and the"damage" that might occur when a metal is exposed to the temperatures of the molten material (approximately). Tin coarings can also be applied by immersing in a bath of molten tin with a covering of flux material. Because of the high cost of tin and the relatively thick coatings applied by hot dipping, most tin coatings are now applied by electroplating. Terne coatcng utilizes an alloy of 15 to 20% tin and the remainder lead.This material is cheaper than tin and can provide satisfactory corrosion resistance for many applications.

CHEMICAL CONVERSION COATINGS In chemrcalconverszon coating, the surface of the metal is chemically treated to produce a nonmetallic, nonconductive surface that can impart a range of desirable properiies.The most popular types of conversion coatings are chsomate and phosphate. Aluminum, magnesium, zinc, and copper (as well as cadmium and silver) can all be treated by a chromate conversion process that usually involves immersion in a chemical bath. The surface of the metal is convened into a layer of complex chromium compounds that can impart colors ranging from bright clear through blue, yellow, brown, oLive drab, and black. Most of the films are soft and gelatinous when they are formed but harden upon drying. They can be used to (1)impart exceptionally good corrosion resistance; (2) act as an intermediate bonding layer for paint, lacquer, or other organic finishes; or (3) provide specific colors by adding dyes to the coating when it is in its soft condition. Phosphate coatings are formed by immersing metals (usually steel or zinc) in baths where metal phosphates (iron, zinc, and manganese phosphates are all common) have been dissolved in solutions of phosphoric acid. The resultant coatings can be used to precondition surfaces to receive and retain paint or enhance the subsequent bonding with rubber or plastic. In addition, phosphate coatings are usually rough and can provide an excellent surface for holding oils and lubricants. This feature can be used in manufacturing, where the coating holds the lubricants that assist in forming, or in the finished product, as with black-color bolts and fasteners, whose corrosion resistance is provided by a phosphate layer impregnated with wax or oil. BLACKENING OR COLORING METALS Many steel parts are treated to produce a black, iron oxide coating-a lustrous surface that is resistant to rusting when handled. Since this type of oxide forms at elevated temperatures, the parts are usually heated in some form of special environment, such as spent carburizing compound or special blackening salts. Chemical solutions can also be used to blacken, blue, and even "brown" steels. Brown, black, and blue colors can also be imparted to tin, zinc, cadmium, and aluminum through chemical bath immersions or wipes. The surfaces of copper and brass can be made to be black, blue, green, or brown, with a full range of tints in between. ELECTROPLATING Large quantities of met.al and plastic parts are electroplated to produce a metal coating that imparts corrosion or wear resistance, improves appearance (through color or luster), or increases the overall dimensions. Virtually all commercial metals can be plated, including aluminum, copper, brass, steel, and zinc-based die castings. Plastics can be electroplated, provided that they are first coated with an electrically conductive material. The most common platings are zinc, chromium, nickel, copper, tin, gold, platinum, and silver.The electrogalvanized zinc platings are thinner than the hot-dip coatings and can be produced without subjecting the base metal to the elevated temperatures of

S ~ c n 35.4 o ~ Coatings

953

coatings are properly applied, bending or forming can often follow galvanizing without damage to the integrity of the coating. Zinc-galvanized sheet can be heat treated with a zinc-iron alloy coating.The 10% iron content adds strength and makes for good corrosion and pittingichipping resistance.In auto applications,galvannealing beats out pure zinc on several counts: spot weldability, pretreatability, and ease of painting. Electrogalvanized zinc-nickel coatings that contain 10 to 15% Ni can be used in thinner layers (5-6 microns) and are easier to form and spot weld. The primary limitations to hot-dip galvanizing are the size of the product (which is limiled to the size of the tank holding the molten zinc) and the "damage" that might occur when a metal is exposed to the temperatures of the molten material (approximately). Tin coatings can also be applied by immersing in a bath of molten tin with a covering of flux material. Because of the high cost of tin and the relatively thick coatings applied by hot dipping, most tin coatings are now applied by electroplating. Terne coating utilizes an alloy of 15 to 20% tin and the remainder 1ead.This material is cheaper than tin and can provide satisfactory corrosion resistance for many applications.

CHEMICAL CONVERSION COATINGS In chemical conversion coating, the surface of the metal is chemically treated to produce a nonmetallic,nonconductive surface that can impart a range of desirable properlies.The most popular types of conversion coatings arc chromate and phosphate. Aluminum, magnesium,zinc,and copper (as well as cadmium and silver) can all be treated by a chromate conversion process that usually involves immersion in a chemical bath. The surface of the metal is convened into a layer of complex chromium compounds that can impart colors ranging from bright clear through blue, yellow, brown, olive drab, and black. Most of the films are soft and gelatinous when they are formed but harden upon drying. They can be used to (1) impart exceptionally good corrosion resistance; (2) act as an intermediate bonding layer for paint, lacquer, or other organic finishes; or (3) provide specific colors by adding dyes to the coating when it is in its soft condition. Phosphate coatings are formed by immersing metals (usually steel or zinc) in baths where metal phosphates (iron, zinc, and manganese phosphates are all common) have been dissolved in solutions of phosphoric acid.The resultant coatings can be used to precondition surfaces to receive and retain paint or enhance the subsequent bonding with rubber or plastic. In addition,phosphate coatings are usually rough and can provide an excellent surface for holding oils and lubricants. This feature can be used in manufacturing, where the coating holds the lubricants that assist in forming, or in the finished product, as with black-color bolts and fasteners,whose corrosion resistance is provided by a phosphate layer impregnated with wax or oil. BLACKENING OR COLORING METALS Many steel parts are treated to produce a black, iron oxide coating-a lustrous surface that 1s resistant to rustlng when handled. Slnce this type of oxide forms at elevated temperatures, the parts are usually heated in some form of special environment, such as spent carburizing compound or special blackening salts. Chemical solut~on% can also be used to blacken, blue, and even "brown" steels. Brown, black, and blue colors can also be imparted to tm, zinc, cadmlum, and aluminum through chemical bath immersions or wipes. The surfaces of copper and brass can be made to be black, blue, green, or brown, with a full range of tints in between ELECTROPLATING Large quantities of metal and plastic parts are electroplated to produce a metal coating that imparts corrosion or wear resistance, improves appearance (through color or lusLer), or increases the overall dimensions.Virtually all commercial metals can be plated, including aluminum, copper, brass, steel, and zinc-based die castings. Plastics can be electroplated, provided that they are first coated with an electrically conductive material. The most common platings are zinc, chromium, nickel, copper, tin, gold, platinum, and silver.The electrogalvanized zinc platings are thinner than the hot-dip coatings and can be produced without subjecting the base metal to the elevated temperatures of

FIGURE 35-14 Basic circult for an electroplatinq operation, cnoiv'ng tile anooe, cathove (workp ecc), and elecrrolytc (conductive solution).

molten zinc. Nickel plating provides good corrosion resistance but is rather e and does not retain its lustrous appearance. Consequently,when lustrou desired,a chromium plate 1s usually specified Chromium is seldom used An initial layer of copperproduces a leveling effect and makes it possible to reduce thickness of the nickel layer that typically follows to less than 0.0006 in. of chromium then provides the attractiveappearance. Gold, silver, and pl are used in both the jewelty and electronrcs industries, where the t h i n layers impart desired properties while conserving the precious metals. Hard chromiumplate,mth Rookwell hardnesses between 66 and 70, can he used build up worn parts to larger dimensions and to coat tools and othm products that ne reduced surface eichon and good resistance to both wear and mosion. Hard eoatings are always applied directly to the base material and are usually much than the decorative treatments, typically ranging from ,003to ,010 in. thick Event layers are used m applications such as diesel cylinder hners. Sincehardchrome not have a leveling effect, defects or roughness in the base surface will be If smooth surfacesare desired, subsequent grinding and pohshiing may be necessa Figure 35-14 depicts the typical electroplating process. A DC voltage is between the parts to be plated (which is made the cathode) and an anode mate is e~ther the metal to be plated or an Inert electrode. Both of these components are mersed in a conductive electrolyte,which may also contain dissolved salts ofthe m to be plated as well as additions to increase or control conductivity. In response to applied voltage, metal ions migrate to the cathode, lose their charge, and deposit o surface.While theprocess issimple in its basic concept, the production of a high-q ty plating requires selection and control of a number of variables, including the trolyte and the conceutrations of the various dissolved components, the tempera the bath,and the electrical voltage and corrent.The interrelation of these features to the complexity and makes process conhol an extremely challengingproblem. The surfaces to be plated must also be prepared properly if satlsfact are to he obtained. Pinholes, scratches,and other surface defects must be r smooth, lustrous finish is desired. Combinations of degreasing,cleaning, and used to ensure a chemically clean surface, one to which the plating mater~al c a na As shown in Ftgure 35-15, the plated metal tends to be preferentially a to apply a nnifo comers and protrusions. This makes it particularly d~fficult to ~rregulat shapes, especially ones containing recesses, corners, and edges. Dm' tures can be incorporated to promote plating nndormity, and improved results be obtained througi~ the use of multiple spawd anodes or anodes whose shap bles that of the workpiece. Electraplating is frequently performed as a continuous process, where th ndual parts to be plated are hung from conveyors.As they pass through the proce

Convex surfaces. Plate un~formly if edges are rounded espec~ally

mmrrmssm

m = = = v

Flat surfaces Not des~rableUse sllght crown t o hide undulat~ons

Concave recesses Platabil~ty depends on d~menslons

rrnmdhdm

Slots Narrow slots and holes should have rounded corners

~-

Blind holes: Must be exempted from minimum thickness requirements. Require vent hole at blind end.

V-shaped grooves: Difficult to plate. Should be avoided

Sharply angled edges: Plating is thinner in center areas. Round all areas.

Fins: Increase plating time and costs Reduce durability of finish.

are lowered into successive plating, washing, and fixing tanks. Ordinarily, only one type of workpiece is plated at a time, because the details of solutions, immersion times, and current densities are usually changed with changes in workpiece size and shape. In the electroformingprocess, the coating becomes the final product. Metal is electroplated onto a mandrel (or mold) lo a desired thickness and is then stripped free to produce small quantities of molds or other intricate-shaped sheet-metal type products.

ANODIZING Anodizing is an electrochemical process, that is somewhat the reverse of electroplating, which produces a conversion-type coating on aluminum that can improve corrosion and wear resistance and impart a variety of decorative effects. If the workpiece is made the anode of an electrolytic cell, instead of a plating layer being deposited on the surface, a reaction progresses inward, increasing the thickness of the hard hexagonal aluminum oxide crystals on the surface.The hardness depends on thickness, density, and porosity of the coating, which are controlled by the cycle time and applied currents along with the chemistry, concentration, and temperature of the electrolyte. The sul'face texture very nearly duplicates the prefinishing texture, so a buffing prefinish produces a smooth, lustrous coating while sand blasting produces a grainy or satiny coating. The flow diagram in Figure 35-16 shows the anodizing process. Coating thicknesses range from 0.1mils to 0.25 mils. Note that the product dimensions will increase, however, because the aluminum oxide coating occupies about twice the volume of the metal from which it formed. The nature of the developed coating is controlled by the electrolyte. If the oxide coating is not soluble in the anodizing solution, it will grow until the resistance of the oxide prevents current from flowing. The resultant coating, which is thin, nonporous, and nonconducting,is used in a variety of electrical applications. If the oxide coating is slightly soluble in the anodizing solution, dissolution competes with oxide growth and a porous coating will be produced, where the pores provide for continued current flow to the metal surface.As the coating thickens, the growth rate decreases until it achieves steady state, where the growth rate is equal to the rate of dissolution.This condition is determined by the specific conditions of the process, including voltage, current density, electrolyte concentration, and electrolyte temperature. Sulfuric, chromic, oxalic, and phosphoric acids all produce electrolytes that dissolve oxide, with a sulfuric acid solution being the most common.

956

c~Am-rR 35

Surface

Engineering

FIGURE 35-16 The anodizing process

has many steps.

In a process variation known as color modrz produce a layer of microscopically porous oxide t h and somewhat opaque on alloys.When this materi lllary actlon pulls fhe dye into the pores. The dye operation, usually performed simply by ~mmersing the ano water.The aluminum oxide, coating is converted to a monohydrate,with acco increase in volume The pores close and become resistant to further staining or ing out oi the dye. While most people are familiar with the variety 4f colors in aluminum athleti goods, such as softball hat$ the actual applicafibns range from giftware,through auto motive tnm, to architectural use. Aluminum can be made to look like gold, copper,0 brass, or it can take on a variety of colors with a combined metallic luster that canho be dupl~cated by other methods. If PTFE (Teflon) is introduced into the pores, ooati ple high hardness and low friction. The porous oxide layar can also be used toe the adhes~on of an adhtlonal layer of material, such as p m t , or carry lubncan a subsequent forming operation. Since the coating is integral to the part, subsequ operations can often be performed without destroying its integrity or reducing its p tective qualttles. Anodzing can also be performed on other metals, such as magnesium, and process is similar to the passivation of stainless steel.

ELECTROLESS PLATING When using electroplating,it18 almost impossible to obtain a uniform plating thick on even moderately complex shapes, the platings cannot be applied to nonconduc and a large amount of energyis required, For these reasons, a substantial effort has b d~rected toward the development of plating techniques that do not require an source of electricity.These methods are known as elecnolas, or ourocatalytic, donsiderable success has been achieved with nickel, some of the precious metals, can also be deposited. In the electroSess ptocess, complex plating ducing agents,complexiag agents,pH adjusters, and st with a substrate ~urface that acts as a catalyst or has been pretreated with catalytic

terial.The metallic ion in the plating solution is reduced to metal and depos~ts on the surn purely a chemlcal process, the coatings are un~form 1n thtckface. S~nce the depos~tion ness. Independent of part geometry Unfortunately, the rate of depos~tion is considerably slower than w~th electroplattng. Probably the most popular of the electroless coatings 1s electroless nickel, and varlous methods exist for its depos~tion using both acid and alkaline solutions.The coatings offer good corrosion resistance, as well as hardnesces between Rockwell C 49 and 55 In addition, the hardness can be increased further to as high as Rockwell C 80 by subsequent heat treatment.

ELECTROLESS COMPOSITE PLATING

A very useful adaptation of the electroless process has been developed whereln mtnufe

partrcles are w-deposited along w~th the electroless metal to produce compos~te-matenal coatings. Finely div~ded solid particles, with diameter5 hetween 1 and 10 in., are added to the platrng bath and depos~t up to 50 vol% wtth thc matnx. Whlle tt may appear that a large variety of materials could be co-deposited, commerc~al applicat~ons have laigely been hm~ted to d~amond, silicon carbide, alummum ox~de, and Teflon (PTFE). Figure 35-17 shows a depos~t of s ~ l ~ c o carbide n particles in a nickel-alloy matnx, where the partrcles conahtute about 25% by volume. The coating offers the same cor rosion resistance as nickel, hut the high hardness of the silicon carb~de particles (about 4500 on the Vickers scale, where tungsten carbide is 1300 and hardened steel is about 900) contributes outstanding resistance to wear and abrasion. Slnce the deposit~onis electroless, the thickness of the coatlng is not affected by the shape of the part. Applications include the coating of plastic-molding dies, lor use where ihe polymer resln contarns s~gnificant amounts of abrasive filler.

MECHANICAL PLATING Me~hunlcalplu~lng, also known as peen plattng or Impact platlng. 1s an adaptation of barrel fin~shrng in wh~ch coatings are produced by cold-weldmg sott, malleable metal powder onto the substrate. Numerous small products are first cleaned and may be

Base metal

FIGURE 35-17 (Left) Photom:croqraph of nirke, carh:de plat'ng pr0d~cl.u by elcctrolru drposit.on. hor're

the un'form rhickness coar:ng on the irregularly shaped prodhct. (Riqht) Hiqli-rnaqriif'c;~tiori cross 5rct'ori through the roatny. (Cuune5y u i Electru-Coatmgs Inc.)

958

CHAPTER 35

Surface Engineering

given a thin galvanic coating of either copper or tin.They are then placed in a tumbling barrel, along with a water slurry of the metal powder to be plated, glass or ceramic tumbling media, and chemical promoters or accelerators. The media particles peen the metal powder onto the surface, producing uniform-thickness deposits (possibly a b ~thinner t on edges and th~cker in recesses-the opposite of electroplating!). Any metal that can be made into fine powder can be deposited, but the best results are obtained for soft materials, such as cadmium, tm, and zinc. Since the material deposited mechanically, the coatings can be layered or involve mixtures with bu chemistries that would be c h e m a l l y impossible due to solubility limits. The fact tha the coatings are deposited at room temperature, and in an environment that doe not induce hydrogen embrittlement, makes mechanical plating an attractive mean of coating hardened steels.

PORCELAIN ENAMELING
Metals can also be coated with a variety of glassy, inorganic materials that impart resi tance to corrosion and abrasion, decoratwe color, electrical insulation, or the abihty t function in high-temperature environments. Multiple coats may be used, with the fi or ground coat being selected to provide adhesion to the substrate and the cover co to provide the surface characteristics.The material is usually applied in the form of am ticomponent suspension or slurry (by dipping or spraying), which is then dried and fire An alternative dry process uses electrostatic spraying of powder and subsequent firm During the firing operation, which may require temperatures in the range of 800" 80UUF, the coatlng materials melt,flow, and resolidify. Porcelain enamel is often foun on the inner,petiorated tubs of many washing machines and may be used to Impart th decorative exterior on cookpots and frying pans

111 35.5 VAPORIZED METAL COATINGS

Vapor deposition processes can be classified into two main categories:physrcal vapo depositcon (PVD) and chemical vapor deposition (CVD). While sometimes used as thou it were a specific process, the term PVD applies to a group of processes in which material to be deposited is carried physically to the surface of the workpiece.Vacu metallizing and sputtering are key PVD processes, as are complex variations, such as io plating. All are carried out in some form of vacuum, and most are line-of-sight in which the target surfaces must be positioned relative to the source. In contrast, processes deposit material through chemical reactions and generally require significant ly hgher temperaturesToo1steels treated by CVD may have to be heat treated again, most PVD processes can be conducted below normal tempering temperatures. Chapter 7.l for additional discussions on PVD and CVD processes.

11 35.6 CLAD MATERIALS

Clad materials are actually a form of composite in which the components are jomed as solids, using techniques such as roll bonding, explosive welding, and extrusion. The most common form is a laminate, where the surface layer provides properties such as corral slon resistance. wear resistance, electrical conductivity,thermal conduct~wty, or improved appearance, while the sub~trate layer provides strength or reduces overall cost Alclad aluminum is a typical example. Here surface layers of weaker but more corrosion^ resistant single-phase aluminum alloys are applied to a base of high-strength but less corrosion-resistant, age-hardenable material. Aluminum-clad steel meets the same objective but with a heavier substrate, and stainless steel can be used to clad steel% reducing the need for nickel- and chromium-alloy additions throughout. Wires and rods can also be made as claddlngs Here the surface layer often imparts: conductivity, while the core provides strength or rigidity. Copper-clad steel rods that caa be driven into the ground to provide electrical grounding for lightning rod systems ar$ one example.

SECrIoN 35.9

Edge Fln~sh~ng and Burrs

959

P J 35.7 TEXTURED SURFACES

While technically not the result of a surface fimshlng process or operatxon,texturedsarfaces can be used to impart a number of desirable properr~es or characterist~cs,The types .. of textures that are often rolled onto the sheets used for refrigerator panels serve to conceal dirt, smudge&and fingerprints.Embossed or coined protrusions can enhance the grip of metal s t m treads and walkways. Corrugat~ons provide enhanced strength and ng~dity. St111other textures can be used to mod~fy the optical or acoustical characterist ~ c of s a material.

35.8 COIL-COATED SHEETS

Tradttlonally,sheet metal components, such as panels for apphance cabinets, have been fabricated from bare-metal sheels Pans are blanked and shaped by the traditional metalfonning operations, and the shaped panels are then finished on an individual basis. This requires individual handling and the painting or plating of geometr~es that contam holes, bends, and contours. In add~tion, there is the time requlred to harden, dry, or cure the applied surface finish An alternative approach is to apply the finish to the sheet material after rolling but before coiling. Coahags can be applxed contmuously to one or both sides of the materseffectively prefinished, ~ awhile l it 1s in the form of a flat sheet.Thus the coiled material 1 and efforts need to be taken to protect the surface during the blank~ng and forming operations used to produce the final shape. Various paints have been applied successfully, as well as a full spectrum of metal coatings and platmngs.The sheared edges wxll not be coated, but d this feature can be tolerated, the additional measures to protect the surface may be an attractive alternative to the finish~ng of ~ndividual components, A second sequence that has some advantages takes the coils of steel that have been cut to length and stamps the holes and notches ~ n t o them to create blanks The blanks are pretreated, dned, powder coated, cured, and restacked Then they are postformed to shape them mto the back, side, and front panels of appliance cabtnets The manufacturers call this biarak coatcng.The mating thickness is about 1.5 mils t 0 2 mil versus 2 mils -C 0.5 mil (less powder, better quality), and rusting at the corners of the holes is eliminated.

11 35.9 EDGEFINISHING AND BURRS

Burrs are the small, bometlmey flexible projections of mater~al that adhere to the edges of workp~eces that are formed by cutting, pnnchlng, or grinding, like the exlt-side burrs formed in themilled slot of Figure 35-18. Dimensionally, they are typically only 0.003 in thick and 0 001 to 0.005 m. in height, but IE not removed, they can lead to assembly failures, short circuits, Injuries to workers, or even fatlgue failures. Heavy burrs on exlt s~de

FIGURE 35-18 Schematic

show~ng the format~on of heavy burrs on the exit s~de of a milled slot. (From 1. X. Cillesp~e, Amer~can Mach~naf, November
remove burrs form on these edges

find burrs as well.

sonic machining, a i d

abrasive flow m a c h i n i i (see Chapter 26) Other burr removal methods mav be auite soecialized.such as thermal-enersv de-

burrs are vaporized in less than 20 ms, including those in inaccessible or d~fficult-I

Process

Edge Radius, mm (in )

Stock Loss,

Surface Em~sh,

SLLT~ON 35.10 Surface lntegr~ty

961

reach locatrons. Since the process does not use abrasive media, there is no change to any of the product dimeusions.The product surfaces are rarely affected by the generated heat, and the cycle (mncludmg loading and unloading) can be repeated as many as 100 times an hour. Unfortunately, there is a thin recast layer and heat-affected zone that forms where the burrs were removed.Thts regon is usually less than 0.001. in. thick but may be objectionable m hardened steels and highly stressed parts. Of all of the burr removal methods, tumbling and vibratory finishing are usually the most econom~cal, typically costing in the neighborhood of a few cents per part. Since m o a of the common methods also remove metal from exposed surfaces and produce a radius on all edges, it is important that the parts be designed for debnrnng Table 35-5 prof the various deburring processes, as well as the edge rad~us, stock loss, and vides a listing o surface fin~sh that would result from removal of a "typical burr" of 0.003 in. thickness.

DESIGN TO FACILITATE OR ELIMINATE BURR REMOVAL

By knowing how and where burrs are likely to form,the engineer may be able to demgn parts to malce the burrs easy to remove or even elimmate them. As shown in Flgure 35-19, extra recesses or grooves can e l i i n a t e the need for deburring,since the burr produced by a cutoff tool or slot milling cutter will now lie below the surface.In this approach,one must deterrnme whether it is cheaper to perform another machining operation (undereunmg or groomg) or to remove the resulting burr. Chamfers on sharp corners can also eliminate the need to deburr.The chamfering tool removes the large burrs formed by facing, turning, or bormg and produces a rel~ef for mating parts.The smallburr formed dunng chamferingmay be allowable or can easily be removed. Often, it may be preferable to glve the manufacturer the freedom to use either a chamfer (produced by machuling) or an edge radius (formed durlng the deburring operation) on all exposed corners or edges.

Surface integrity has become thesubject of intenseinterest because the traditiona1,nontraditional, and posttreatment methods used to manufacture hardware can change the material's properties.Although the consequence of these changes becomes adesign problem, the preservation of properties is a manufacturing consideration. Designs that require a high degree of surface integrity are the ones that display the following qualities: Are highly stressed Employ low safety factors = Operate in severe environments e Must have prime reliabihty Have a high surface areas-tccvolume rabo Are made with alloys that are sensitive to processing
Recess eliminates deburring

Groove minimizes milling burr

Not this Burr contained wlthin V-groove

7C-l

FIGURE 35-19 Designing

extra recesses and grooves into a part may eliminate the need to deburr. (From L X . Ciiiespie, American Machinist, November
1985.)

Not this

This

ingineering Surface integrity should be a joint concern of manufacturing and engineering. Manufacturing must balance cost and producibility with design requirements. It bears repeating to say that engineering must design components with knowledge of manufacturing processes. A reduction in fatigue life resulting horn processing can be reversed with a posttreatment.This is another example of design for manufacturing. I t is important to understand that the various manufacturing and surface-finishing processes each impart distinct properties to the materials that will influence the performance of the product.The achievement of satisfactory product performance obviously depends on a good design, high-quality manufacturing (including surface treatment), and proper assembly.The failure of parts in service, however, is usually the result of a combination of factors. A brief survey of features associated with surfaces and surface processing follows. Each of the various machining processes produces characteristic surface textures (roughness, waviness, and lay) on the workpieces. In addition, the various processes tend to produce changes in the chemical, physical, mechanical, and metallagical properties on or near the surfaces that are created. For the most part, these changes are limited to a depth of 0.005 to 0.050 in. below the surface.The effects can be beneficial or detrimental, depending on the process, material, and function of the product. Machining processes (both chip forming and chipless) induce plastic deformation into the surface layer, as shown in Figure 35-20, The cut surfaces are generally left with tensile residual stresses, microcracks, and a hardness that is different from the bulk material. Processes such as EDM and laser machining leave a layer of hard, recast metal on the surface that usually contains microcracks. Ground surfaces can have either residual tension or residual compression, depending on the mix between chip formation and plowing or rubbing during the grinding operation. If sufficient heat is generated, phase transformations can occur in the surface and subsurface regions.
I

New

/surface
Workpiece
-\,'-

I.

&Of

. '

Depth

damage

FIGURE 35-20 Plastic

deformation in the surface layer 6 .W. Kruszynski after cuttlng. ( and C. W Cuttervelt, Advanced Manufactunng Eng~neenng,
Vof. 1, 1989.)

S ~ c r r 35.10 o~

Surface Integrity

963

Plast~cally

deformed layer

LD

oo

0.125shot
A

- 13.2-mmi stainless
steel shot
steel shot

Peened with 0.725-dia.

0 O

Peened with 0.156-dia. - 13.9-mm) stainless

500
c

1
J

'$
+
n

50 m_ m

Surface
la1
FlGURE 35-21

1 mm deep
lbi

2 mm deep

(a) Mechantsm for formation of resldual compresswe stresses in surfaceby cold plast~c deformation (shot peening). (b) Hardness Increased in surface due to shot peenlng.

Processes such as roller burnishing (described in Chapter 18)produce a smooth surface with compressive residual stresses. Shot peening (and tumbling) can increase the hardness in the surface and introduce a residual compressive stress, as shown in Figure 35-21a and 35-21b Welding processes produce tensile residual stresses as the deposited material shrinks upon cooling. Similar shrinkage occurs in castings,but the resulting stresses may be complex due to the variation of shrinkage or the lack of restraint.Tensile stresses on the surface can often be offset by a subsequent exposure to shot peening or tumbling. In summary, the surface and subsurface regions of a material can be significantly altered due to (1) plastic strain or plastic deformation, (2) high temperatures, (3) differential expansions or contractions due to temperature changes or variations, and (4) chemical reactions. To illustrate the complex nature of surface effects, consider Figure 35-22, which shows the depth of "surface damage" due to machining as a function of the rake angle of the tool.To increase the cutting speed (and thereby increase the rate of production), an engineer might change from a high-speed tool steel cutter with a large rake angle (such as 30") to a carbide tool with a zero rake. While the resulting surface finish may be similar, the depth of "surface damage" is doubled. Failures may occur in service, whereas previous parts had performed quite admirably.

The depth of damage to the surface of a machined part increases with decreasing rake angle of the cutting tool.
FlGURE 35-22

Rake angle, degrees

layer

Peened wirh 0.125-dia

0.125shot
0 0 Ln

a - 13.2-mmi stainless
Steel shot
Peened with 0.156-die

500

0.156 shot (3.9 mrn)

- 13.9-mmiStainless
steel shot

P
c
0

0.

400
Y

Surface
(a)

1 rnrn deep

2 rnrn deep

I b)

FIGURE 35-21 (a) Mechanism for formation of residual compressive stresses in surface by cold plastic deformation (shot peening). (b) Hardness increased in surface due to shot peening.

Processes such as roller burnishing (described in Chapter 18)produce a smooth surface with compressive residual stresses.Shot peening (and tumbling) canincrease the hardness in the surface and introduce a residual compressive stress, as shown inFigure 35-21a and 35-21b. Welding processes produce tensile residual stresses as the deposited material shrinks upon cooling. Similar shrinkage occurs in castings,but the resulting stresses may be complex due to the variation of shrinkage or the lack of restraint.Tensile stresses on the surface can often be offset by a subsequent exposure to shot peening or tumbling. In summary, the surface and subsurface regions of a material can be significantly altered due to (1) plastic strain or plastic deformation, (2) high temperatures, ( 3 ) differential expansions or contractions due to temperature changes or variations, and (4) chemical reactions. To illustrate the complex nature of surface effects, consider Figure 35-22, which shows the depth of "surface damage" due to machining as a function of the rake angle of the tool.To increase the cutting speed (and thereby increase the rate of production), an engineer might change from a high-speed tool steel cutter with a large rake angle (such as 30") to a carbide tool with a zero rake. While the resulting surface finish may be similar, the depth of "surface damage" is doubled. Failures may occur in service, whereas previous parts had performed quite admirably.

FIGURE 35-22 The depth of damage ta the surface of a with machined part ~ncreases dscreasins rake angle of the

cutting tool.

964

CHAPI~ 35 R

Surface Engineering

b(

Rotating beam with bending results in cycles of tension and compression

,
FIGURE 35-23 (Top) A

stress
~ ~ e s str? i ; ~ ~

cantilever-loaded (bent) rotating beam, showing the normal distribution of surface stresses (i.e., tension at the top and compression at the bottom). (Center) The residual stresses induced by roller burnishing or shot peening. (Bottom) Net stress pattern obtained when loading a surface-treated beam. The reduced magnitude of the tensile stresses contributes to increased fatigue life.

+4-

deformation 10~Urface

Beam glven roller burnlsh or shot peen to cold work surface and develop a restdual compressive stress

Beam in service with reduced peak tensile stresses due to additive nature of applied and residual stresses

INFLUENCE OF SURFACE FINISH O N FATIGUE Fatigue iailure occurs as the result of repeated loading at some point typically below the yield strength of the material. Fatigue failures have been shown to almost always, nucleate on or near the surface of a component. Fine surfacc cracks begin at discontinuities (such as microcracks. grooves, ridges, cavities, machining marks, imbedded particles. ctc.) at the surface, and the cracks propagate with repeated cyclic loads. Tensile residual stresses in the altered surface layer havc an additive effect on th applicd stresses in the component. This means that tensile residual stresses in the material add to cxternal stresses to the component, reducing its fatigue strength. Altern lively. as shown in Figure 35-23, compressive residual stresses subtract from tensid external stresses, and since tensile stresses are those ultimately responsible for fatigu failure. the fatigue strength of the material is increased. Figure 35-23 shows how residual stresses couple with applicd stresses to aff product performance. Suppose that a round beam has a load applied to it so that it is be while rotating. A t the top of the rotation, the surfacc is in tension, and at the bottom, is in compression.The result is a condition of cyclic fatigue and the likelihood of a ser life limited by fatigue failure. Tf the part is roller burnishcd or shot peened, compressive residual stress pattern of the middle Cigurc is added to the applied str es, producing the net pattern shown at the bottom.The net effect is a lowering of the tensile stress experienced by the surface and a related extension in fatigue 1ife.n cific results will depend on the details of the process.For shoipcening, the key varia include shot size, shot velocity, exposure time, distance between the nozzle and surfacc, and the angle of impact. Figure 35-24 presents the results of a study in which specimens were prepared milling and turning and then either polished, shot peened, or roller burnished. If an plied stress between 41,000 and 42,000 psi is experienced in a fatigue application,the fcrcnce in fatigue life between a milled specimen and one that has been milled andro burnished is 610,000 cycles (90.000 cycles as opposed to 700,000 cycles). In esseuce,ro burnishing serves to induce a sevenfold extension to the fatigue life of the pr0duct.S ilar results have been observed in the resistance to stress-corrosion cracking. As the data above show, both the designer and the manufacturer need to bea o l the effects that manufacturing processes can havc on the performance of a pro Maintaining the proper sequence of operations may be as important to the surfacep crties as the selection of the processes and control of the operating parameters.

a " i
Y)

41 * Y)

Surface-fin~sh~ng processes used to manufactore specimen

a Milled = MLD

3 i t 4 F a t i p liie of mtating beam 2024-T4 aluminum specimens wrth a variety of surface-f~n~sh~ng operations. Note the enhanced performance that can be ach~eved by shot peening and roller burnishing.

<
32
-

ep

o Turned = TRD
A Turned
7

and p o I i d = TPD Turned and shot peened = TSP o Milled and roller burntshed = MRB .Turned and roller burnished = TRB
N, fatigue life (cyclesto fallurel

."

Key Words
&bm$&eoleaning &cidpiokling alkslim cleaning hard chromiumplate hot-dip coatmg mechanical cleanmg mechanical plating shot peensolvent cleaning spray painting

media

su&&&c@ty * = I @ " =

buffing burr removal case hardening centniugtll barrel tumhlidg &mica1 sonumihi maring chemical dean& ehemical wapw deposihon

elecrrafommg electzolqs composite p k n g dpctroless pjaung el*ctroplating ekctrqbtishing elecbo%k%tic deposition fiwsixi~ compwnd8

porcela~n enameling powder coatlug prime coat restdual slresses roller burn~shing roughness sand blastmg

ult~iisonic clewing vacuum metalliking vapor dweaging vibratory hn~shing waviness wire brnh'mg

I I Review Questions
1. What are some possible objectives of surface engineering processes? 2. What are some of the factors tha* should be considered when selecting a surface-modification process? 3, How are the surface and its integrity altered by the process of metalcutting? 4 . Two surfaces can have the same microinch roughness but be different in appearance. Explain! 5. What limits the resolution of a stylus-type surface-measuring device in finding profiles? 6. What is the general relationship between surface roughness and tolerance? Between tolerance and cost to produce the surface andlor tolerance? 7. What are some of the sources of foreign material on the surface of manufactured products? & What are some the common abrasive media used in blasting or abrasive cleaning operations?
9. What types of quantities and part sizes are most attractive for barrel finishing operations? 10. Describe why there might be an optimum fill level in barrel finishing.How might you find the &ptimumrotational speed? 11. Describe the primary differences between barrel finishing and vibratory finishing. 12. What are some of the possible functions of the compounds that are used in abrasive finishing operations? 13. How is electropolishing different from electroplating? 14. What are some of the mechanisms of alkaline cleaning, and what types of soils can be removed? 15. What types of surface contaminants cannot be removed by solvent cleaning? 16. In view of its many attractive features, why has vapor degreasing become an unattractive process? 17. What is the primary type of surface contaminant removed by acid pickling?

C H A ~ 35 R Surface Engineer~ng
32. When anodizing aluminum, what features determine the thick-

What is the difference between coating and cladding operations? What are someof the reasons that paints may be specified for manufactured items? What are some of the functions of a prime coat in a painting operation? What features are desired in the final coat? What produces atomhation and propulsionin airlessspraying? What features make industrial robots attractive for spray painting'? What are some or the attractive features of electrostatic spraying? Why would it be difficult to apply electrostatic spray painting to products made from wood or plastic? Whatare some of the metal coatings that can be applied by the hot-dip process? What are the two most common types of chemical conversion coatings? How can nonconduclive materials such as plastic be coated by electroplating'? Whal are the attractive properties of hard chrome plate? What are some o l (he common pl.ocess variables in a n electroplating cell? Why is it difficult to mix parts of differing sizc and shape in an automated clectroplating system? How is electroforming different from electroplating?

ness of the resulting oxide when the oxide is not soluble in the electrolyte? When it is partially soluble? 33. What produces the various colors in 'the color anodizing process? 34. Whal are some of the attractive features of electroless plating? 35. What types of particulate composites can he deposited by electroless plating? 36. What is mechanical plating? 37. What are some of the attractive properties of a porcelain enamel coating? 38. How are burrs made by the milling process? See Figure 35-18 39. What dehurrrng processes are avdllable that were not described ~nthis chapter? 40. D o all machln~ng processes leave a res~dual stress? 41. Why would the depth of damdge (i.e., plastic deformatton) Increase w ~ t h negatlve rake angle cuttlng tools? 42. What types of surface features or surface mod~fications result from machmmg-type processes? 43 Why m ~ g h tprocesses that produce resldual compresswe stresses on product surfaces be attractive for mechanically loaded products?

[ I I

Problems
b Large quantities of steel bolts that are intended for use ~n ouldoor constructlon. They have been fabricated from 4140 steel, have a shank diameter of 'Iz m , and have been quenched and tempered to a frnal hardness of Rockwell C 45 c The handle of a household utility knlfe (retractable-blade cutter) has been made as a two-part zrnc die castmg. f an Inexpenswe ice cream scoop that d. The scoop portion o has been made as a zlnc die casnng. cdhinct that is made as e A decorative handle for a k~tchen d zinc die casting f. The exterior of an off~ce tiling cabinet that has been made from low-carbon steel sheet g A h~gh-qualitycomblnatlon wrench (open-end and boxend) that has been forged from 4147 steel bar stock h The case of a moderately pr~ced wnstwatch that has been fabr~cated from vellow brass (to have a gold appearance). i. Tubular frame of a lightweight bicycle that has been made from age-hardened aluminum. j. The basket section of a grocery-store shopping cart that has been fabricated from welded steel mesh. k. The exterior of a n automobile muffler to be fabricated from steel sheet.Descr~he how the coatlng treatment mlght best be integrated into the fabncation sequence. 1 A n inexpensive interior door knob that has been fabncated from deep-drawn cartr~dge brass sheet. m HI&-quality steel sockets for a socket-wrench set.These have been forged from AlSl4145 bar stock and subsequently heat treated by a quench-and-temper process. n . Refrigerator door panels that have been fabncated from
text~lreri ATST l n qteel . .. -. . - l f l- - qhept -- - - .

F~shermen are among the most snper$titlous people in thc world, and thcir superstitions affect the type of equipment that they use. As a result, hook manufacturers generally offer thelr products in a wide range ot colors and fin~shes Your company manufactures a range of hooks fromAISI1080 carbon steel wlre, forming them to precisron ahape (eye, bends, barbs, and po~nl) and then heat treating them by a quench-and temper treatment. Cons~der the size and shape of the product and the various properties that are required Thc hooks must he stron8 enough to resrst bending, hut not so br~ttlethat they m~ght break They must be corrosion-resistant to both fresh andsalt water, and the deslred appearance must be provided without foullng the polnt or the barbs. If the surface IS applied before heat treatment, it must endure that process and maintain its appearance If n is apphed afterward, it cannot weaken or emhnttle . - the ~hook. ~ ~ the various surface.modifkation processes, which ones a. might be attractive for such an application? (Note; Make sure that the process is appropriate! F~~examplo, barrel plating would probably produce a hopelessly snarled mass
~ ~

-e...:---~\ "L W l l C S , ,

b FOP each of the poss~ble processes, descnhe the advantages, cost. l~mitat~ona, posslble colors o r finishes, and relat~ve Select one or more of the following products (as directed) and recommend a surface treatment or coating Conslder the ap propnateness o i the technique to the slze, shape,quantity,and mater~al. Cne the spec~fic features that make the recommended treatment the most attractwe What, if any, are the pnmary l~m~tationa or production concerns? a. The exterlor housmg for themotor and drive umt of a chain saw that has been made as a magneaum-alloy die castmg.

-- - - - - - - -

and exterior surfaces of a lm0-ganon ~h~ storage tank that 1 s fabricated by weldrng 5000 serler

p. g. r
r.
t.

alum~num plates-Thewater wdl be held at room temperature and is tntended for human mn$umptlon A standard office paper c i ~ p A flashltghiht case that has h e n fabricated from deepdrawn yellow brass sheet. High-speeddnll brts that have been fabrrratedfromMU twl steel. Injection-molded ABS plastic wheel covers for cars that arelotended ro look like chrome-plated metal. Inexpmsive household ~cisqors that have been east from gay oast iron.

u lhc hiaJc c ~ J i tnyh-qualit! scr~adri\er th:tt has hcc~t ft~rgcd from.\lSI 11153stcel and quenched and tcrnpcrvd L o Kock-

well C 55, v. The eMerior df high-qualtly, thck-walled cast ahminum cookware. w . A b t h r m m smk basin made from deep-drawn 1008 steel sheet. x B e hody s ~ t i o o n f a child's toy wagon thilthasbeendeep drawn from 11308 6teel sheet.

Diana LynnWFagwe Lesson

ana Lynn has lust come from her INSY 3000 lab in the manufactunng processes course. The purpose of this lab was to introdurn her to merai fatigue and its bascc prmciples. Metal faague anses from the cyclic loading below the yield strength. It can be greatly influenced by the surface flnish applied to the metal. Dr. Payton, her instructor, ssid fatigue most lhkely accounts fot 90Y0of ail mechanical failures, so it Is Important for an englneer to understand how matefiais respond to fatigue cwrdlhons. The procedure of thts lab was for each student to fin~sh the aluminum specimen with the emery paper, then load the specimen into the fatigue machine (see Figure CS.35A). Students were required to record the number of cycles that was heeded to fracture the specimen. It was importapnthat thestudent wipe thespecimen dean and Inspect the speclmen for burrs, ndges, or flats before inserting it into the drive spmdle The presence of stress raisers can dewease the cumulatrve number of cyoles n e & d to sfart a fatigue crack and cause it m fall, hence reducing thefat~gue life and static strength of the metal. Small surface cracks, surface flows, ar machintng marks are examples of stress raisersTherelore, rt is important that one strlve to ellminate sfress raiser or surface flaws in the spsclmgn that will tie exposed 1 0 cychc ~oadigs. The expertmental factors for ffiis exljerimentw?ethe $U&CB finish [gill w e of energy peper), applied stress fete!, m d e e d i s e C f t mof lhe svrface finlsh (parallel or perpendrcular to *hespwtmen axis). Dr Payton explained that repeated applicatmns of stress can cause metals to fracture, even if all of the strcsscs are less than the yield tensile strength and less than me JII'-

mate strength of the matertal. Surface oondltionscan heavtly lhfluence fattgue life because most fattgue craoks start at the surface of a metal.Th6 fatigue data in Table CS 35A is to be analyzed using stahstical experimental design techniques and summarized in the ANOVA table. Factor A was the applied load (5Oand 55 lbs), factor B was the treatment (fine and coarse emery paper), and tactor C was the surfaoe flniSh direcfton (horuontaland vemcal).The experiment was replicatedso arotal of 16t e t s were run by the dass (2 x 2 x 2 x 2).The speclmen was tapered so that the stressactlng along the kst section (shaded) 13 constant.To make the specimen, you have to calculate 8 and the diameter o' (sf@ Flgure C5.35B). Here are some of the queStlons that Dana Lynn had to answer. 1. Analyze the f&tgu+a dafacollected by fhe dass using statistical experiment (facfonal expertment) des~gn techniques. Include all calculattons and summanze your results m an analwls of varLance (AMOVAJtable. 2. What are the effects ofstress lwel, surface fMlsR (grlt size in our experiment) and surfacetinish direction. Are there any tntemctrons between them on Fatigue life? Drscuss. 3. What effect would abrupt surfacechanges, s w h as tool mark8 or surface flaws, have on fatLLue life?

4. The c y W m specihen is desighed suGh Mat the applied stress actihg along f i e test section is cornstant.
WhV? ..

5. ~ $ is t the endurance limit f o r a metal?Why is the end~ranco limit an important criteria in many design applications? Does alum;num have an EL?

968

CfiAa~n 35

Surface Engineering

Hvrizontai
FINE
50 lb-~n
55 lb-in. 167000 145600 8g600
98800

YerticllI

COARSE
126700 116800
61300 59800

FINE
1Q2200

COARSE
88600
92600 49400
41200

786W
56500
h3200

Specimen,

Bear~ng

Weight,

Figure CS 35A Rotary beam fatigue testlng rnach~ne,used for

cyl~ndr~cal specimens

Drrve sp~ndle I collet

Constant stress

Load arm / collet

wmax =

MC I

"

Applied load = P
= -----

Figure CS 35B Cyl~ndr~cal speclmen w ~ t h constant stress sectlon d Determine the ahgle 0 and the d~arneter

. Makmg FC Studies by the Traditional Methods H~tograms Run Chart or Diagram Process Capability Indexes Discusz~on o f Process Capabdlty Scenarios 36.3 I N S P E ~ O TO N CONTROL QUALI~ StatisticalProcess Control (SPC)
-

DETERN~INA~"~ON FROM cbNTROL CHART DATA 36.S DSTEBMIWG CAUSESFOR PROBLEMS INQUALITY Samplmg Errors Gage Capability Design of Expenmentr (DOE.) and Taguqhi Methcrds Mvtorala's Six SIgma Total Quality Control (TQC) Lie Stup in Lean Production

Implementing Quahty Companywlde Making Quality Visible Source, SelE,and Sllccessful Checks and Paka-Yokes Teams (aka Quallty Circles) Superior Qual~ty in ManufaeturinglAssembly C a a 36.6 SUMMARY Case Study: BORING QC CHART BLUNDBRS

r\ll rn~lllui3ctl1riiig procr,scs displa! ,onlc Ic+clof \;tri:~t~on. No t\vc) itcnii corningfr,~ni the ptoccss will he cx11~11!,111~. s~n~c.'l'hc pri~n:iryohj?:ti\r. ofqu.ilir!. enginccring is t l ~ c systematic reduction of variability, as shown schematicallyin Ftgure 36-1.Variability is measured by sigma, rr, the standard deviatim, which decreases w ~ t h the reduction in variahrlit)?.Early on, acceptance sampling techniques were used to screen incornlog goods. This was followed by statistical proce6-s control (SPC) efforts, whch gave way to companywide quality control (CWQC) and Total Qual~ty Control (TQC) programs techniquas,like multiVariation can be further rcduted by the appIlcation of s~atrstical ple variable analysis, designed experiments,andTaguchi methods, techniques fiat are a routine part of six sigma efforts many compankz are implementing.The drive toward zero defects has been led by Toyota, which achieved exceptional levels of quality by redesignzng the manufacturing system so that eachstep in the makrng of the car and aU

Uppa? specification limit

L SL-L ow~s~ec~ I ~ca ilmi5 t~on - - Methadologies and Acwptance technquueb sampiing

U ' I 1 i I
-- - - -

- - - -

----

$PC

TOM

6s1gma

SQC

TQC
CWQC

DOE
Taguchl
metbods

- Lean MSD Poka-yoke

Over many years, many techniques have been used to reduce the vanabiliv in product5 and processes.
FIGURE 36-1

969

mg how well the product conforms to the specifications. And within that area, we concentrate on PC s t u d ~ e s that are directed at the machlne tools used in the p

shots), whereas preclslon reflects the repeatabil~ty of the process The objective 1s

PROCESS CAPABILITY
The nature oftheprocess refers to both the vanuhiltfy (or inherent un1formi"tyJ and accuracy or the a m of the process Thub in the target-shooting example, aperfect proc would be capable of placing ntne shots right in the middle of the bull's-eye, onenght

Accuracy hlgh, preclston low

Accurate and
precise

FIGURE 36-2 The concepts of

accuracy (aim) and precision (repeatability) are shown in the four target outcomes. Accuracy refers to the ability of the process to hit the true value (nominal) on the average, while precision is a measure of the inherent variability of the process.

Accuracy poor, prec~sion poor

low

precision

Accuracy poor, good

Precbsion (Processvar~ab~litv)

SECTION 36.2 Determining Process Capability

971

top of the other.The process would display no varlabillty with perfect accuracy. Such performance would be very unusual m a real Industrial process. The vanability may have assignable causes and may be correctable if the cause can be faund and eliminated.That variability to which no cause can be asslgned and which cannot be eliminated is s a ~ d to be Inherent m the process and is therefore its nature. Some examples of assignable causes of vanation m processes mclude multiple machines for the same components, operator blunders, defectwe materials, or progressive wear in the tools dunng machining. Sources of inherent variability in the process include variation m materlal properties operator variability vibrations and chatter, and the wear of the shdiig components in the machine, perhaps resultmg in poorer operation of the machine These kinds of variations, which occur naturally in processes, usually display a random nature and often cannot be eliminated. In quahty control terms, these are referred to as chatzce causes.Sometimes the causes of assignable variation cannot be eliminated because of cost. Almost every process has multlple causes of variability occurrlng simultaneously,so it is extremely difficult to separate the effects of the different sources of var~ability during the analys~s.

MAKING PC STUDIES BY THE TRADITIONAL METHODS The object of the PC study 1s to determine the inherent nature of the process as compared to the deslred specifications. The output of the process must be examined under normal conditions, or what is typically called hands-offconditzons. The inputs (e.g., materials, setups, cycle times, temperature, pressure, and operator) are fixed or standardized.The process is allowed to run without tinkering or adjusting, while the output (i e ,the product or unlts or components) is documented with respect to (1) time, (2) source, and (3) order of production. A sufficient number of data have to be taken to ensure confidence in the statist~cal analysis of the data.The capability of the gage (its precis~on) used to measme the products must exceed the expected tolerance on the part by one order of magmtude (See the discuss~on of the rule of 10 in Chapter 10.) Prior to any data collection, these steps must be taken: 1. Design the PC experiment (standard method) Use normal or hands-off process cond~tlons; specify machine settings for speed, ieed, volume, pressure, material, temperature. operator, and so on 2. Define the inspection method and the inspection means (the procedure and the instrumentation). In selecting the gage, consider these aspects. a. Features that the gage will be checking b. Speed or rate of operation c. Level of accuracy and precision d. Skill of the operator e. Portabrlity of gages or part, or both E Env~ronrnent (clean and stable, cuttmg fluids) g. Workpiece (clean, lubricants present) h. Cost (initla], maintenance, dally)
3. Decide how many items (measurements) will be needed to perform the statistical analysis. 4. For a standard PC study, use homogeneous input materlal, and try to contrast it with ~rormal (more vanable) Input mater~al 5. Data sheets must be deslgned to record date. tlme, source, order of production, and all the process parameters bemgused (or measured) while the data are bemg gathered. 6. Assuming that the standard PC study approach is being used, the process e run,and the parts are made and measured.
specified Now follow the steps outlined in Figure 36-3. Assume that the des~gner the part to be 1.000 z0.005 m. After manufacturing engineering has developed a process plan, some units are manufactured according to the process plan without any adjustment of the process. Each unit is measured, and the data are recorded on the data sheet.

972

CHAFTER 36

Quality Engineering

START

CAD drawing

Designer specifies nominal (desired) size of 1.000 and specifications (iTolerances i0.005) USL = Upper specification limit = nominal + tolerance LSL = Lower specification limit = nominal -tolerance

Information t o manufacturing engineering

manufacture the part

(mach~ne tool + Raw mater~als

lnspen by measurement device or gages

Dtshlbutio!, of~ndividuala x's

la hlstograml

(A t u n chart of x's)

U6L-

- -

Tim% From the data, calcuiate a) Mean afthe pmceas t o m m p a r e t o nominal. b) Standard deviation of procewrr, to compare USL tSLm 3v

So the process
i s producing

parts which, when measured become the parent

population but mean

-Parts

SEC'I'I~N 36.2

Determining Process Capability

973

Manufacturmg

process

All the parts the

process wdl make ln lts Ilfetrrne.

The parent papulatlon I*= populat~on mean v = population standard dev~at~on

standard

dev~at~on not known

t h e first 70

parts from the process and makea h~stogram

So we look at a sample of

__t

estimate rr a n d p the foliowlng analysis IS used

To

where is estimated by 2 n = sample size = 70

and
s=

JZ(X,-T)~
n-1

where U. 16 est~mated by s.
of70 measurements of a parameter H~stagrarn The deslgn speclficatlonwas 1 001 + 0 005 in

The nominal was 1.001 in.

FIGURE 36-4 Example of

calculat~ons to obta~n est~mates of the mean (p)and standard of a process devlat~on(u)

A frequency distribution, in the form of a histogram, or a run chart is developed.7his histogram shows the raw data and the desired value, along with the upper and lower specification lirnirs, where LSL represents lower specification limit and USL the upper specification limit.The statistical data are used to estimate the mean and the standard deviation of this distribution.The run chart shows the same data, but here the data are plotted against time. The mechanics of this statistical analysis are outlined in Figure 36-4.The true mean of the distribution, designated p, is to be compared with the nominal value specified by deviation, designated cr (sigma), is used the designer. The estimate of the true st~zndard to determine how the process compares with the desired tolerance. Thepurpose of the analysis is to obtain extimates of @and uvalues, the true process parameters, hecawe they are not known.

974

CHAPTER 36

Quality Engineering

Centering (average1
FlCURE 36-5 The normal or bell-shaped curve with the areas ,i Zu, and -t 3ufor within t 1r a normal distribution; 68.26% of the observations will fall within i-1 ufrorn the mean, and 99.73% will fall within -t 3 0 from the mean.

-3u

-2"

-1 v

+1c

t2r

+3u

The process capability is defined by i3mor 6u.Thus p t 3udefines the natura capability limits of the process, assuming the process is approximately normally dis tributed. Note that a distinction is made between a sample and a population. A sampli is of a specified, limited size and is drawn from the population.The population is the largi source of items, which can include all the items the process wiU ever produce under th~ specified conditions. Our calculations assume that this distribution was normal or bell shaped. Figure 36-5 shows a typical normal curve and the areas under the curve a defined by the standard deviation. Other distributions,shown in Figure 36-6, are possible but the histogram clearly suggested that this process can best be described by a norma probability distribution. Now it remains for the process engineer and the operator t~ combine their knowledge of the process with the results from the analysis in orde to draw conclusions about the ability of this process to meet specifications.

HISTOGRAMS A histogram is a representation of a frequency distribution that uses rectangle whose widths represent class intervals and whose heights are proportional to the corre sponding frequencies. The frequency histogram is a type of diagram in which data ar grouped into cells (or intervals), and the frequency of observations falling into eac interval can be noted. All the observations within a cell are considered to have the sam value, which is the midpoint of the cell. So, a histogram is a picture that describes th variation in a process. It is good to have this visual impression of the distribution c values, along with the mean and standard deviation. Histograms are used in many way in QC, for example:
To determine the process capability (central tendency and dispersion) To compare the process with the speclficat~ons To suggest the shape of the population (e.g., normality) To indicate discrepancies in data, such as gaps There are several types of histograms. A histogram shows either absolute frequency (ac tual occurrence) or relative frequency (percentage). Cumulative histograms show CI mulative frequency and reliability cumulative frequency. Each type has its ow advantages and is used in different situations. Figure 36-7 shows frequency versus loc; tion for 150 measurements. The aim (accuracy) of the process is a bit low, but all th data are well within the tolerances.The disadvantage of the histogram is that it does n c show trends and does not take time into account. We can take the data from the hi togram and spread them out over time to create a run chart or run diagram.

RUN CHART OR DIAGRAM A run diagram is a plot of a quality characteristic as a function of time. It providr some idea of general trends and degree of variability. Run charts reveal information th;

SECTION 36.2 Determining Process Capability

975

Common probability distributions

Distribution Form Probabilitv fundtion Comments

Normal

Applicable when there i sa concentration o f observations about the average and it is (X-p"2 equally likely that y= e 7 UG 2u observations will p = Mean occur above and below the average. rr = Standard deviation Variation in observations is usually the result of many small causes

y=

-1 e
P

-5

Applicable when it is likely that more observations will occur below the average than above describing a wide f variety of panerns o variation, including depanures from the
normal a n d
,

n=
y =
X . .

B = Shape parameter
Location parameter

, ap(X Scale
=

- ,,]P-le-*rx-y)@ parameter

Poisson*
p = 0.3
'=

(npl'e" rl

r Binomial*

n = Number of trials r = Number of occurrences (less than 0,10) on p = Probability of occurrence each trial

exoonential Same as binomial but particularly applicable when there are many opportunities forevent occurence o f an but a low probability

n = Number of trials

the probability of r occurrences in n trials

I
;

each trial

dtscrate d~stribut#ons but shown as curves for ease of comparison

FIGURE 36-6 Common probability distributions that can be used to describe the outputs from manufacturing processes. (Source: Quality Control Handbook, 3rded.)

histograms cannot, such as certain trends over time or at certain times of day. Individual measurements (not samples) are taken at regular time intervals, and the points are plotted on a connected line graph as a function of time. The graph can be used to find obvious trends in the process, as shown in Figure 36-8, a run diagram with measurements made every hour over four shifts. Run diagrams are very important at startup to identify the basic nature of a process. Without this information, one may use an inappropriate tool in analyzing the data. For example, a control chart or histogram might hide tool wear if frequent tool changes and adjustmenls are made between groups of observations. As a result, run diagrams (with 100% inspection where feasible) should always precede the use of control charts for averages and ranges

976

CHAPTER 36

Quality Engineering Tolerance

45

b -

39

Process Capabiilty Mean = +

4 -

"

38

40

42

44

46

48

50

52

54

56

58

shows the output mean ir. from the processversus nominal and FIGURE 36-7 H~stogram the tolerance specifled by the designer versus the spread as measured by the standard dev~at~on LT. Here nominal = 49.2, USL = 60, LSL = 38, p = 50.2, o = 2.
The lengths of manufactured components are measured A run

how the process IS behaving. diagram IS constructed to determ~ne Dur~ng 34 hours, measurements are made every hour (60 minutes) and plotted in the order that rack bars are produced. First shlft Secondshlft Thlrd shlft F~rst shift
35 24 15 16 40 23 13 5 27 28 28 9 30 15 8 13 30 23 20 16 34 17
9

10

26 16 5 9

31 21 11 10

5
25A

FIGURE 36-8 An example of a run chart or graph, which can reveal trends in the process behavior not shown by the h~stogram.

10 5 1 2 4
8 10

16

22

26

30

34

36

Order of product~on or tlme

PROCESS CAPABILITY INDEXES

The most popular PC index tells y o u if the process has the ability t o meet spc This process capability index, C ,, is often computed as follows: spread USL - LSL c, = tolerance 6u 6u

A value of C,

1.33 is considered good

SECTION 36.2 Determining Process Capability The example in Figure 36-7 has
- LSL - 2 1 --= c, = USL 6u 12

977

Theprocess capability ratio, C , does not, however, take into account the location of the process mean, p, with respect to the nominal or the specdications. C, mearly looks at the variability or spread of process (compared to specifications) in terms of sigmas So another process capability ratio has been developed for off-center processes. This ratio is called CPk, where

USL - p. , c,, = 3a

3u

C , , is simply a one-sided ratio for the specification nearest to the process average, p To compare the two, look at Figure 36-9. All the histograms have the same standard deviation (u= 2) and the same USL - LSL specifications (62-37). For Figure 36-!la, the two indexes are the same. For Figure 36-9b, the mean has shifted to p = 54 so Cpk = (62-54)/3(2) = 1.5. Can you calculate C,, for (c), (d), and (e)? Answers are on the figure

LSL

USL

(a1

c, = 2.0 c,, = 2.0

(el

38

ah=
,C , = -0.5
62
65

44

50

56

FIGURE 36-9 The output from the process is sh~fting toward the USL, which changes the C,, ratio but not the C, ratio.

Engineering
The

process is very

capable, w ~ t h
6 r <USL-LSL,

LSL

U5L

well centered and ( a ) stable (mean not d r ~ f u n g ) The process is very capable but has drlfted off center
Corrective actlon needed to realm or
(b)

,
38
44

c,=

2.0

50

56

62

Capab~lity ti 6rr
Nominal

Out Of C,,=20

center the process

38
44

value
I

cpk=oo
56
62

50

The process 1 s j u s t barely capable, 6 w = USL-LSL stable, d n d centered but the v a r ~ a b ~ l should ~ty be
(c) decreased.
38
44

C,= 1.0

50

56

62

The process 1 s centered but

IS not capable, 6 rr > USL-LSL. Corrective actlon needed on the bas~c process to (dl reduce varlabll~ty.

C,< I
h

CP, < 1

62 Out Of

FIGURE 36-10 Five different scenarios for a process output versus the designer's specifications for the minimal (50) and upper and lower specifications of 65 and 38 respectively.

The process has shlfted and has Increased the variab~l~ty (el Multiple causes are likely.
38
44

Outof

\
50 56 62

C,=1.33 Cob-0.66

The capability indexes can tell you about the variance, where the wldth of the histogram 1s compared with the specifications (Figure 36-10). The natural spread of the process, brr, is computed and is then compared w ~ t h the upper and lower tolerance limits. These situations can exist:

1 . 6 u < USL - LSL or C, > 1,or process vanability less than tolerance spread. See Figure 36-10a. 2. 6 r < USL - LSL but process has shifted. See Figure 36-1Oh. 3. 6 u < USL - LSL or C, = 1,or process var~ability is just equal to tolerance spread. See Figure 3&10c. , -.: 1,or process varfability is greater than tolerance spread. 4 . 6 r < USL - LSL or C See Figure 36-10d. 5. The process mean and variability have both changed. See Figure 36-10d.
DISCUSSION OF PROCESS CAPABILITY SCENARIOS In s~tuation F~gure 36-10a,the machlne 1s capable of meet~ng the tolerances applied by the des~gner. Generally speaking if prooess capab~lity is on the order of two-thirds ta three-fourths of the design tolerance, there is a high probabil~ty that the process will produce all good parts over a Long period o f time. If the PC iS on the order of one-half or less of the design tolerance, it may be that the selecled process is too good; that is, the company may be producing ball bearings when what 1s called for IS marbles. In this case, it may be possible to trade off some precis~on in this process far looser specifications elsewhere,resulting m an overall economic gain. Quality in well-behaved processes can be maintained by checking the first, middle, and last par$ of a lot or production run. I f these parts are good, then the lot is certain to be good. This is called n = 3.

Naturally, if the lot size is 3 or less, this is 100% inspection. Sampling and control charts are also used under these conditions to maintain the process aim and variability. When the process is not capable of meeting the design specifications,there are a variety of alternatives, including the following: 1 . Shifting this job to another machine with greater process capability 2. Getting a review of the specifications to see if they may be relaxed. 3. Sorting the product to separate the good from the bad.This entails 100% inspection of the product, which may not be a feasible economic alternative unless it can he done automatically. Automatic sorting of the product on a 100% basis can ensure near-perfect quality of all the accepted parts. The automated station shown in Figure 36-11 checks parts for the proper diameter with the aid of a linear variable dzerential transformer (LVDT). As a part approaches the inspection station on a motor-driven conveyor system, a computer-based controller activates a clamping device. Embedded in the clamp is an LVDT position sensor with which the control computer can measure the diameter of the part. Once the measurement has been made, the computer releases the clamp, allowing the part to be carried away If the diameter of the part is within a given tolerance, a solenoid-actuated gate operated by the computer lets the part pass. Otherwise, the part is ejected into a bin. With the fast-responding LVDT, 100% of manufactured parts can be automatically sorted quickly and economically. Sortmg to find defects by automatic inspection 1s bad because you already p a d to produce the defects. Also, automated sortmg does not determine what caused the defects, so this example is an "automated defeit finder." How would one change this Inspection system to make it "inspect to prevent" the defect from occurring? 4. Determining whether theprecr~ion (repeatability) of the process can he improved by: a. Switching cutting tools, workholding devices, or materials b. Overhauling the existing process andlor developing a preventive maintenance program c. Finding and eliminating the causes of variability, using cause-and-effect diagrams d. Combinations of (a), (b), and (c) e. Using designed experiments and Taguchi methods to reduce the variability of the process In Figure 36-10c, the process capability is almost exactly equal to the assigned tolerance spread, so if the process is not perfectly centered, defective products wiU always result. Thus, this situation should be treated like the situation in Figure 36-10d unless the process can be perfectly centered and maintained.Too1wear,which causes the distribution to shift, must be negligible.
clamp J

Parts

f'

activated ejector

FIGURE 36-11 A linear variable differential transformer (LVDT) is a key element in an inspection station checking part diameters. Momentarily clamped into the sensor fixture, a part pushed the LVDT armature into the device winding. The LVDT output is proportional to the displacement of the armature. The transformer makes highly accurate measurements over a small displacement range.

differential transformer

:HAPTF.R

36 Quality Engineerfng

PC studies can evaluate the ability of the process to maintain centering so th avefa&e of the distribution cOMes as close as possible to the deired nominal v Most promses can be re-aimed. Poor a can be eliminated. In addition to d k t

duction prcfceeses,and vice ver$a. If uct are carefuuy recnrded, i n f m a t i m about the instantaneous found and compared with the repeatability of the prgcess w~th respect to t i e time variability). More important, since almost a l l proeeases are generate information about machine-to-machine variabrlity, G shooting esample, suppose that nine different guns were used, type.The results would have been different, just as having nine marRmnen urre the gun would have resul a h u t the homogene highly dependent on the skill (the capability) of the*'worlrer."Procwses that

ple, themean of apmess @) will usually shift after a tool decrease or remain unchangsd. Machjnes tend to beco varrability within a sample) after they have been * surfaces have smoothed out becqme of wear) but less repeatabrlity) due to poor fits between chine under varying lwads. Other variables thazcan dominate processes are setup, parameters, and even information, In many machining processes in use today, the task of tool se placed by an automatictool-positioningcapability, which means that one source a h i t y in the process has been eliminated,m&ing the proms$ more repeatable, %melight, it will be very important in the future f o manufacturingenginews ~ to the process capability of robots they want to use in the workplace. Thediscussion to this point has assumed that theparent uibuted, that is, has thcdassic bell-shaped disttibution in wh in Figure 36-5)are dictated by the number of standard devi or mean. The shape of the histo to thelefior theright (u Drilled holes exhibit s possibility is a bimodal dsiribution (two distinct peaks), often caused by two being mvred together. S a~?d you were using a check 12. The time to perform an aesembly are bimodal? Intheddesign,shown shared by two operators.The data the same way, since there appem to be a 5-or 6to complete this task. CIearly further study is nee was. For the next check sh&et,usea differentxymbol for each worker. Tht- check sheet is an excellent way t o view data whrle it is being collected be c m t r u c t d using predetermined parameters based on experience system. The appropriate interval is checked as the data are being colle f thp; data to be seen.The allows the oentral tendency and the spread o provide basically rhe same information as the the check sheet is formatted].lge possibilities in$ of all !ha sources of the data to track down the factors that result i and accuracy in the process Rapid feedback on quality 1s perhaps the mast fact~r, s o these data-gathering tasks are done right an the factory floor, by the

SECTION 36.3 lnspect~on to Control Quality

981

FIGURE 36-12 Example of a

check sheet for gather~ng data on a process.

Time (in seconds) to perform an assembly task in a cell.

W 36.3 INSPECTION TO CONTROL Q U

A L ~

In virtually all manufacturing, it is extremely important that the dimensions and qual~ t of y individual parts be known and maintained. This is of particular importance where large quantities of parts, often made in widely separatedplants,must be capable of interchangeable assembly. 0therwise.difficulty may be expenenced in subsequent assembly or 1n service, and costly delays and failures may result. In recent years, defective products resulting in death or injury to the user have resulted in expensive Iitigation and damage awmds against manufacturers. Inspection 18 the function that controls the quality (e.g., the dimensions, the performance, and the color) manually, by using operators or inspectors, or automatically, with machines, as discussed previously. The econooucs-based questlon "How much should be inspected?" has three possible answers:

l . Inspect every item being made. 100% inspect every item being made; 100% cheeking
with prompt execution of feedback and immediate corrective action can ensureperfecf quality. 2. Sample Inspect same of the product by samphng and make dalsions about the qual~ t of y the process based on the sample. 3. Nom. Assume that everything made is acoeplable or that the product is inspected by the consumer, who will exchange it if it is defective (This Is not a recommended procedure). The reasons for not inspecting all of the product (~.a,forsamphg) mclude the following: process is continuing to make the item--so we have to look at some before we are done with all. 2,The test is destructive. 3. There is too much product for all of it to be inspected. 4. The testing takes too much time or n too complex or too expensive. 5. It 1s not econom~cally feasible t o inspect everything even though the test i s simple, cheap, and quick. Some characteristics are nondissectible; that i$ they cannot be measured during the manufacturingprocess because they do not exist untilafter a whok series of operations have taken place. The final edge geometry of a razor blade is a good example, as is the yield strength of a rolled bar of steel. Sampling (looking at some percentage of the whole) requires theuse of statisucal techniques that permrt decisions about the acceptabtlity of the whole based on the quality found in the sample.This is known as statistical process control.

1. Everything has not yet been manufactured-the

Quality Englntling

STATISTtCtAL PROCESS CONTROL (SPC) Looang at some (sampling) and decidfng about the behayforof the whole (the parent popuiaiion) is commonin industrialinsp":ction operations T h e mast widtry w d basic SPC technique is the contr01 cham Ioparticular, c a m 1 & ~ ~ ~ f ~ ? v & b l eace S used to monitor the outpur of a process by sampling (lookiig at Some),itpmeasnring$electedquality chxracterhtics,by plottingthe -It? data on the dart,mdtbenby

a r used ~ in place o f R chart%,

(mean]Z = 7
n = simple size

( f W e l R = -%IIOH - xmw Sometimes the stantlard deviatlan IS cahuiated.

n = sample size

Dlstrlbutlon of lndivldual measurements

of x

-----After ksarnples, calculate the average range Rand the process average X Sample no ortime
t

3 c x = 3 c - 3 - = AR 2R 4 7 - d2fi

control chart calculations O n the charts, FIGURE 36-13 Qual~ty UCL plot and R values oveltlme. The constants for calculat~ng and LCL valuesfor the X a n d R charts are based on 3 standard dev~at~ons.

The samples are drawn over time. Because some sample statistics tend to be normally distributed about their own mean, X values are normally distributed about R values are normally distributed about R, and u values are normally distributed about F. Quality control charts are widely used as aids in maintaining quality and in achieving the objective of detecting trends in quality variation before defective parts are ac~ually produced.These charts are based on the previously discussed concept that if only chance causes of variation are present, the deviation from the specified dimension or attribute will fall within predetermined limits. When sampling inspection is used, the typical sample sizes are from 3 to about 12 units. The chart tracks the sample averages values).The R chart plots the range values (R values). Figure 36-14 shows one example of X and R charts for measuring a dimension of a gap on a part called the retainers. Twenty-five samples of size 5 were taken over six days, and this sample data will be used to prepare the control charts. The centerline of the X chart was computed prior to actual usage of the charts in control work:

F$

(x

where X was a sample average and k was the number of sample averages. The horizontal axis for the charts is time, thus indicating when the sample was taken. XI serves as an estimate for p, the true center of the process distribution. 7 is also centerline of the X chart. The upper and lower control limits are commonly based on 3 standard error units,3cr, (ux is the standard deviations for the distribution of X's about F). Thus, UCL7
=

upper control limit on Xchart

LCL,

lower control limit, z c h a r t

=p -

3u,-

(see Figure 36-13 for A, values). The upper and lower control l~msts are entered as dashed lines on the chart.The X chart is used to track the central tendency (aim) of the process. In this example, the samples were bang taken 4 times a day.The R chart is used to track the vanab~lity or dispersion of the process A u chdrt could also be used R is - xLOW).The value of E 1 s calculated as. computed for each sample (xHICN

where R represents the average range of k range values.The range values are normally distr~buted about% with standard deviation r,.To determne the upper and lower control hmits for the charts, the followmg relationsh~ps are used. UCL, LCL,
=
=

upper control limit, R chart lower control limit

=

R + 3uR = D,R

= 2.11Rfor n = 5

R = 3uR = D3E = 0

where D, and D3 are constants and are given in Figure 36-13 For small values of n, the distance between centerlme R and LCL is more than 3u,, but LCL, cannot be negative, as negative range values are not allowed, by definition Hence, D3 = 0 for values of n up to 6.

984

CHAPTER 36

Quality Engineering

VARIABLES CONTROL C H A R T X & R Averages &Ranges PanIAsm Name Retarner Part No 1234567 Parameter Gap D m "A"

1 operation 8end ~ l r p 1 Depalfment 105

Machtna 030

1 Spec~ficatlon.SO- .90mm I NomlnaI Slre 70 mm Gage Depth Gage Mlcrometer I Sample S~zeffrequency 512 Hours I

X and R charts and the data set of 25 samples [k =25 of size 5 (n = 5)]. (Source : Continuing Pracess Critical and Process Capability Improvement, Statistical Methods Office, Ford Motor Co., 1985.)
FIGURE 36-14 Example of

After control charts have been established, and the average and range values have been plotted for each sample group, the charts act as a control indicator for the process. If the process is operating under chance cause conditions, the data will appear random (will have no trends or pattern). If R, or u values fall outside the control limits or if nonrandom trends occur (like 7 points on one side of the central line or 6 successive increasing or decreasing points appear), an assignable cause or change may have occurred, and some action should be taken to correct the problem. Trends in the control charts often indicate the existence of an assignable cause factor before the process actually produces a point outside the control limit. In grinding operations, wheel wear (wheel undersize) results in the parts becoming oversized a n A

r,

SEC~ION 36.4 Process Capah~l~ty Determ~nation from Control Chart Data

985

corrective action should be taken. (Redress and reset the wheel or replace it with a new wheel ) Note that defective parts can be produced even if the polnts on the charts mdicate the process is m control. That is, it is possible for something to change in the process, causlng defective parts to be made,and the sample point still to be within the control limits. Since no corrective acbon was suggested by the charts, a type I1 error was made. Subsequent operations will then involve performing addltioilal work on products already defective. Thus the effectiveness of the SPC approach In lmprovlng quality IS often deterred by the time lag between the d~scavery of an abnormahty and the corrective action. With regard to control charts in genera1,lt should be kept in mind that the charts are only capable of indicating that somethmg has happened,not what happened, and that a certain amount of detective work will be necessary to find out what has occurred to cause a break from the random, normal pattern of sample points on the charts. Keeping careful track of when and where the sample wa? taken wlll he very helpful in such invest~gatious, but the best procedure is to have the operator take the data and run the chart. In this way,quabty feedback is very rapid and the causes of defects readily found.

After the process 1s determined to be "under control," the data can be used to eshmate the process capabihty parameters. As an example, examine the data in F~gure 36-14.These are measurements of the gaps in 125 retainers. They are supposed to have a nominal slze of 0.70 mm. The populat~on is assumed to be normal, and only chance vanations are occurring. (Could you make a histogram of these 125 measurements?) The mean for all the data can be obta~ned by equation 36-4

where x, la an indiv~dual measurement and the number of Items was 125.The mean 1s a measure of the central tendency around which the indrv~dual measurements tend to group.The vanability of the lndrvidual measurements about the average may be Indicated by the standard deviation, u, where

7/%("; -q2
(7

,=1

where n

125

But more commonly the sample data are used to make detetminatlons of the process control. A sample size of 5 was used in this example,so n = 5.Twenty-five groups or samples were drawn from the process, so k = 25. For each sample, the snmple mean T and the sample range R are computed. [For large samples (n > 12), the standard deviation of each sample should be computed rather than the range.] Next, the average of the sample averages, F,is computed as shown m Flgure 33-14 This IS sometimes called the grand nverage.This n used to estimate the mean of the process,p.Thestandard deviationof the process,whlch 1s ameasure of the spread or varlabllity of theprocesqis est~mated fromeither the average of the sample ranges,z, or the average of the samplestandard dev~atlons, u,us~ngeither~/d~or~/c factors ~ . T h d2 e and c,depend on thesample sizen and are glven tn the table in Figure 36-13 So , un estimated by 7 and 61sest~mated by R/d2or F/L, The standard dewation of the distribution of the values about 5?.OF, to the standard deviation of the parent population, u, by = a/&.
IS related

Now these estimates can be used to determine the process capabihty of the process In the same way the histogram was used

CAUSES FOR PROBLEMS IN QUALITY

The best way to quickly isolate qualrty problems i s tomake everyone an Inspector This means every worker, foreman, supervisor, engmeer, manager, and so Corth is responslble tor making it right the first tune and every time. One very helpful tool in thls effort is the fishbone d1agram.A~ shown in Figure 36-15, the fishbone diagram can be used In conjunction with the control chart to root out the cauyes of problems.The problem can have mult~ple causes, but in general, the cause will lie in the process, operators, materials, or method (i.e., the four main branches on the chart). Every trme a qnahty problem i5 caused by one of these events, it is noted by the observer, and correctwe actlon 1s taken. As betore, experimental des~gn procedures to be discussed later can help ~dent~fy causes that affect performance Cause-and-effect (C&E) dragrams are also known as fishbone diagrams because of their structure. Init~ally developed by Kaorw Tshikowa in 1943, this diagram organlzes theones about the probable cause of aproblem On the main line is a quality characteriqtic that is to be improved or the quallty problem being rnvestrgated. F~shbone l ~ n e are s drawn from the maln 1ine.These lines organlze the main factors that could have caused the problem. Branchmg ti-om each of these factors are even more detarled factors. Everyone takmg part in rnaklng a diagram galns new knowledge of the process? When a dlagram aerves as a focus for the discussion, everyone knows the topic, and the conversatron does not stray. The diagram 1s often structured around four branches the machine tools (or processes), the operators (workers), the method, and the material berng processed. Another version of the diagram IS called the CEDAC, the cause-andeffectdidgram w ~ t h the add~tlon ot cards.The effect is often tracked with a control chart The possible causes of the defect or problem are written on cdrds and rnserted in slot I n the charts. The three main applications of C&E d~agrams are as follows:

I. Cnuse enumerution: Every poss~ble cause and snbcause 1s listed. a. V~~unlpresmfntrons are one of the most widely used graphical techniques for Q
b. A better understanding of the relationship$ wrthln the process yields a bette. understanding of the process as a whole

Causes

Material
Contamination

One speciflc quallty characterlst~c of the Control chart track~ng devi defects in electron~c

------..----

N~trogen air guns Method

FIGURE 36-15

Example of a fishbone dtagram using a control chart to show effects

1 1 . Dispersion analysis involves grouping causes under similar headings; the four M's are men, machines, materials, and methods. a. Each major cause is thoroughly analyzed. b. There is the possibility of not identifying the root cause (may not fall into main categories).
1 1 1 . Process analysis is similar to creating a flow diagram.

a. Each part of the process is listed in the sequence in which operations are performed. In summary. data are gathered to develop the early PC studies and to prepare the initial control charts for the process.The removal of all assignable causes for variability and proper setting of theprocess average requires that operators and engineers work together to find the causes for variation.After the charts have been in place for some time and a large amount of data have been obtained from the process output, the PC study can be redone to obtain better estimates of the natural spread of the process during actual production. On-line or in-process methods such as SPC and off-line (Taguchi) methods are key elements in a total quality control programs.

SAMPLING ERRORS
1 1is important to understand that in sampling, two kinds of decision errors are always pos-

sible: see Figure 36-16. Suppose that the process is running perfectly, but the sample data indicate that something is wrong. You, the quality engineer, decide to stop the process to make adjustments.This is a type I error. Alternately, suppose that the process was not running perfectly and was making defective products. However, the sample data did not indicate that anything was wrong.You decided not to stop the process and set it right. This is a type I1 error. Both types of errors are possible insampling. For a given sample size, redncing the chance of one type of error will increase the chance uf the other. Increasing the sample.size or the frequency of sampling reduces the probability of errors but increases the cost of the inspection. It is common practice in control chart work to set the upper and lower control limits a t 3 standard deviations.This makes the probability of an alpha error very small and the probability of a beta error quite large. Many companies determine the size of the errors they are willing to accept according to the overall cost of making the errors plus the cost of inspection. I t for example, a type I1 error is very expensive in terms of product recalls or legal snits the company may be willing to make more type I errors, to sample more, or even to go to 100% inspection on very critical items to ensure that the company is not accepling defective materials and passing them on to the customer. The inspection should take place immediately after the processing. As mentioned earlier, i n any continuing manufacturing proces$ variations from establishedstandards are of two types: (1) assignable-carrse variations,such as those due to malfunctioning equipment or personnel, to defective material. or to a worn or broken tool; and (2) normal-clzunce variations,resulting from the inherent nonuniformities that exist in materials and in machine motions and operations. Deviations due to assignable causes may vary greatly. Their magnitude and occurrence are unpredictable and thus should he prevented. However, if the assignable causes of variation are removed from a given opera-

_\\
Based on the sample, you decide the process has changed Based on the sample, you decide the process has not
changed

The process is running. Changed It really has: Not changed (making defects)
No error T V PI ~ n error

Type II

No

FIGURE 36-16 When you look

Type i l o error) SaylnQthe process has changed when r has not changed Type I1 (perrbrl Say~ng the process has notchanged when it has changed

a t some of the output from a process and dec~de about the whole (I e ,the quality of the process), you can make two klnds of errors.

988

CHAPTER 36

Quality Engineering tion, the magnitude and frequency of the chance variations can be predicted with great accuracy.Thuqif one can be assured that only chance variations wdloccur, the quality of the product will be better known, and manufacturing can proceed with assurance about the results. By using statistical process control procedures one may detect the presence of an assignable-causevariation and after investigation to find the cause and remove it before it causes quality to become unacceptable.Also, the astute application of stabstical experimental design methods (Taguchi experiments) can help to identify some assignable causes To sum up, P C analysis and process improvements help to get the process "under control." Control charts help to keep the process on center (Xchart) with no increases in variability (R chart or uchart).

GAGE CAPABILITY
The instrument (gage) used to measure the process win also have some inherent precision and accuracy, often referred to as gage capability.In other words, the observed vanation in the component part being measured is really composed of the actual process variation plus the variation in the measured system; see Figure 36-17. The measurmg system will display: Bias: poor accuracy or aim Linearity, accuracy changes over the span of measurements Stabzlrty. accuracy changes over time Repeatability: loss of precision in the gage (variability) Reproduczbdity: variation due to different operators Measurements made by d~fferent operators wiU have different means and different variation about the mean when performing a measurement. Determining the capa. bility of the gage is called an R and R study In selecting a gage, the engineer tries to get the vanation in the gage (as measured by the standard deviation) to be less than 10%

Manual, published by the Automotive Industry Action Group.

DESIGN OF EXPERIMENTS (DOE) AND TAGUCHI METHODS

ity, as evidenced by the major emphasis (reemphasis) on SPC in American industry.Thi drive toward supenor quality has led to the introduction of Taguchi methods for i m provement in product$product design, and processes. Basically, SPC looks at process es and control, the latter loosely implying "improvement." DOE and Taguchi method however, span a much wider scope of functions andinclude the design aspects of prod

have been incorporated into all phases of production. The Taguchi methods incorporate the following general features:
1. Quality is defined in relation to the total loss to the consumer (or sonety) fromle than-perfect quality of the product.The methods include placing a monetary vahe quality loss, Anything less than perfect is waste

Cage capabil~ty (variauon) contributes to the total observed variation in the measurement of a part
FIGURE 3617

Variat~on~n the process

Var~at~on in the measurement system

36.5 SECTION

Determining Causes for Problems in Quality

989

2. In a competit~ve society, continuous quality improvement and cost reduction are necessary for staying in business. 3. Continuous quality mprovement requires contmuous reduction in the var~abll~ty of product performance characteristics with respect to their target values. 4. The quahty and cost of a manufactured product are determtned by the engineering des~gns of the product and its manufacturmg system. 5. The variabihty in product and process performance charactenstics can be reduced by exploitmg the nonlinear (interactive) effects of the process or product parameters on the performance characteristics. 6. Statistically planned (Taguchi) experiments can be used to d e t e r m e settings for processes and parameters that reduce the performance variation 7 . Design and improvement of products and processes can make them robust, or less sensitive to uncontrollable or difficult-to-control variations, called nolse by Taguchi.

f all of the input parameters, at In the Taguchi approach, specified combinations o vanow levels, that are believed to influence the quality characterist~cs being measured are used.These combinations should berun with the objective of selecting the best level for individual factors. For example,speed levels may be h~gh, normal, and low, and operators may be fast or slow. For aTaguchi approach, mater~al is often an Input variable specified at different levels: normal, homogeneous, and highly variable. If a material is not controllable,it is considered a noise factor. Desisned experiments andTaguch1methods can be used as alternative approaches to maklng a PC study.The Taguchi approach uses a truncated experimental design (called an orthogonal array) to determine which process inputs have the greatest effect on process vanability (i.e., precision) and which have the least. Those inputs that have the greatest influence are set at levels that minimize their effect on process variab~lity. As shown in Figure 36-18, factors A, B, C, and D all have an effect on process variabd~ t V. y By select~ng a hlgh level of A and low levels of 3 and C, the inherent var~ab~lity of the process can be reduced. Those factors that have little effect on process variability,
Zchart or Rchartto measure process var~abilrly Iprectsion) Sporadic departures
from standard
hard to detect

process at new

vald

Before Taguchi
0

T~me

After Taguchi

FIGURE 36-18 The use of Taguch~ methods can reduce the inherent process variability, as shown in the upper figure. Factors A, 6, C, and D versus process variable Vare shown in the lower figure.

Selecting h~gh and low values of input parameters changes the process output.

b k d 6 b g&uses for Problems in Quality

$2@

2, Id a EarnpetitiYedety, continuous quality improvement and cast reduction arenet+

ewuy faro r t a y * i p w 3, Contmuous quaJity impfovment requirk% continuous redaction in the variability vf produd performance charac'terhtics with respect to th9ir target value$,
4. The quality and cost of a mtmufactuted product are determined by the engineering designs of tlta producr and its mufwturing system. 5. B e variabilitJ' in product and process pdonnance characteristic$can be redncedby exploiting the fiohliiear (interactive) eBec&of the process or prodact parameters 4 x 1 the pperfoxmance oharnctaisties. & StaWcaUy piamed (Taw&) experiments can be used to determine settin6s for pracesses and parametem that reduce the performance variafian. 7, Design and improvementbEprodu& and proocesms can make them rubwt,or 1 % sensitive to uncontfoUable ot diffimlt-to-control variation&c&ed nohe by Taguchi.

In theThguchi qpraach,specified cambin;urIons o f all of the input parameters, at various le~els, that are believed to influence the quality chsacteristb beingmeziswed e + r e used.Thesecombmations should be run With the objective of sslFCting the best lcvd for tilrdividnal fadeis. Far exampie, speed leveh may be high, normal and low and operators may be fact or slow. For aTaguchi approsh, material is often an input variable specified at diffgrqt level%aomal,bomogeneeus, and highly variable. E a material is not ~ontrdlable, it is emsidemd a nohe factor. Dm@& expe~bmmts andTapehi methods can be useti asaltemlativeitgproaches t ~ f makiig a PC study. The lbguchi approach user a tnmc@ted experimental design (called an orr&a$onrrl army] to determine which process inputshave the greatest eflect m proof%$ variability Ci.e., pfeci~ioa)and wxch have the 1east.Those inputs that have the greatest influence.are ser at levels that mini&& their effect on prOse.5~ variability. As shown in Q u c e 36-18,factow A,B, C,and D all hav&an effect on process uariamity V. By s e l d ~ a ghigh level of A and low levels of B and C the inherent variabarty
Worn standard
hard to detect

Sporadic departures

Running the pmcess at new levels of input

@remeters

FWlRE 36-18 The use of Rguchr methods can reduce the inherent p r m s variability, as shwn in the uppef flgute. Factars A, B, C,and 0 vasus process varieble V are shown in the ~ O W W Rgufe.

lmp:bvh
t

T~ma

'4

LOW

nigh v'

LOW

H i g h
R

Low

Mgh II

Low

Hleh

I
A

I 1

f
6

Selecting high and low values of Input parameters changes the process output.

Quality Engineering like factor D, are used to adjust or recenter the process alm. In other words,Taguchl methods seek to minimlze or dampen the etfect of the causes of variability and thus to reduce the total process vanahllity.This is also the goal of the six sigma method, which is discussed next. The methods are. however,more than just mechanical procedures.They mfuse an overriding new philosophy into manufacturing management that basically makes qud~ty the prlmary issue in manufacturing The manulactunng world is rap~dlybecoming aware that the consumer 1s the ultimate judge of quality. Continuous quality improvement toward perfect quahty is the ultimate goal.FinaUy,it is recognized that the ultlmate quality and lowest cost of a manufactured product are determined to a large extent by the engmeered designs of the followmg.

1. The product 2. The manufacturing process technology and the sequences of processes 3. The manufacturing system (integration of the product and the process)
So, anew understanding of quality has emerged. Process vanability is not fixed.It

MOTOROLA'S SIX SIGMA

estimated from Tiid, or UIC, sample data), that a slgma 1s a standard deviatio sigma measures (sigma measures the repeatahilify or variability or lack of p a process). In essence, the six sigma concept calls foi the process to be improved where there are 12 standard dev~at~ons between USL and LSL. As the van

Here 1s an example. A foundry was having a problem with cores brea

TOTAL QUALITY CONTROL (TQC) The phrase total quality control (TQC) was first used by A. V. Feigenbaum Quulrty Control in May 1957. TQC means that all departments of a Cn participate in quality control (Table 36-1).Quality control is rherespons~bili
ers at every level In every department, all of whom have had quahty control

SLCTION 36.5

Determlmng Causes for Problems In Qual~ty

Mean

991

Four slgrna capability

. .-.

s;x ~ i g m a capability

FIGURE 36-19 To move to six sigma capabtllty from four slyma capabil~ty requires that the process capabil~iy (vanability) be greatly ~mproved (mreduced) The curves in these f~gures represent

histograms or curves fltted to h~stograms.

in then processes This is probably because many manutactul~ng (mechanical) engineers have not had any coursework in this area. Also, a change in company culture 1s often needed to give the respons~b~lity for qualrty to the worker, along wlth the author~ty to stop the process when something goes wrong.An attitude of defect prevention and a habit of constant improvement of quallty are f~ndamenlal to lean production, a system developed by Toyota. They have control) by extensive education accomplished TQC (they call it company-wide qnal~ty of the workers, giving them the analysis tools they need (control charts with cause-andeffect diagrams) to find and expose the problems Workers are encouraged to correct the~r own errors, and 100% lnspect~on (often done automat~cally) is the rulc. Passing defect~ve products on to the next process is not allowed.The goal 1s perfection. Qual~ty clrclcs,now popular in the Unlted States, are just one of the methods used by Japanese ~ndustnes to achleve perfection.

LINE STOP IN LEAN PRODUCTION A pair of yellow and red lights hang~ng abow the workers on the assembly line can be used to alert everyone in the area to the status of the processes Many companies use Andon boards, which hang above the aisles The number on the hoard reflects stations

4,

+
-I
I

dg -Irr

ia

ib

dr

3d

4r
1 I

&

e;r

& fwr stgma

des~gn specification width I-

. - Ir

-60

-b

la

4~ 4 r r

-I-

la

2v

a*

s
-1

6n

+S'X 848ma deslgn specIRCBtmn width

FIGURE 36-19 To move to SIX sigma capability from four sigma capab~lity requlres that the process rapabiltty (variability) be greatly improved (rr reduced). The CUNeS in these figures represent histogram or Cucves fitted to histograms.

in their prcrcesses.Thf9 is probably becausemany manufacturtng (mechanical)engineers have not had any courscwark in this area. Also, a ckange in company culture is oRen needed to give the respcsnsibility for quality to theworker, along with the authority tostop the proems when somethmg goes wrong. An attitude of defect prevention and a habit of mnstant impcovenrent of quatity are fuitdamental to l8an production, a system developed by Toyota. They have accamplished TQC @heycan it bornpany-wide quality control) by extensive education ofthewarkers, giving them the analysis tbols they need (control charm with cause-andeffect diagrams) to find and expose the problems. Workers are encouraged to correct their own mom, and 100% inspection (often done &utomatically)is the mle. Passing defective products on to the next pmcess ienot allowed.The goal is parfection. Quality cutlet, now papular in the United %tatates,arejust one of the methods usmi by Japanese industrim to achieve p6dection.

LINE STOP IN LEAN PRODUCTION

A pair of yellow and red lights hangiag above the workers on the a~sembly line can be used to alert everyone in the azea to the status of the processes Many companies use Andon boards, whieh hngabove the aisles T h e numbw on the board reflects station8

TQC Category
1. Orgalllration
2. Goals
r

TQC Concept
r Manufacturing engineenqg has respons~bility for qualrty-qual~ty circles

3 Basxc pnncrples

4 Facrlltat~ng concepts

$ Techtllques and aids

H a b ~of t lmptovementfor everyone everywhere m the manufacturing system * Perfecnon-zero d e f e d b n n t a program, a gaal Procearcon~rolbdefect preventron,not detection Easy fo-see,quality-quahty on display so cuvtomers can seeand mpeet ptocesse.--2asy to understand quality Tns~st on ompllance mth maintenance L ~ n stops e when something goes wrong Correct your own errOH 100% check m manufactunng and subassembly cab. QC department acts as facllltator Audrt suppliers Help in quality Improvement projects Tremlng workets, supervisors,suppklen Small lot m e s through rapid changeaver Hoqsekeepmg Less-than-full-capanty schedulmng Tolal preventive mamfenance. (TPM) 8-4-84 two-shift scheduling + Remove some h\renrory, i.xposeproblems,solve problems Defect prevention,poka-yokes for checklng 100% of p a m n = &for chsckrngfirsl and Iast ftem m lot (orn = 3, for largelots) Analysts tools Cause-and-effect dmgrarns ktbgrams, and run charts, check aheets Control oharts, (7, R, Scatter dbagrams Pareto charis Process flaw charts Taguchi and DOE methods

.
-

Sowce R~cherd Sohoenbrger. 1983 Japaness Mmueactunng Tzchm~uer:

on the 1ine.A worker can turn on a yellow light when assistance is needed, andnearby workers will move to assist the worker having a problem. The hne keeps moving, how. ever, until the product reaches the end of the station. Only then is a red light turned on if the problem cannot be solved quickly and the line needs to be stopped. When the problem is solved and everyone is ready to go again, the red light goes o f f and everyone starts back to work, all in synch Every worker should be given the authority to stop the production line to correct quality problems. In systems using poka-yoke or automation, devices may stop the line automabcally.The assembly line or manufacturing cell should be stopped immediately and started again only when the necessary corrections have been made.Although stop plng the tine takes time and money, it is advantageous in the long run. Problem can be found immed~ately, and the workers have more incentive to be attenfive because the) do not want to be responsible for stopping the line.

IMPLEMENTING QUALITY COMPANYWIDE

The basic idea of integration is to shdt functions that were formerly done in the s t d organization (called the production system) into the manufacturing system. Whal happens to the quality control department? The department serves as the facilitator m d

SECTION 36.5 Determining Causes for Problems In Quality

993

therefore acts to promote quality concepts throughout the plant. In addition, its staff educates and trains the workers in statistical and process control techniques and provides engineering assistance on visual and automatic inspection installations. Its most irnportant functions will be training the entire company in quality control. Another important function of the QC department will be to work with and audil the vendors.The vendor's quality must be raised to the level at which the buyer does not need to inspect illcoming material, parts, or subassemblies.The vendor simply becomes an extension of the buyer's plant. Ultimately, each vendor will deliver to the plant perfect materials that need no incoming inspection. Note that this means the acceptable quality level (AQL) of incoming material is 0%.Perfection is the goal. For many years this country has lived with the unwritten rule that 2 or 3% defective was about as good as you could get: better q d t y just costs too much. For the mass-production systems, this was true.,% achieve the kinds of quality thatToyota and many others have achieved, a company has to eliminate the job shop (a functional manufacturing system) and restructure the production system,integrating the quality function directly into the linkedcell manufacturing system, L-CMS. The quality control department also performs complex or technical inspections, total performance checks (often called end item inspection), chemical analysis. nondestructive testing, X-ray analysis, destructive tesrs, and tests of long duration.

M A K I N G QUALITY VISIBLE Visual display on quality should be placed throughout manufacturing facilities to make quality evident. These displays tell workers, managers, customers, and outside visitors what quality factors are being measured, what the current quality improvement projects are, and who has won awards for quality Examples of visible quality aresigns showing quality improvements, framed quality awards presented to or by the company, and displays of high-precision measuring equipment. These displays have several benefits. When customers visit a plant to inspect processes, they want to see measurable standards of quality. Highly visible indicators of quality such as control charts and displays should be posted in every department.Everyone is informed on current quality goals and the progress being made. Displays and quality awards are also an effective way to show the workforce that the company is serious about quality. SOURCE, SELF, A N D SUCCESSFUL CHECKS AND POKA-YOKES Many companies have developed an extensive QC program based on having many inspections. However, inspections can only find defects, not prevent them. Adding more inspectors and inspections merely uncovers more defects but does little to prevent them. Clearly, the least costly system is one that produces no defects. But is this possible? Yes, it can be accomplished through methods such as self and source inspection where quality control is in the hands of the operators. Many people do not believe that the goal of zero defects is possible to reach, but many companies have achieved this goal or have reduced their defect level to virtually zero using techniques snch as poka-yokes and source inspection.When you art: inspecting to find defects, the components are compared with standards and defective items are removed. Sampling inspection is used when 100% inspection is not an option, but this assumes that defects are inevitable and that more rigorous inspection can reduce defects. The truth is, to reduce the defects within a process, it must be recognized that defects are generated by the process itself and that most inspection techniques merely discover defects that already exist. To achieve zero defects, a lean manufacturing concept must be developed where all operators are responsible for quality You perform the following kinds of inspections:
Successive checks, where the next operator checks the work of the previous worker Self-checks,where the operator checks his own work before passing it on * Source inspection, where preventive action takes place at the error stage, to prevent errors from turning into defects

CHA~LR 36

Qual~ty Englneerlng

FIGURE 36-20 Source

lnspectlon lnvolves defect prevention; that IS, preventing errors from turnlng Into defects. (Source. Ach~ev~ng Zero Defects Mtstake-Proof~ng-The Zero Qual~ty Control System, O 1999, Productw~ty, Inc All rrghts reserved)

Defect provsntlun cyolle

Source inspection (seeFigure36-20)involves rethinking theinspection part of the manufacturingprocess Fust off,although it is necessary to have efficient inspection operations, they add httle value in the product. Even the most efficient inspection operations are merely efficient b r m s of non-value-adding activity. Inspection plays a passive role in manufactnring and cannot by itself reduce defects. Defects and errors are not the same thing in manufacturing. Errors can cause defects For example, not setling the oven temperature correctly can burn the roast (too high) or cause it to be undercooked (too low). Incorrectly loading the original into the fax machime [the error) results in sending blank pages (the defect). When you discover the effect,you make corrections,but you have already spent the money to produce the defect. Table 36-2 outlines some common exrors associated with manufacturing. So source inspection looks for errors before they become defects.These techniques either stop the system to make corrections or automatically compensate or correct for the error condition to prevent a defective item from being made. There are two ways to look at source inspections:vertically and horizontally.Vertical source inspections try to control upstream processes that can be the source or the cause of defects downstream.It is always n m a r y to examine source processes because they may have a much greater impact on quality than the processes being examined. Finding the source of a problemrequires asking "Why?" at every opportunity. Here is an example. Some steel bars were be~ng cylindrically ground. After grinding, about 10% of the bars warped (bent IangttudinalIy)and were rejected.The grinding precess was studled extensively, and no sure solution was found. Looking upstream, a problem wtth the heat-treating process that preceded cylindrical grinding was detected. About 10% o f the bars were not getting a complere,uniform heating prior tb quench. Agkmg "Why?", it was found that these bars were always ly~ng close to the door of the oven, and it was found that the door was not properly sealed,which resulted in a temperature gradlent inside,Quenching of the ba~s tnduced a residual stress that was released by the grindug

Omissions in processing step% Enon insetting up the lob in the mach~ne Omission m the assembly process (mlss~ng parts)
Inciuslon of iacotrenpart S=e errors due to rvrong measutitment Errors due to adjustments Errms in cutting-tool geometzy.or eutbng-1001 settlng Errorsm prwesmgcomponents (heat treamg)

SECTION 36.5 Determining Causesfor Problems in Quality

995

and caused the warping. Asking "Why?" at every opportunity uncovered the source of the problem. Horizontal source inspections detect defect sources wlthm the pracesses and then mtroduce corrections to keep from turning errors into defects. In metalcutting and forming, this is commonly called adaptive control (for preventing defects) or in-process quality control. One of the best ways to prevent errors from occurring 1s through the use of poka-yokes. Poka-yoke is a Japanese word meanlng "defect prevenhon." Poka-yoke devices and procedures are often devlsed mainly for preserving the safety of operations. The Idea is to develop a method, mechanism, or device that xv11l prevent the defect from occurring rather than to find the defect after it has occurred. Poka-yokes can be attached to machines to automatically check the products or parts in a process. Poka-yokes are source inspections that are usually attributes mspectlons The product~on of a had part is prevented by the devlce. Some devices may automatically shut down a machme if a defect is produced,preventing the production of an add~tional defective part.The pokainspection to guard against unavoidable human error. yoke system uses 1130% Modern cars are equipped with many poka-yoke devices: you can't turn off the motor unless it is in park, you can't open a door while the car is moving, the headlights come on wlth the windshield wipers-all are devices to prevent you from making a mistake. Such devices workvery weU when physical detection is needed, but many Items can be checked only by sensory detection methods, such as the surface finish on a hearing race or the flatness of a glass plate. Variat~ons m nonvlslble cond~tlons (ax pressure, fluid velocity, temperature, electrical voltage, etc.) requlre detectlon dev~ces where cr~tical conditions are readily visible. For such problems a system of self-checks and successive checks can also be used.

TEAMS (AKA QUALITY CIRCLES) Several popular programs are built upon the concept of participative management, such as quality circles, improvement teams, and task groups.These programs have been very successful in many companies but have failed miserably in othersThe difference is often due to the way management implemented the program. Programs must be integrated and managed within the context of a lean manufacturing system design strategy. For example, asking an employee for a suggestion that management does not use (or cannot explain why it does not use) defeats a suggestion system. Management must learn to trust the employees' ideas and decisions and move the decision making to the factory floor. A quality circle is usually a group of employees within the same department or factory floor area. Meetings are held to work on problems. An organization structure is usually composed of members, a team leader, a facilitator, a manufacturing engineer, and a steering committee. Quality circles usually have the following main objectives: provide all workers with a chance to demonstrate their ideas; raise employee morale; and encourage and develop workers' knowledge, quality control techniques, and problem-solving methods. They also unify companywide QC activities, clarify managerial policies, and develop leadership and supervisory capabilities. Quality circles have been implemented in U.S. companies with limited success when they are not part of a lean manufacturing strategy. It is possible for quality circles to work in the United States, but they must be encouraged and supported by management. Everyone must be taught the importance and benefits of integrated quality control. SUPERIOR QUALITY IN MANUFACTURINC/ASSEMBLY CELLS In lean manufacturing subassembly and manufactur~ng cells, the cells are des~gned tor a "make one-check one-move one on" (MO-CO-MOO) strategy. The part receives successive checks after each processing or assembly step. For successrve checks to be successful, several rules should be followed. All the possible variables and attributes should not be measured, because t h ~ would s eventually lead to errors and confusion in the inspection process. The part should be analyzed so that only one or two points are

inspected after each step in the proce$s.This is the heart of MO-CO-MOO. Only the most important elements just produced are inspected. Another Important rule 1s that the immedlate feedback of a defect leads to Immediate action Slnce the parts are produced in an Integrated manufacturing system, this will be very effectwein preventing the product~on of more defective parts. Suppose the cell has only one or two workers and they are not in a position to dlrectly check each other's work after each slep. Here is where the decouplers can play an actlve role by providing automatic successive chccking of the parts' critlcal features before proceedlng to the next step. Only perfect parts are pulled from one process to the next through the decoupler In assembly lines, a worker may inspect each part immedlately after producingit. Thls is called self-check~ng. There IS an lmmedlate feedback to the worker on quality. However, it would be difficult for many workers not to allow a certain degree of bias to creep into their inspection,whether they were aware of it or nol,since they are Inspecting their own work. Wlthin cells operated by multiple workers, the operator of the downstream station or process can Inspect the parts produced by the upstream operator. If there 1s a problem with the parts, the defective item is immediately passed back to the worker at the previous station Thcre the defect is verlfied and the problem corrected. Act~on is immedlately taken to prevent any more defective parts. While this is going on, the llne is shut down.

The des~gner of the product must have quality in mind during the quallty design phase, seek~ng the least costly means to ensure the quallty of the des~red functional charac: teristics. Major factors that can be handled during the early stages of the product desigd cycle Include temperature, humidity, power variations, and deterioration of materials: and tools. Compensation for these factors is difficult or even impossible to implemend after the product 1s m production. The distinction between superios- and poor-quah products can be seen in their variahil~ty in the face of Internal and external causes.Th is where Taguchi parameter deslgn methods can be unportant. The secret to successful process control is puthng the control of quality in th hand? of the workers. Many companies in this country are currently engaged in SQ (statistical quality control), but they are still inspectmg to find defects.The number of tects will not be reduced merely by mak~ng the inspection stage better 01 faster or tomated You aresimply more efficient at discovertng defec(s.The trick is to inspect prevent defects How can this be done? Here are the baslc ideas. Use 7ource inspecti techniques that control quality at the stage where defects originate. Use 100% insp non with immedlate feedback rather than sampling. Make every worker an inspect Mintmize the time it takes to carry out corrective action Remember that people human and not infalhble. Methods and devlces can be developed to prevent them fr making errors Can you think of such a device? Does your car have a procedure t prevents you from locking the ignition keys inside the car? That is a poka-yoke. Do not s~mply rely on inspection to control quality. Sorting to find defects rnspectlon is bad because you already paid to produce those defects. Process improvement should drive toward defect prevention. In older to achie the highest levels in quality, you have to implement a manufacturing system des~ (MSD) that has the highest objective (zero defects) built Into (mtegratedmto) theM Concentrate on making processes efficient,not simply on making the operators operations more proficient. Continuous improvement requires that you redeslgn manufacturing system continuously, reduwng the time required for products to rn through the system (i.e., the throughput time).Thls approach seems to be the Arne stumbling block. Industrial engineers can do operations Improvement work such as ing a better machine or improving the ergonomics of a task. However, they need to more systems improvement work.Too often fancy, complex, computerized soluhons devised to solve complex manufacturing process problems Why not simpllfy the rn ufacturing system so that the need for complex solutions disappears?

ls Key Words
accuracy control chart control lunlts fishbone diagram hlstogrem nom~nal parent population poka-yoke precision process capability (PC) quality control (QC) range (R) sample mean sample size jn) self-checking source inspection specification limit standard deviation (u)

(z)

statlsticirl process control (SPC) Taguch methods total quality control (TQC) vanabmhtv

M Review Questions
1 Define a process capabmlity study in terms of accuracy o r prewslon. 2. What does the nature of fheprocess refer to? 3. Suppose you have a'plstol-shwtmg" process that 1s accurate and precise What m ~ g h the t target look like if, occasionally whlle shooting, a sharp gust of wlnd blew left to r~ght?

15. Which type of error canlead to legal actlon from the consumer for a defectrve product that caused bod~ly inluq7 16. Define and explain the difference between each of these a. u and wT b uandF c- U Z , u", and cz
~.

6.
7.
8.

9.

10. 11.

12 13

of a given process, such as rolbng,extmdmg, or t m n g ? What areTaguchi or factorial expenmen@,and how might they be used to do a process capabllity study? How does the Taguchi approach differ from the standard expenmental method outlmed In thls chapter? Why a r e Taguch~ experiments so important compared to classical DOE-type experiments? Here are some common, everyday processes wxth which you are familiar What variable or aspect to the process might domlnate t h e process in terms of quahty,not output? a Baking a cake (from scratch) or grilling a steak b Mow~ng the lawn c. Washing dishes In a dishwasher Explain why the diameter measurements for holes produced by the process of drilling could have a skewed rather than a normal distnbut~on. Name some common manufactured ttems that may receive the following. a. 100% lnspechon b . No final inspection c Some fmal mspectlon, that ls,sampImg VJhat are common reawns for sampllnglnspectlon rather than 100% inspection? R11 in t h ~ table s wlth one of the four Eollowingstatements: no error-the process is good; no error-the process is bad; type I or alpha error; type I1 vr beta error. In reality, if we looked a t everything the process made, we would know that had: not

. . ~

18. The designer of a component usually sets the non~inal and tolerance values when designing the part. How do these decisions affect the decisions of the manufacturing engineer

19. What are someof the alternativesavailable to you when you have the situation where 6u> USL - LSL? 20. CPi, is also a process capability index. How does it differ from C,? 21. In a sigma chart, are values for the samples normally distributed about F? Why or why not? 22. What is an assignable cause-,and how is it diiferent from a chance cause? 23. Why is the range used to measure variability when the standard deviation is really a better statistic? 2 4 . How is the standard deviation of a distribution of sample means related to the standard deviation of the distribution from which the samples were drawn? 25. In the last two decades, the quality of automobiles has significantly improved.What do you think is the main cause for this marked quality improvement? 26. Figure 36-12 shows a bimodal check sheet indicating that the two operators performing an assembly task (in a cell) do the !,>ur.!c~~rti~tm:~~J hctc ' lobs :el d ~ f m c r r>tc\ ~~~~ \4'It;tt r w*,t~lJ 27. ('onlr,~lrlt.~rt\usc upp6r:mnJ l u s c r c ~ ~ n f limttit\. ~ o l I c . , i:CL fh; same as a USL? 28 In F~gure 36-14, what are the USL and the LSL and why dre they not shown on the charts? 29. What are four malor branches (fishhones) on a cause-andeffect d1agram7 30. How does vanation m the measunng devtce (mnstrument) affect the measurements obtalned on a com~onent? 31. Explain what happed to Improve the process in Figure 36-18 32 in Table 36.1, explain these a. MO-CO-MOO b. 8-4-8-4 scheddmg c n = 2 inspection d. Pareto chart 33. What is a quality cucle,and how might you apply thls concept to your college hfe?

(MfE)?

1
1

The sample changed suggested that the process had. not ch.dng& 14 Explaln why, when we sample, we cannot avoid makmg type I and type I1 errors"

I I Problems
1 For the items listed in the folIowing chart,obtain a quant?ty of

48 Measure [he lnd~cated character~stlcs and determ~ne the process mean and standard d e v ~ a t ~ o Use n a sample slze of 4, so that 12 samplo are produced Item Flat washer Paper clip Coin (penny, dime) Your choice Characteristic(s) You Can Measure Weight, width, diameter of hole, outside diameter Length, diameter oI wire Diameter, thickness at point, weight Your choice

2. Perform a process capability study to determine the PC of the process that makes M&M candy You wiU need to decide what characteristics you want to measure (weight, diameter, thickness, e t ~ . )how , you will measure it (use rule of 10).and what kind of M&Ms you want to inspect (how many bags of M&Ms

you wish to sample).Take sample size of 4 (n = 4). %take a. histogram of the individual data and estimate p and uas out.. lined in the chapter. If you decide to measure the weight char;. acteristics, you can check your estimate of p by weighing allthe M&Ms together and dividing by the total number of M & W 3. For the data given in Figure 36-4,compute the mean andstan- , dard deviation for the histogram and then C, and CPk,makng any assumptions needed to perform the calculations. 4. For the data given in Figure 36-7, compute C , and C & . 5. Calculate and and the control limits for the X and R control charts shown in Figure 36-A.The sample mean, F,and range,R,for the first few subgroups and the data for each sample are given in the bottom of the figure. There are 25 samples of size 4.Therefore k = 25, n = 4. Complete the bottom part of the table and then compute thecontrol limits for both charts. Construct the charts plotting X and R as solid lines and control limits as dashed lines. as shown in Figure 36-14. The first four data points have been plotted and the points

,.

FIGURE 36-A

1000

C B A36 ~ Quality Englneerlng

ou have tecently beell hlred by the Pippen Company as the 6 slgma Black Belt Your new boss, Gebby Sorenson, has one of the lead~ng textbooks m manufacturing engineer~ng (NOT DeGarmo) and she shows you this problem After read~ng the dlscusslon and studylng the figure, Gabby asked you the following questions 1. What k~nd of control chart is this? 2 What do you estlmate the standard devratlon (at) of the parent populat~on to be, assuming that the control l~mits on this chart are based on three s~gma? labeled-mean*the mean 3 is the dashed line the figure s equal to or the average? (The mean is X and 1 0.00017.) What 1 s the llne really called? 4 From the infomat~on given ~n the dlscusslon and In the f~gure, defermine the values of C , and C, 5 What IS the most glaring error in the figure (Hint: somethlng one never does wlth control charts) and why e it

The workplece shown In F~gure CS 36 IS made of gray cast Iron and is bored lo the tolerances rndicated (5.512515.51 15 in.).These parts were bored on a chucking machine. Each of the 19 pointsplotted on the vert~cal axls of the control chart represents the average of bore diameter measurements made on tour parts (sample slze). The hor~zontal broken llnes at t0.0005 and 0.0005 repres*nt upper and lower specified limits, respectively. The sol1dline X= 0.00017 In. 1 s the estimate of the process capabil~ty based on a study of several samples bored on the machlne. The upper and lower control llm~ts are then calculated from X .We note that samples 4-9 show a definrte trend toward undersized bored holes. If the operation had k e n continued without any changes, the success~ve bored holes very lhkely would have been Out 0f t0kmnCe To avo~d thls sltuatlon (out of control), the boring tools were reset toward the upper control llmlt before parts in sample 10 and the rest were bored. (Source ASM lntefnat~@nal.)

4 measuremenZ?

Tool was reset

-

--

-----

Sample number
Figure CS 36

2 4

6 8 10

12

14

16

18

-I location pnnciplc, 680 3jaw Jacob, ~ h " ~ k . 6 4 3 3~phase equll1briuln.77 5-:,xis mschlnlng mnlcr. 723 Ahlation.516 Abrasion,370.561 Ablaaivc j e t machining.(AJM).787 ~brabivc 1nachine.524 Abrasive machining proccirer. 757 Ahraslvc mochining. 17,756,765 Abrasive media.942 Abraslvc warerlel culling, (AWC). 784 Abrasircs,757 Absolute vddrcss systcm.715 Accuincy,219, 677.970 Acc,ylmc,%l A c ~pickl1ng.948 l Acouiticemiri~on munitoring,255 Acourf~c holography 257 Acrylic enamel. 948 Acrylics. 91 6 Adrlendum.Xil9 Adhcsion,370 Adhcsivc bonding. IY.352,915 Adherive joinrllg design, 920 Adheswei 171,347,915.920 Ad~usv~blc resmer.653 Advanced lihci-rcinlurced composller. 186 Aduanccd iligh~strmglh nccl, 129 AFS.(American Foulidry Sociely).2XX Age hrrde"ing.93.101 Aging.94,96 Airbend dir.434 Airhardening u ~ usteel, l 135 Airless spa) painting.Y49 Alriet molding,295 AISI ( ~ m e i i c a n lron and stcel Initi~utr), 12) AISI-SAF. ,dcnliflcation $yrlrm. 125 AISI~SAEsrandard steel designation$,126 Alclad. 148 Alkalille clcalling. 946 Alkyds. 948 Allotropic. 60 Allrrwa,,ce,220 ~ l l o raci, y 124. 125 Alloys. 6X, 139 Alpha brass, 142 Alpha-ierritc. 80 Al~cmdling Currenl.(AC). 850 Alumina,(Ai20,), 178. 181,757,943 Aluilllnum ailoyr.283 Alurnlzluln casling alloya. 1Sf) Aiu,n,n"lllcastingall,,yr,iablc, I51 Aluminum cilrusro~~. 146 Aluminum foam, IS2 Alum,num lilhillm.lil")s,150 Alulnillurn oridc,(AI,O,) 572.759 Alum1num.144.313 Alum!num-hronzc alloy, 144 Aluminum~ulicun alio)s,Y05 Amcrlchn Welding Soclel) (AWS),RlX Amorphrlur mctnl. 463 Alnoiphour riructurc59,177 Ampcrc.214 Anarnih~ca. 9 16 Andon,9YI Anlri>tmpicproperty. 66 Anisofropy376 Anneaihng. 90.376 Anodi7i"g 955 Anlioxidunls. 17u *-parent denriLy,463

Apron. 607,609 Arall, 183 Aramvid. 185 Arc cutting, 865 Arc furnacc,326 Arc welding powcr sources, 864 Arc welding. 849 Arc-welding electrode designation,852 Anfhmeticroughncsi auerage.934 Aititiclal gar atmosphere, 113 Aspect ratlo. 186 A~~emb jig.695 ly AsIembly iine(s).13 Assembly yieId,496 Assembly. 343 Atomic bonding. 57 Atomic number, 57 Atomic rfrucfure.57 Atnmi,:alion.461 Affribufe.214 Altririon.758 Auiforming, 109 Ausfemper, 108 Aurlrmpsred ductile iron,85 Ausienitr.8fI.81.90 Ausfinef~crVainlcrs steel, 133 Autocatalytic plating. 95h Auiogenotis wcld, 893 Aulrlmated guided vehiclc. (AGV). 20.708 Aufomslic hill forging, 397 Aufomaficspraying,Y49 Aulornatlc storage and ictriervl system, (ASIRS).2ii Automatically programmed louls. (APT),707 Automafion.20.702 Auluqurnchin: 110 Bubhi11.159 Back rake anpla,539.577 Bucklash,980 Backward impact cxrrurion.41o Bnlnlle,98 Bake hardenable steel, 129 Ball gild arra).498 Ball lead screw716 Band-hling nlachine.752 Bundsaw744.748 Bar drawing,407 B~~rcilder,431 Barrel hu1n1rhing.942 Ba,rclfinrhing.Y41 Barrel lumblmg.941 Base circle. 809 Basc, 645 Llaslc inapecflonmerhods,227 Batch rurnocc, 112 Bcad welds. 829 Bed. 607 BsdMype milling machiae, 671 Brll furnilce, i l Z Bcif susdlng,Y45 Bcnci>type lathc,6lO Bending blank size.433 Bending opcrafionl422 Bending. 363.392.430 Beryllium, IS9 Bevel gc;ir.812 Bcvel orolrack,r.230.241 Bickfnrd drill. 633 Bilateral rolerunco.222 R1licf,382,401 Bismulh. 144 Blackening.953 Blank coating, 919

Blank lhc~ldc~ iorce.441 Bianklng,423,425,812.820 Blast cleaniilg.94Il Blendiing.461.464 Blind riser.274 Blind nveiing.415 Blockin~,392 Bioom.3K2 Blow mold!ilg,33h. 348 Blou~oul.405 BMC (hulk~molii1ngcompound).354 Body crnlcrcd cubic.60 Bolsler plsLc.426
n o n i l 7R7 -"

Borazao. (CBN).758 Bor!ngTrcpannmg Arsocrat~on, (BTA). drill, 636 Boring. 528.599. 602 BoLL,,m tap.798 Bottoming dlr.414 Boule.493 B O X furnacc.112 Box jig, 686 Bralding.358 Braising, 19 Brass. 141.283 Bnac wcidlng, 90Y Brazed joilnf dcsign.9113 U~;mng msierials.904 Biazingmclhudr. 906 B~azmg. 836.9li1.929 Breaklng strenglh.3.? Bnne, 105 Brinell hardness tcst.37 Rrittie fiscturc.582 Biitflmrsr,33 Broaching. 17,524,531,736,815 Bronre, 142.283 Buffi"g.945 Buill&upedge. (BUE),542.557.572.606 Bulglng.443 Bulk lluw363 Bulk forlning pnwrraea. 381

Bulk-moldiiigo,rnpound,(BMC),3i4
Bull whcci, 732 Buin~end jleid.496 Rurnishing.416 Burn~.,ll opcr;,flon, 468 Buir.959 Butllcdd,498 Bull weld.834 1null~oinl c1earunco.902 Bull-lnpjoinf.903 ButtGwcldcdpipe. 451 CADICAM.7 Caicndai. 347 Cniendunng. 342 Cam driven prcss,452 Casdela,214 Callilcd machining mofine.717 Cannmg. 470 Capillary acfiao. 353 Cnrblde lnlcrr cutting fool. 6 9 Virhldes.568 Carhun cquiuulenr,82,84 Cnrbnn-caihoi, c0,"poiiier. I88 C2,,bonifrld,ng, 111 Carhottom box furnace. 112 Ca~huriiin: flame. W4 Curburir811g. Ill Carriage. hOl.hO9 Cartridge hrar,, 142 Cast cobalr all,,vs.i07

1002

Index
Clily lnafnial~ 178 CidyY.148 ? I t . * " r m , 4% Cieanlng wiJlngs.329 Clcanrng 310 4K5 Clierann Et.221l Clinchiilg.501 Clc~mi & e l fuam plaw.Y41 C l w d i l i r it~mng,?91 CMC (cerdmlc-mulnh wmpos~Ie9).Ivo CNCgearm~penwn maehme 8 2 . 2 CNC LlthC.m5 C ~ c m s d j ~ thtz1.7 ne CRC lornmg ceOfer,b7KTn F02l~~ct.517 894 CoulescenCe.XZ7 (li~tcd uboulves77~ Cl),xt~~lca~hrde culung t o o l ~ ~ i u Cdalid ~lemrodc.851 Ca~ted ~ 1 ~ 5 6 5 C o a u a 948 Cdhslt .tiluyr.YOS

Computer lomog~aphy,256

Con' mandrel. 621 Cot~lnr~n pmsx4iffl CDnlelty 2a1 C~(rsnl~4s$ion prwemw. R 2 7 rmstantan 143

Cemcnls 179 Ccntrr cqre dillL634 Ccnwr drlll 638 Center h ~ a d 230 Cenler lvp. iyllndrlcal Qnnlilng 774 Centerlesr px1dmg7l5 Ccnledesr rhrsatignnddng805 Cenlcr-ty~endteed thread gudw 8 0 5 Cenln1ug.d bnrrsl tumhlm&Y42 CenInlu~41 cmlng,322,353 Centrigus anlnf"@Tl castmcg 323 mm" pmcWL471 rCerRm,c ir(Lrn 2 ' 7

Cobalt. LSY.568
Meffluentof tnc,*rm,570 Cohorzn2y. Y S Coh~~vcncss,ZX7 Cml-malcd shcct metal Y i Y Cniaing Q0.434 Culmq$415.473 Cold diaumg,lM7 M l d emwwn,911 Cold fbrming,409 C d 4 heddmn~409 Cald molding 337 Cold mer.274 Cold mU-fomng proeem436 Cold mUtng3Xrl Cold saw 7 U Cald Sl~uh270 Cold wclding,%<1 Cold wotkmg. M, 373 Cold-boxmcthod 3(m Cold-chnmbeidm m=tlng.3lX Cold.mIl fur~urng 818 raid IWI sne~i13s C'di"*ilbti,Iy,ZB7 r"lll,p*'"g tnp,799 CaIW shack643 cbllcl.622 . Colonng ~ p n t s170 C~,bnngmetuls, 9S3 Colmn,bBi Column-and knee m~ll!ng nrleh,m,67u CoIumua~ronc.26Y Comblmr~on dnIl.63Y Coniblnatron fourlaw rumbrnanon chuck.621 Combmatian cet.230 Conlpactabii~ty 2M Cnmpudod $rapha& cart imn 4 Cnmpurtm:, 4 M Compilron squmce 465 Cornpaa1on.461 ~ o n l r d(CWUC).U69 Company widequal~iy Compmilcmatettslu 183 ilompmiLeL 11%

Contour turnlng 599 Contounngcontiol.714 C*bnlounns 712

Cermeli 179.572 Ceral8eSear,Xl2 Chnnnei jihW6 Chnplets.301 a . 832 CUrpyirnpmfl L Chnrpy les1.42 t:Utl<r*dn tbn;i($hue,, 1 'U i I

Cooling rate 2117 r n -e n l Cooidinaro m e ~ ~ ~ i r m Com.ZGS. 284

(CMM). 240

Check,il@.
Chcmjcal~~hiiaion,5i6 Chsmi~lhnrl~shLi7 Chcplical Man'dng, 4% Cbcmieal cleaning. 946 ChcmknI pwerdon nr;iiings. 953 Chemical failurc.5Xz Chcmicalflax cuftin&X# Chcmimln,aehining (flHM),485 Chemical mi~chining d~:Ei-~:n rXscarir,4Y11 Chcplicdl m ~ m a r h i l ~ l n g p r a c ~ ~ 4 ~ 5 Chemioalmduclion. 4b2

Copper. ldfl Cappr-hrrylllum alloy. 1411 Coie him2h5 Cbre maklny, 7YX cumprint 26r.iiit Colod. 77 C~IC-o~l procss,3(X) Cws.to.facc bond 9211 CMOJOJO~ rcslsBne 140

Chill mne, 2 ' s Chiil&.cx(rmul.2iS Chilh/nternili.l7S Chip di&,,<n;iI. fim Chip groovr;>Tl Chip ahooles.PI? Chip thinkralio.542 Chipless tapping. 806 Chipping.361 Chips 19 Chise1snd.m Chokc,27i Chramltt: wnvrrsbn mntingsYS1 Chuakingieamsr.651 Chrutinav's rule.mX. Cirnular blade sw.75i saw 76 Cir~uinr Chddlop, Y48.958 Uampin.q,$77.hXZ Clampwr ] n g6% Uappar box 732 Classes of ILL% 272

C0mp0~11~~,Eubn=~han.3~1
Compunddia.lia Compound rcu M17 Componndr 949 Campres?ab,lny 4h* Comprei~on hc"dsn&435 Compiesrton moldmg 336 356 Cvmprwl'on test 36
CompiesilM stram 2Y

ailv mnleni w~

ComprewVC itrm&th289 Compremvc strcm.29 Cumpttleinriiod deslghCAD 23,707 Cnmpular mdtd manuiactwmug CAM 23 Compubr atdsd lestmg ilnd mspacuon, CATI.ZI Cnmpubr mntrai h w e r 3W

Index
Ugstalluq 165

1003

&bre aemn mirrd~ ( ~ 5 7 3' i ,m

Cabre d r ~ ~ l ~ 63 rn, e t q ~ ~ * & , 4 4 2 e"ple9r3ZS Cuprndckd,M Curin&915 Coriantsyeie a13 Cut and-peel method,485

Direct ~ x * ~ o E ~ o " , ~ V Diredl fuel-fired rumaee.326 Direct numerical control. (DNC),707 Dired-arc Turnace. 326 Diieal~curien!electrode-negative, (DCEN). 850 die casiin&318 D~nccf~injection Direaiion ofshear,sn

Ctnbff Wrmfbnn 4% eutaftS&6tl2,937 Ciirlerg%~&r 7% CUttarMf%et, 7n

Cuitingfiu>db591,649 1 7 t . 8 m C~tOngieaoe,r?3.Y43.6[U.6h.9 E-iltttngrw"i vs mmpemNie,5% c U n n ~ ~ p W . s Z WO~b2s.?31,9S7 S, W~sp~warq&t,7B6 C"U,,!gstdfU~,53s.5~ cuttin~fme,3ZY.4a1e,83U133 Cuttw@Ume,t ~ p m g8 ,0 1

w f m sWI ang1cI.sn

CaUii t w l ~ e t r y . 5 T 7
W t m g t ~ phol&r,6l6 l Cuttmg f))ol matenai spleci~o11,561 C u m toot materb1SX41 C m m g twl mator~aisp19nart1es.56r) CulUt)E.ti%%% l$.5766S9S Curtw761 c ~ m l = w , r i a Cyi%niVrcd mathlning apeations 332 cyimaticity.241 CmhraMke methbd,49? Darmwne,capaww,36,81 Data-w2+senauntt. (DPU11712

DW tdepth.ot*ut aw),576
$ 2 7 4 Detnmmg pao:ss.s6b Daca00%4l(l Dsde"d"m.639 Deep iVswm& 440 newhalp ddiis, D~&~U60.633

Wad ~

DW ~ e e ~ e e , h Z O

IMmahon pro~d~ses,36%
Dagli*yficrlm. 121

Dagassmg270
Detaysd t s & a d b , w ,

a4

Deiiatm.fm
Da4xldyihw,121

Depth of cut (DK?,f25.6W. Ori1.7JI

Deptkni rmcrsioq(D0I) 861 Dept~.oi~th~(DCrJ,576 Depth 1 ~ h e t raho, w 63s

Dts~pnformanuPaetming,(bFM),ll, 7 8 7 h a f w t ~ z l
Dedp dsqmdmenls(D0E). 988 D e s i p efPMprmdaeth476 Desrgnp=mlW M ~ e t ~t v e ~ ~ & 2 4 Demtdcaum$4@ h a f i c a t r o n , 142 bra1 mdimter.244 D&amoUr.ggt6s D ~ w ~ l p t o h . Ro9 ( ~ , D~amands~~l58 Die csa~g,313,316,& Dleskoe,426

DlG'lZ?
Dm-cmtmtdl pegpcmes, 320, W1 Dte~c~ahn&*11nys.155 Die.fal,!g mash",*. 732

DdfuslanhORd*~&%

hffusnw wetdTx~$(DRW),8s
Drglld mMmMm,3!33 D~mm"nsla e m $ 6 %

D,&mgrn Dlp bmeng,XJ8

DIP s~ldclmnb Pt3

Dlppcn& 3G6.949

DirectionalsoIidi6catian,a66 Discontinuous chip.542 D~smete hfener~Y2S Disk saw746 Disk-grinding machinrr.778 Disk-typo frling maen,nil.752 Dislocation, 64 Dispersion hardetling.YZ95 Dispersion-rtiength&d malmak.IR4 Dirlorlrun. 106.277.537 Disirihuted numerical eontroi, (DNC).70? Do8 plale, 620 Doping.493 Dormant llaw.257 Double frame diop h~mrumru~~3W) Dosbleactirm press441 Double-cut hroach, 740 Daublc-eut file.751 Double-howingpiancr,735 D m gnndmg.757 Down milling.662 Draft,265,277 Drag2h5.284 Draw beads 442 Dmw bench.407 Iliiiw bending.435 Draw hlo.~k.408 Draw ralro.441 Draw-cut rhuper, 733 Dm,uin cdiet. 622 Drawing oprraiiuns.422 Drawing 342,410 Drcsingrtick.766 Drcsing. 7 6 DrifL22S Drill hushing, 684 ~ r i ipres. l 64s Drill selection. 642 Drill shes, 633 D~illmol hoCicr,643 Drilling axial thrust, 646 Drilling horsep~wer.631 Driilingproblcmg.635 Drillikgpr-sse~ compared, 638 Dtilling. 17,524.510.599,MN,628 Diivr cmnk rpm,73l Dmsr,26Y Dry sand p,,urZ,Zn9 Drying, 951 Drysatad mald.293 Dual phase steel. 129 Dual-inline package (DlPj.497 Dubbed drill, 342 Dubbiog, 632 D U C ~ cast I ~ EIron, 85 Duotiie IU brittle tranriliun temperature.48 D~stillly,33,374 Durnp-bor,296 Dump-cMe hox,2$$ Duplrf stainless steel.133 Dupiicaling attachment. 605 Dup1ic"t"l. 672 Duraiumm. 148 Durometcr.40 Dury eyclr.51 I Dynamic properties,42 Ecccntticpress drive mechanirm.453 ECG metal removal ntr,509 Economiccutting rpeni.589 Econmnin of mncluns.588 Eddy-cuncnt tesfiig,Z53 E d ~ chipping. z St2 Edge damp.695

Edge dislocation. 61 Edging.iY2 E i i s e l placcsr.2YK Ejector drill.636 Ejector pin. 337 Eioctic delo1maooo.62 Elarliciim~f.32 Eiastlc springback.375 Elnstomcr%173,915 Eln~lumcrs. rvblc uf pmperurs 176 Eic~lric:ileonduetiviiy,53.68.140 Eleclrical dischnr~e machine [EDM).S06,510 Elcfrical inducfim heating. 113 E i c c t r i c a i 1 c r i a t i v i r i ; S 3 Elcdio~hrmicnl deburring.508 E!&c%hemik&! @indinp.(ECG). $dl? Eleilroel~ernical ~~achining. (ECM).StM Eiecrrochc~nical rnicrumachinin:. (EMM),506 EIe~trocoati~~g.YSli Elecli~di. R44 Electrii4cposition.W mi'hininp.934 El~tdiscilutge Ekcftofommg.451.YS5 Eluctroless co~npmilr p1;ltin.g. 957 Etrrtroles~ plat$"&,456 Elrdroliti~ tough-pltch (ETP). q p w . 141 Eieciclrcin hole. 69 Elccrmn,56 Eleclmnhenm hardening.1lll Eleetmn-hmm machinine (EBM1.515 Elccmm-bcam welding, (EBW). 841 EiccLr0plaling,Y53 Eicctrop~~llshing. 946 122 E l e ~ l r o ~remitiny ly Elecliosiag welding,(ESW),XYIl Eiecftobtattc chuck.096 EledrOCldliC dcp~rilicm.9511 Elsdmrlslii~prayfinirhing.949 ElrctruSLieam driiii0g.5116 Elevator furnace. 1I3 Elongation.2Y tlusiomer.335 Emh'nring.445 Emer*, 757 Emulsilicsliun. 94h End mill, b N End miiiin&53<l,650,661 Endurance lirnit.44

Entcrprike.i Environmental rondderutiong. 201 Eporio.916 Equiuued zonc.269 Equilibrium phase diagram,71 E~ch rearr.485 ELchin&485 Eutesiicstructure. 78 Eutritic-cumpusition silo), 267 Eutouliod neal, Yj Eulrcloid. 78, l C Evapoialivc odherive.918 Excitner lasc~. 517 Eroihermi~ malciini.275 Expandable paphim molding. 304 Expausion fit,925 Expansron roameq h5Z Expcndablen~i~id carting proccsscs.283 Explosive bonding.352 Expl<sivciivot.415 Explosive wcldiug. (EXWj.886. 'YzR Exlrusion.3411.348.4111.ti21 face centered cubic 60 Face mi11.665 Face milling.530.659 Face pMc.624 Faen& 52X.SB8.6iiL Failure modc%ndhenivrjoints.922 Fasfonerr.827.YZ4.925

1004

Index
0 0 , 1 8 6 Gmpbxte.82 1 Govtty drop hs~ww, 390 Grap=.%timn,83 314 h n hardnm289 GTaen maehinmg. 350 Cr- s d ~ a s ~ n g , j " j Gztreeo ~ d p o u ~ l i y Ezeenstieo~!h463 embg pmamsters,7&S CINidulg safety,?lb Gmdingwhsel ~d~ntiRcal~oo,?hX

Furiguc crmRgiwth iavz 52 m t w fail"ro..M,Y64 Fatigue slm11glh.4S.934 fsligw.44 Fs3C,m m t e 82 FeeSgerlmlb,bBI Feed tbte 661 Feed md,rB.m Ferns gaar+h&per.Rlh Terr~le.80, X1,90 T%sr,lw w r n l m ~ me1 132 F l m 7 ~ a U a ~ d a ~ d i t o aI19 tw~. Fm~~~metdhll8 fib- rcmfemdrompoe. 184,332,35$ hbc%353 Rlamcnl d m g '3% me rtJt 4v Rlc.751 Film& 751 Flllcr melsls844.&, 90I.YO5 Fzllm 170 Fi1letwcld.W

rd,m

Fllle&,m

ZB MI*T$.~& F i n . hlank~np.422.W Finrrh alluwance.277 Fidrahingdcmfmnm229 Fmshbg pFMEAG9,la P~ntshlllrratmcnta,Y4lJ F~*mn,llo.343 firms 349 Fishbare dtagiam Y86 Ftt.92s-926 Fiad atttannlmn.itl2 EX& 1eqlleaIit4 f"n**.72U FZiure, bii, 6% Flame hardenmng I10 niima reliviLcnL 131 FlameVr;ugir6Nhg.X4& F r q m g a3'1 Flank wsar.St2 583 Flash werd&,(FW). 8 0 Fidrh.8431 l a r k , z 6 3 . 345 El~~~urlaccmaeh~ 53J r~ng
F , I ~ C ~rnrrtaag,ng R

I?rllenng,392 Full fmrd s h e . 3% F~fLm~ E(IStmgprwmI(B ld fUt"Ucc t i r a m s 901 P W~ ~ I G I ~ M W S ~ ~ B P w m welang.a,833,m F w l o s i weldin$~ointd=iip831 G mrin 763 . GagrbloCks,216 G a w fapabtllW. Zl9,ZZK as4 Gtaivrn*, 1% Gsn8msndrs1,BZl &**"In"@ m&e, Gap irame press 433 Gap bed lam%6x0 oarnet, 759 Gda ~t~miiaMe.dbl. 06% c d ~ b m l wl,l l 08s flmme 270

Gia'.vmg,StS Group jl&692 Gmup technolqgy,(GT) 15 710 Cucnn pioas$443 Gull vnngicad.lP8 Gundnlls 635 ai-wd.719 Hack%w 743 Hafnium. I59 Hslf hanf$h&et,%4 HB%tiilay-up,?% Hmd lammtag.29@ H.~ndrnm*~,Sl

H W cpraxsmg,V4O Hans fop.798 Kafti a u t e m i l l i a . n r 702

Hard bnked.4% Hard ohmmiuntplaanp. 95.1 Hard f a w 838 HardGmhhty, l M FiaMaa Conuenicm rBblcd1 HardnedS E-1 37

natoes.m.xI
Flexfor",r*>*,444
Rsnible.wmm@an,7m Ge.rea.284
G a 3 l n g s Y s t m , w . 2 7 0

nex+blemaehimngsyn~,~PM5~,7UI
REUbEmanufpcrunnge~llI F M 0 , l I l Fltpdhtp wchnob%y,4Yl mdltingp1up,408 Rou'lmei413 Flaw *o(A3*12 PIQWslsd~l~n~~72 Flaw Iuti"n&4w nuUlIy.270. %I3 Bluidaed bed furnace 113 mrwh pm sags 244 Flut<.@8 FiUrd-ehuckmgresmer,651 FluteM #I% irpS W F k - m d arcwetdmg,fFCAW),8Sl,XS3 Mxelm bra%rn&$09 Flmelfss mlrlcnU~B91J Ftuxesw'i FLU*% bzmnz. me Flyolnre<.66Y fa~m molditz&341 F C I E~lmtc'r U~~~ 172.341 FnVd *hiati.",548

Gcnt box.4lO

G*wfn,sbin&gZ1
CCar&cBcFJtlQg 81s Geirgnndmg, 822 Gwls~lng,8t3 Gear makmg p""s%lJ Gem mkhns8aS 06urshnpwrUlt~i,~l7 W S h a Q b g 813 OesrShuv~ng,BZl Grm~m%lh,lWX OearIqp,;VTI Gra~taaab nomanclaturq808 u m s t aliper,sZ3 Oe&r-~mtb 0.51 o u l l l ~ d s Oapmetnc mmiiiwat1nmd01 Gepmetrk tolez~~~ces. 224.2ill Glam-seram~c134S Glassy sew I n Globular lransIer.g(5 6 . 3plqge&age,W Go~neokdlerasbnp318 &ds,763 Gein b o u n d q 61 Gruin@c~wth.61,6&26B.m Or,;" rdm-"t,%66 Grain susrefinueanh92

Heat aeatmenr2D,B9 Heliarc waldme 85Q Hahcal anglc,791 HdlcSlff~I~$Z\? Heiieol gear,Oll H&dm1Ilu2gcutter, 667 Helix a&e,632 H8*$gond cWe P R B W H a h speed swd [HSS),505,3S,663 H@ w e d =el a1ttiogfouLdI6 i o u l ~ i ~ l136 s, H d i sa~rd

~W&IB& tool siees 135

B i p h - e n c ~ - z a t ~ f - ~ p i H E ~44g l,

~tghi~tbaesieel, 12-3

Fmc-dl%$lscemeniSSJ
FUrgwrldmsr&X7Y kwv~mwold'ig k7*

HzSlo~~m 973.04 . Xob.rn8.819 Hahhn&812,813,818 Hold-down prasure.d<l Bole wttpr,ax Hole saw, 6% Homogenizatloa,$Z Hmeywmbsdatruaurz.352

Honlngrn
H o d lc,oth *w hlade,lS Hmka's law, 32

F a r g l n p . 3 8 9
Fonn swm.XI4 Form " , , l l " ~ ~ w ~ e r . ~ ~ FOE~I miIIta~,813

Orah 97~.2$!8,7
G m n strosiure.61

Index
Hori?oatai push cut s~lspcr.733 Horizontal spindle n i l i ~ l nmacl,ine, g 657 Horn plcss. 454 Honepower,s2~ Hol hardncss.561 no1 1rortatxcrtccl.566 Hot mcir.917 Hot prcrsing. 353 Hot rolling. 383 Hot spots. 279 Hot ?"my p;iin,lag,94V q0L teilrs.273. 2911 301 wclrking.66.371 <ol-hox nlelhod. 100 +31-chamher die c;isting, 318 rot-compression molding, 336 d01dill co;lline.952
Inlermctall~ccompnui~da,79,~36

1005

i ~ ~ l ~ i s o i t a t ~ c p i e (HIP). s s l n r 166,348,469 lot-plutc w'rldlnp.92K I-piocess. 2V3 luhhlng.415

Hyperrufectoid stee1.81 Hypoeulcctold s~cel.81 Hypoid gcur.812 IC packaging. 497 Impact sxtrusion,410 Impact rtrmglh.563 1mp:at testr.42 Imp~~clor.3'13 I1"pl;int ~ c l d 1 ~ g , Y 2 H Impregnafiirn.329.47d llnprcrsioll d m p forging die, 392 llnprcrri081vdic derixn.394 linprcrrion-die h:imn~erbrgiog.3y 1 Inhar. 157 lncllilahle prcii454 lncomplcte fusion.829 lnc~lllplele pcnc,r;,tio,,, X L Y lnconcl 117 ~. . lncrenlental address rystcrn,715 lndext~blr drilling prublcln ic,lving, 641 Indexable insert drill,640 Indexable-lnserl drill.specd and fecd,630 Indcxlng. 674 Indirect cxt1usion.40~ Ind~rcct fuel-fired iuin;ice.326 lllducflon brazing.908 lnductnon furnace.327 bducuun hnrdc818ng. 110 Indur~rial c#lglncsrs,11 Incil pasle.469 Inertia \ r e i d l ~ ~ g . ~ 8 ~ Inked. 757 lnliitral1on.329,474 Infrared radiation. 929 ~nlecitor~ muld~ng.3li4.347.348.358 Inner ~uetaii~ccumpound,6x Innculat~on.X5.266 I n w t tooling, 664 Inscited teeth raw746 Inscrti344 Inrerl~loofh niililngcutter, 668 losolubility 76 Inspeclion e q ~ ~ p m ~ ~ r . 2 2 8 Inspection 111ethods.227
~~

Inrpect1~rn.~,20,2i3.98~
insulating slecves. 275 Integral lasfencrs. 924 Illlegrated clrcuit. IC, 19,492 Inferlerence hands236 Interference iit.2211

Intermediaie~igconcept.693

1"ernaliondl Organiznrion for S f a n d a r d s , ( 1 s ~ ) Intemal~onai Systcm of u n i t r z 1 4 InIcrpoiafion.721 lllteriuptcd cutting, 656 li~ters~~f afoms,93 ial Invcrfmenr ~ ~ s l i304,820 ~g, Involulc carve, 809 Inuolulc gear teeti, formulas. 810 Inn carhurinng. I l l Ion implanration.58I Ion plaLin:. 112,181 I"". oegative. 57 Ion. pos~tive. 57 1on~hcam machining (IB.M),sir Ionic hood.57 lonlticdlng,I I1 lion age. I Iron carbon phnrc diagr.lm.9~) Iron, 119.283 Iron-cslrholl rquilibrlum diagram.79 lronlng,448 khlkcrivl dlagr;lm, 086 Isiandr.488 I S 0 system of limits and firs.22d Ironer. plastics. 163 Irortalic conlpacthon.467 lrobiaiic prenszng. 348 lrothcrrnnl aooes1,98 1salhcrm:~i ioiging.394 laolhcxm;il iornllny.377 l s ~ t h u r n ~ l ~ t r s ~ ~ s f c ~ (IT) r m adiagram, t ~ o n 96 Isutropic p l o p c i f ~ 66 In>d t o f . 4 2 J'g and lint~8re economrs. 698 Jlg dcr~gni. 685 Jlg, 677.679 J-lead, 498 Job shop. 12 Job. \cc sfatlon.5 John P:~r~001,7115 JoiaahiilfiR39 Joisrng ~oicc1an~ics.350 Joining oiplustm.Y27 joining pmcesscs, 19,827 l,,iinr dcsigns.921 sc,lling.291 Ji>minylest, 102 Jump~rooth hmach.73~ Kanhan, I4 Kerl: 743,846 Kcvlar. 185,354 Kcyscelrc~. 734,818 Klck presc.452 Knee-;ind-colun~n mrlling machinc,669 Kiinfrlng.358 Knuckle joint pros, 452 Kn~rl~~p.605 Ladle mefalluigi 120 Lamlnal c o m p o r ~ ~1 c8 , 3,i~i ILnminntii~n,355 l a p , 782 L;ipping,7XZ Lap~uelded pipe, 452 Laerr infcrfrrometcr. 236 Laser powerdensity 516 Laser sinfering.350 Lssrr spot w r l d ! n ~ . X ~ h Lascr typcs,517 Lnser wsldmg. 929 Laser-hem, c u ~ t i n g . ( ~ ~ c ) , 8 9 5 Llscrhcsm hardening. I10 L"SG"">rdm lalchining (LBM), 516 h ~ r ~ k wclding. c m (LBW),~Y~ Lathe cenlcr,619 Lathe dog.620 Lathe, 18 Latiicc ~trucfure, 59 Lay ~ymbols.936 Lead alloys, I5R.313

[Lead piah.498 Lead pot. 113 Lead screw795 Lend shcil nlold.331 Lead~frcc rlloyb, 144 Lead~frcc s<,lden.911 Lcadtmerr. 610 Lcad-rilver alloys, 91 1 Lcsfjlg. 686 Lcak tentmg.256 Lcan manuirc<unng,969 Lcail product1nn.3 Length.214 Levcrl;lw 77 I.FG,ilt>w~iorce groove), 578 1.ife~qclr curuu I Llghlrreight mctalr. 106 Limirr.zz? Liiic sl"p,Y91 Lincnr frlctlon wciding. 927 Linear nieasuring inrfruments,2~9 L'"emvvaiishie dillcrential transformcr, ( ~ v ~ ~ j . 9 3 6 Linrorily 228 i4 Linked ccll manufacturing syrlem, (L~CMS), Llp. 628 Liquid uIoo~iinlion,462 Liquid carhuriiing, 111 Lxquxd penetrant inspec1ion,247 Liquid-phase slnleiing.350,468 Liquiilur, 74,267 Lwaded whcci, 766 Locating, 677 I.ogdrifhimicStraln.35 Losf-loam caillng.308 Lnstii.ax casting. 15,305 Low~alio)rLcel. 124 L~w~clrho sfeel, n 123,375 Lower cuiitml lilnir. (LCL). 9x3 Lowrr natural folerar~cc limit, ( L N T L ) , ~ ~ ~ Lower rpcciticaf~onlimit. ( ~ ~ ~ 1 , 9 7 3 Lower yxeld p,,~nt, 32 L o w h i c e groove ( L F C ; ) , S ~ X LDW~PXSSEUIC prrmancnl-mold (LPpM) L ~ ~ - s l r cgnnding,765 rs Luhrrcaiiun,369,370 Ludcn bands.375 Ma~hlaah~iit~50.589 Machirie connvi unlf. (MCc),702.712 M;ichine icamer. 651 Mnchi~~ tr~ol,5.18,526.59~,647 c Machinczcro pulllt,713 Machln~ng ailowancc,277 Machining crnter.18.702,703,711 Machining dy11smics.546 Mach1ninghonepower.53~ Machining mechanics. 543 Macllining 0 i u r r n m ~ ~ ~ 3 s u Macilln~ng of plastics. 342 Machin~ng PIM productr475 Machining powcr,533 Machining procrrscs.534 M21chlning. 17,523 Machiosr 1ule.229 M;icrorcgrrgation. 77 Magnesium 152.283 Magnesium alloys, table, I 54 Magnesium.313 Magnetic chuck.695 Magnerlc particle insprction.248 Magncfic workholder,695 Mag"ciita(Fe,o,),757 Magnihcufion.228 Malleahlc cast imn 81 Mandrel. 403.407.620 Manncsmann pmccss.414 Manualpreis.452 Manufuciunngceii. 14 Manufacturing concerns. 202 Monufacturmg cost, 2,203 Manufacturing enginee,~, 11

Manufacturing proccss. 3.6 Mnnuf8cluring syslrm delip. 12.23 Mnnufaeruting system, 3 MAPPgns, 844 Ma~agiogrferl, 131 Margm. 628,631 Marking Jyacnl. 768 Manernper. 108 Martensite. 84.98 Manensidc aa~nlasstecl. 132 Mirkanl.485 Masking.4RF Mass finishing methods.944 Mos~.Z14 Malsachusettslnstifute aiTechnology. (MlT).702,705 Masscrjip.692 Match pihie mulding.292 Match-plate pauern.285 Marpnal avaiiabiliLy.21M Maie"al handling, 2 1 1 Material relzcih,n. 195.198 Material ~utriliHlIi011.207 Matciials and processes compaiibality.2ffi Matuials engu~rsra. I1 Maitrials procasing families.263 Matetirls processing. 262 Matrix.353 Mar5353 Measured force, S43 Measurement,ZL3 Mechiin8cul &ive,734 Mitchani~alfasteners.19.929 Mcchanical fiistcning.924 Mrchaniral plrling, 957 ~ e i h a n i t spress394 l Mechanical prnperlies,ZV.201.176 Mechanical propcnies PIM pmducls,475 Media, 943 Medium-cutbonnreel. 123 Mflturg proces%.264 Mdtine 325 hl. '.'hi, ' b~1.i.: I. (3.1 ' 2 " hl,nh,#.<% . . : : . d,&:::~;u hlcrzhrr ; m - . : . ' : . ' Metal cuttinc. 18 523

Madnllar tooling, 427 Modulus of eiasricity.32 Modulusof resilienee.32 Modulus of toughness, 34 Maisture fonteat.288 Mold cnvity 264.265 Mold conJtant,268 Mold haidnes%M8 Mold remaval.264 Mold,2% Molding plates 890 Molding. 15 Moleeuiarntructure of plastic%163 Moleeuiustructurc.59 Mol0culg 56 Molten poo1,844 Moncl. 143.157 Monaohromatif light source,236 Monomer, I64 Monotmtic.78 Mortled zone.84 Mounted abrasive whcol.779 Moving assembly h e . 13 Muller, a87 Multiple-spindle sutamnticscrcu machine.613 Multiple-spindle drillingmachi"e.698 Multiple-tooth cuuing, 570 Multiple-use mol'k,264

Multiple-we-muldaastingprocessff, 313
Moltislidemachine. 456 N d , 925 Natural abrasives.943 Natural struin, 35 NC-word.719 Nd.YAG lascr517.894 N8amct shape forging, 400 Near-nct shape.313 Near-shapeforging.400 Neckin& 33,599 Neutral %me, 843 Neuln)n,56 Ntchrumc,i57 Nickel alloys905 Nickel b s e d aiIoysl57,283 Nickel silver, 143 Nifriding. 111 Niuocarbonizlng, I'll Nitrocellulose lacquer, WR No-bake sand, 2 9 S . m Nodular cast irnn.85 Nodulher.85 No-go gage, 243 NominaL220,970 Non consumsblc elcctrodesrc welding,859 Non~nsumaMo elccmde procesq 850 Nondetiuctive (estmg.246 Nowfenis metali 139 Nowfenir metalipr~pcitiei 140 N0nmetilll1cmaterials,29,162 Nonmetah29 Nonstructural sdherivv.918 Nontraditional machining (NrMj.484 Normal curve, 9 4 Normal di$Lribution.nl Normalizing,Ql Nose radius.582 NTMprocess rummary, 4x6 9 n-type remicandueor, 6 Nodcation, 61.266 Numerical control, (NCj, 14.7132 Nut and ball.925 N-word, 719 Obliq"emachining.535 Offsct yield rtrength,32 Onepiccr pattcm,285 O n s t of shear aogIe.539 Oosrt of sh.hea1.545 Opsn rixr,274 Open-back press.454 opefi.tell roam p ~ a s t i c . ~ t

Mela1 removal. 17.523 Mctu1f"rmi"g. IS M c m h b0"d,5K Mc~allicglass 159 Metallic mateciais29 Metallidng. 898 M e t a l f rir camponi~ci'lRX.35Y Motsi-oxid< renlicanductor (MOS).494 Mcttie to English conversion table.217 Metric to Englirh converrion.216 Metmlogy.227 Micro stmcture.56 Miuoulloyed rtaelz 128 Miuochipping,576 Miaodrilling.642 Miwohardnrss test.39 Memmctcr dcpth gage.234 Micromuier.23ll Milkol11mc.YM Mill~ngattaehmmt.6115 Mlliing cutter. 656 Millingfuture, 689 Milling. 17,124,656 Mirnwsve hearing.92Y Misiun.27O MlT.705 Mixing.4M Modular lixlurr. 690

Open-die hammer forging, 390 Open-loop numerical conuo1.704 Opm-moldprocesing.357 open-side plane5 735 Operation,see process,5 Optical comparator or projector, 235 Oplirll flak236 Orbital fo~vrming.415 Organic Snishingmaterials.949 0"fice gas.862 Orlliogonsl d ~ s k machining 538 Orthogonal machining,538 Orthogonal platemachining.538 Orthogonal lube rurning.538 Osprey proces$451,471 Oveiaging, 53 Oveihang.488 Oxyacetyene-metal spraying gun, 899 Oxyacetylenecutting, (OFC-A),847 Oxyacetylene p r m s &843 Oxyacetylene welding torch. 842 0xydyung flame, 843 Ory-fuel cutting, (OFC). 846 0xyf"elpas welding. (OFWj, 842 Oxygcn lance cutting, (OLC),846 Oxygen-free high-~onductivity (OFHC) coppcr. 141 PIM forging,473 Pack rolling,386 Packaging, 19.29.497 Pack-carbutizing. I l l Painting, 948 Pallet changer,709 Parallel-plate hydiafoming.445 Parent population, 980 Pnison,336 Part piagramming,718 Particle shape.461 Partide size.461 Particulate composites. 184.351 P~rtingiine, 264 Paning plane, 276 Parting rurtrfcr,266 Paiting,5%,602 Pattern allawance&277 Pattern board,284 Patterns276 PCB assembly process502 PCBN.po1ynystalline cubic boron nirride.574 PCD.palynystalline diamond, 573 Pearlile, 81 Pearlite, coarse, 98 Pearlite. fine. 98 Pcarlitic cast iron. 84 Peening.416.838 Penetrameter.253 Penctralon,270,290 Percent eompactability.289 Percent elongation, 34 Percent reductionin area, 34 Percussion welding, (PEW), 897 P e r f m m c e . 28 Peripheral milling,656 Perifecif, 78.80 Pcritectoid.78 Permanent-mold casting processes 313 Permeability number.288 Permeability.287.468 Phasediagram, 267 Phase rransformatiun.92 Phosphate coating. 953 Photoshemieal machining praces,489 Photochemical miIling.488 Photolithography proces,495 Photaresist.493 Physical faiIure.582 Physical properties.29.ao2 Physical vapor deposition (PVD),565,579.958 Pick-and-place machine, 503 Piercing,425 Pig iron.119

Index
Pifiow 'brmzg. 4 8 ' Pin grid arras 498 Piu-"I hole tcchnolagy 498 Pitch mrcl+ 808 P i t a diilmeter.(PD).7Y1.808 Pitch pu,"t.WI8 Pilch,736.791 Plain-carbon slcal.123 Planertype milling machiin+ 671 Planetary gear shaper.816 Planmg. 524,731 Plasma arc welding. (PAWI.I(h2 Plasma-arc cutling (PAC).518,MO Plasnla-srcspmy gun.nY9 Plaster mold casting 302 PIartic deiormaliirn.32.63.363 Plariie formin&348 Plastic muldingli21 Plasticity.363 Plvsticizcm 170 P1asfii-r. 163.334 Pl-tia properties bi l6R Ptdfejig.686 Plale. 382 Pla~en grindcr, 780 Plowmg.761 Pluggage.243 Plug tap,79R Plug weld.830 Point-ta~point cunlrul.714 Feint-m-p~inl machine 712 Poison's raliu, 62 Pobyokc.W3,YV5 Poierizatcon.5Y P~lycryslailine cubic boron nittide. (PCRN),57.1 Polycryrtslline diamond, (PCDl.573 Polymer.164 Polymerizsrion, 164 6 4 Pr~lymarphis Polystyienc,308 l'op-rivet. 415 Porccldin enameling. 958 Poring temperature. 267 Porwily.468 Pouring ladle, 328 P o w h ~ p p r u c l i328 ~. Pourlop techniquc.TOi Powder coating sy;trm, YSZ Powder cualing.951 Powder mc~allurgy (PIM).4M),810 Power h.a~ksaw.748 Powcr injection moldillg (PLM),471 Power m*lallurgy foul steel, 563 Powerhead. 645 Vm pmgrammcdrnach~nining roulinm.717 Prealloyed powder.462 Precipitation hardeningstainl~ss slrcl. 133 Precipitation hardeding, 92 Ptecision plssrnv s r e m r l i ~519 . Precision forging.400 Prccis~on plasma cutting, 8(17 Pracis<un,ZlY. 677 Precision. 9711 Precoated slcel sheet, 131 Prefarm.337 prepreg.353 Press bending. 43 P r e s brcak.431 Press fcuiing deviees.450 Prcsr forging. 391 Pirrr-and-sinter parts.467 Pressed fif.926 Pressure angJa.808 Prwsuie carfing.353 Pressure gas wrldink (VGW).845 Prerjure induced ductility.4n5 Pressure-assisted rintering.410 Press",-bag moldin&,356 Prr~auie~elrsilive adhesivc,Ylb: Pressure-lnnpemlnrctr diagzam.72 Wssurc-to-press~tre eutrusron.405 Prmlcd~i~ board. ~ ~ i t(PCB). 19.51111 Probability di$uibut1ona.Y75 Probe. 724 Piilcess nnneal.Yl Pt-slcapab~lil) calculatians.985 Pb'ocesr capahiliry index.976 P r o e e s caprbilily stud$ V72 P m c e s ~apahiiit)1724.970 ProcebS damping.552 Procrsfmoaeling, 367 P r o m s yield: 4% Prmxesing.2L 357 Producrr goods. 1 Yrnducl liability. 207 Piaduct life cycle.21 Productron design.200 Produdion sysrem.3 ProGIer, 672 Progrsmin?ble uutomaci~~n, 702 PrOgreBSive b m h . 7 4 O Propensive die ~ ~ ~ 4 2 2 Project shup, 14 Projectiun wclding (RPW). 877 P c ~ u~ r ~ i 24s n g Prilpeitier of matcnais, 28 Pn!ron,56 I?ototype, 199 p - t s e iemiconductoi.69 Pull brwch.736 Plllscd spray rrgngfer,855 Pulsod-cunenr ECM, (PECM), 5116 Pulsc-echo technique, 251 Puitmion,354 Pulvetiz~tion.462 Punch hblder. 426

Punch.423
purging468 Pufh-typc b r u s h . 738 Quad mslpackpackagc4Y8 Qlwlily circles,995 Quality control chsit calcularion$982 Quarterhard shcet,3X4 Quam.758 Quench and lemper10i Qucnch cracking, 104 Quench media. 105 Quenching94 Quickchangr chuck. M4 Q u a - r e l e a s e lastaner.925 Quick--stop device, 543 Quill.60R Ra, 937 Rock.812 Racon drill, 633 Rddial drill press 4 ~ ~ dm i n~g i a 8 Radi;>licm~ouring adhzsives.'i20 Rddio~~aph iimpectioo.2S2 y Radius gage.244 Rnkeanglc.761 L~keitamh saw,746 RamEDM.Sl1 Ram. 732 Ram-typemilling machine 6711 Ram-type lumet l a t h c h l i Rapid prototyping. I8 Rasp-cut file. 751 Reacrion injection molding.339 Reaoiion sin~erin&350 Reaming. blJ5.fi51 Rr.Za5t mrUl.844 Rrcryslallizalion. 66.91.92.172 Red brass 141 Kenow oven,SU3 Relractorineis, 287 Refracton fraciorymetals.15X Ref?actdrymarcriab. 178 Relicved hclical drill. 633 Repcat ncwlraru.28

Rexiliance.32 R e ~ i n ~ t n a t f i ~ ~ ~ m187 p~sile. Resinoitl bond. 764 Resin-transfor rnolding.356 Resistance hraring. 908 Resisran=rcam welding. (RSEW).876 Resisanrc rpor weiding.(RSW),874 Xi2 R c s ~ ~ a n welding cc ci-I. Rcsisiann weldmg heating.872 Reststance w l d i n ~ , M 1 Rc~icli~itymefh&s,2i6 Resal"fioh,27,8,d9h Reaolver.714 Resulted knee, 544 Kef;iinedaurrm~te.YY Rctramhle core.317 Reurahlc mold.276 Rruerbeiaioiy furnace, 326 Krveise pnlarily direcl nlrrenf. (RPDC) KeverNrcdraw442 Rcvolutiuns per m h t e (rpm).526,h?Y Rheo casing procesr.322 Ringgage,243 Ringiig.686 Ringrollrng.387 Rise per tooth,(RPr), 737 Rirer d r s i w . 271 Ri$er.265,284 Rivct.YZ5 R i ~ t i n g415 , RuckweU hardncss iest,38 Rod drawing.407 Roll b"dina434 R d l bonding. (ROW).351,88U Ruil exlrusian,414 Roll hnishing,822 Roll forging398 Roil forming.436 Re11 Ievding.437 Roil rhuightaning437 Roll welding,sffl) Roller burniahing,417,940,963 Rouingmill m l l ~ 3 8 R Rolling mill.385 Roiling.342,382 Rolluucrjlp,686 Root-mean-rquarc ualuc.937 Ruseclrucking rebluci. 651 Rotary gzarshapor.8l6 Rotary picrcing,413 Rolarytyp; lahle mlilingmachinc. 671 Rotational moldmg.340 Rotor-tooth hroach.738 Rou&nessproflle.934 Roughness, 606 R"undness,241 Ro~ings, 353 Rubber mold carting.31UI Rubber tooling. 442 Robber, 174.346 Rubber-hondd wllcel. 764 Rubbiqg,lhl Run chair.973.974 Runner well.271 Runnei.M.284 SaddlClypc tuircl lathe, 612 Salt bnth hunac+ 113 &If-bas13 brazing, YD7 Samplingerrois.987 samptins rs1 Sand blasting, 941 Sand c~aing.284 Sand molding propenics 2% Sand muilrt.W7 Sand sIing~r,2911 Sandwiched aliuCLrrk,352

Index
SmK-re ofmstcri*28
Sfruaum,763 Slld W i t & 8SB Styb r m f i l s d c ~ 937 SublsnddrilLb39 Sahlimntiw.n S ~ b w w dm h l d b s , (S~w),851,856 Subpres+e. 427 SW&II,~~~ ~8lIU.916 k micmww,ur *Pnbraaregrinding whoel, 764 SYpcrhiihing,ll superhoa(, q 7 SupeW*.abeel fmning.443 did &ll116W,94 ~w-s=h=md suclacd mid, ",i,!,wm SwfaQ liaishhg y65 SU* grindar.m Susface gibdiqmeth&777 surface &xi"% 757 S " * ~ ~ 1 W Surf- hmrdnu4 961 S u m integity chwacleristiep,935 Surface integrity, 933,%1 @m%e m t teebnoI~,498 s * Plnti~delme~ion.9~ SsrfIb . rm s n u f ~ a w o g 9 3 9 S ~C ~ s hfn c s .604 ~ 8211.924, D37 s u btexiu~,ll.%2 smraEe w m o m p ~ < & % S U r b c % UMmenta P M p m d u m 4 X Suifacd lrulmen9567 Surflein&898 Slauining t e m ~ L o ~ y . 2 Swdn&398 wigt type zittomatic snew mnehinc,65) S Y h sm-snemting & i ~ , X 1 8 SYnleCtis78 Tab~scquenthi forrrm.720 ~metbEd9,Pl. B l a O h n ~ r n TailJmcL.~.MI T%bm mningmiil.386 mna634 ' h m l G a l fom.543 Thermal e x p n s h o . 5 ~9 .9 0 3 nKrmsl fatigm.317 Thermal methods.256 Ibennsl prOfiie.503 Thermal w a y deparition.900 Thermal spraying,W Thermit weldin& (nV),m T h e r m ~ f o r m340 i~ Themomcchani~al~&3~~( n e r m o p l . ~ t k ,i65.167,335,915 Thcrmowt.335 Therhcrmoaetting plastic, 165 Ihcrma~sttins wmkrm.ting951 l k m o a c a i r g . 915 Thko casting pmeen* 322 Thorium, 159 Thresd chain& 798 Evead cutting CNClathe.797 Thread cutting dies.m Thread cuttiq.790 Thread cum% lathc.795 Thread derignatiom 794 Thread gti"dingXU5 Thread milling. 803 T h m d milling.NC, S(11 m i l d mlling,387,7Y0,805,8~ Tbmad rapping. 798 Threading. SW.7W Truc ntriun. 35 rnre stress 35 True stress-tue strain diagram,m TNing.766 Fslot milligcutter66y Tube bending435 X b e llrawing,407 mbe hydmformin~.es Ihberinking.408 h b l e jig.686 Tunp-.tenearbides.568 Tungsten incrt gar welding, mG),n5y Turbulent flow.271 Ihnring. 17.524,528,598.m1 Xrret lathe cutting tooi~619 m n s t iathc.611 X n e t press454 V ~ mI t F driliingmachine.648 m m t type punch prsss.429 n"4ype mrllingmachine,670

1009

Three-jawchuck.62I
Thrilling (drilling + thnading),m Throwaway insert cvtting t m 1 . 6 ~ 7 E X ) technology,4s%.4,49v.~o~ .Ihrough-halc V l k w t r o e , 62% m 3 T~~~ourpermanent-mo. Eaning.315 Time 214

Time-lemperaturc-tran~fo-tim. (~~~),disgram.% Tm allop158 Tin brome. I41 TiN C0~t.d h i g h S p d s t e i , % 6 Tim mating.953

nu 7P8

T i n i o p p r nlloy~Yll Tmned mating.913 Tin-zinc allon911 Tltsnium a l i o p l S 5 Titaoium wbido,569,579 Titanium niuidr. 579 Toggle damp.694
To& mechmim,e52 Toleran~e.220 Tool failure.58~ Tmi grinder, n 8 Tool life, 582,587 TWl Po31.732 Tml settingpoint,717 Toat steel AISl grad- 135 T O O 1 S r C e k 134,563,565 Tooiin&5. 9 Tmlmakera m-rop, ~15 Tml-pon grioder sos Tooth set.745 Topdwr.274 Toreh bruin&POI mrch.842 TOrped0,337 Total indiearormnout, ( ~ 1 ~ ) , 6 3 4 Total quaticymntml. (Tw),969, vw Toughness 34.42.561 Tows. 353 Toyotn Motw C o m p n y , m T o ~ Production a Srstem.(TPS),u %Wer dk 429 Trannler moldin&337 'Rmslor -454

Twis~ dri11.631 Ultimate ntrength.33 Ullimate tmsiic rtiength.33 Ullr+high*pecd machining ~ ~ ~ ~ . ( u H D M c ) , ~ Uitramnie cleaning. y47 Ultrasonic impan @ding, 783 Ultraonic it>spec60n,zo Ultrasonic machining. (USM),7u U l t r m d c welding, (USW),8K5,sn Undcreooling.266 Underwater cutting torehh848 Uniaxial tehtile ~ S S I . ~ UniFd thread Elandard.791 Unilalcral tolerance, 2 u Unit esii. 59 u n i t powez536 Unit atrein.29 Univenal dividing hcsd.673 Universal jig.688 Univrml milling sttarhmcnt.673 Univenal millingmachine,6n) Unpoluymsrilrd plastic, 336 Up grinding,757 Up milling.662 Uppcr mntrol Iimit.983 Upper oatunl tolerance iimit.zzi UPPEPspecification limit.(USL). 973 Ilpper yield poinl32 1Jwet forging. 389.395 u w t w e l d k (uw).an Upretling,W Urethsne.YI7 b u m arc remelting, 122 Vacuum chuck.By6 Vacuum evspratian,581 Vacrivm ind~ctionmcltin~. 122 Yacuum intilLration.353 V m u m molding297 V m u m permanent-mold castiw.317 Vacuum sintcting, 468 Vacuum, 270 Vamum-bagmo1diog.3~ v.l- clcctron.5, van dm wads f0ree,59 Vapor dsgreasing. 947 Vaporjacket, Ill5 vapor.ph== reflow solderiog,913 Yapor-phase soldering.5azaz~i3 Variability. 970 Variable &ion manufaatring rys(om,710 Vsriahlc pitch saw blsde.76 Vatiahie polarity.X50 Variable%214 ~ ~ t i . vJ.dependent, b ~ ~ 367 j Velocity d i m . 542 ventiog 314 Vcna.266 Vcmier caliper. ZW Vcnical broaching maehins.743 Vertical laLe.610

TaE.7cmtina 349 TW-4UrOmated M i n g TAB),*^^ T l p r ream% 6S3 IhpcrIuv.798

BPr t M 5 % %mfhrud.m

TW;sh&. 632 Tappi& 790,798 %y1m tab1 lice w T a y h t w l Meequation.YYI.5~ o m Tcmpvl~inrw nW bpcratUre.Zl4 -Tstweoornpooitan dingram.n kF.-d a'nss, 348 'Rmpedna 100 h i k impact la;* Tensile strain. B 'fdlc.tteng~h a t dsnain P P D .
.((aaaf IW+WU~C,

~1

shenglb m yieIdstrewh. l n

B n a i k rtrcnvh. 33.374
TensiIemss. 29 164 mting.m.z3 mfam9S9 T-red Xhrmal m42m Thermal & m i d rmchCnqm mermal cooductiviw,n,140 'h-~ c@ins ~ ~ ~ S S d&fi~t!mn,829 S S 7hcrmal amin~RZ8.816 l X m a l dcburrlng.s-m ThSrmd dcm&llpWoN, 462 MW o l e t w sm
lhC&mst,

T - s f o ' m a ~ o n - i n d ~ d ~ l ~ i c i r*tsel.(~~1~).129 y 7kaments.Y l k p m i n g gun drill.637 * i b I w , 370 'Ilimming.42 PIP eoaud m b i d e tool, 571 Trve centrifugal casring.372

Vertical pit furnace, 113 Vercical shrper. 734,816 Vertical spindlr milling machine, 644,703 Vertically parted flaskless molding. 292 Vias, JW Vibralmn in machine.548 Vibration welding927 Vibratory finishing, 943 Vickcrs hardness test, 39 Viscous now glasses.347 Vision systcms,239 Virtitication, 347 Vittified bond.764 Vixen-cut 61e.751 V-process 297 wafer. 19 Warm ~ a m p ~ c t i o n . 4 6 7 Warm forming.376 Water atomization.462 Water glars.764 Water jet machining3w.l Water-hardening i w l srcels 135 Waterjet cuttinp,(WJC).784 Wave soldering, 5W.912

Wave-set tooth saw.746 Waviness,934 Wear resistancq563 Wear, 370,561 Weavul~ 358 Web I h i m i ~ g634 , Web.628 Weldnugpt, 872 Weld pool, 833 Weld pressure, 872 Weldahility,50,839 Weldingproeesses,dassification,aL
Welding19.827,927,928,929

Wet pressing. 348 Wellability, W2 Wheel lathe,610 White cart iron.84 Wire bonding, 497 Wire hrushi?g, 945 Wire drawing die.409 Wire drawing machine.4OY Wire qpwing,$8 Wire EDM.511 Woodmff keyseat curter,669

Word~addrcsr foimet. 720 Work hardening, M,374 Workholder design miieiia,680 Workholder design principles. 682 Workholder3 9,677 Wmkholding device, 526 Workholding deuice$lather,619 Workpiece conflguialion, 10 Workpiecf velocity.757 Worm gear, 812 Wrought alloys, 141 Wrought aluminum alloys properties, table, 149 Wrovght aluminum alloys, 147 Yarns, 353 Yield poini.31.375 Young's madulu$32 Zero reference poyi~.713 Zinc alloy$Z83 Zinc b a e d alloys, 154 Zinc.3U Zincgalvanized coa,ing, 953 Zirconia, 181 Zirconium, 159

Acronyms
AC Adaptive Control AFM Abrasive Flow Machining AGVS Automated Guided Vehicle System A1 Artificial Intelligence AFT Automatic Programming of Tools AQL Acceptable Quality L i t (or Level) ASCII American Standard Code ASIRS Automatic StorageIRetrieval System ATE Automatic Test Equipment . . AWJM Abrasive Water Jet Machining BASIC Beginner's All-Purpose Symbolic Instruction Code BTRI Behind the Tape Reader Interface CAD Computer-Aided Design CADICAM Computer-Aided DesigdComputerAided Manufacturing CADID Computer-Aided Drafting and Design CAE Computer-Aided Engineering CAM Computer-Aided Manufacturing CAPP Computer-Aided Process Planning CAT1 Computer-Aided Testing and Inspection CDC Cutter Diameter Compensation CHM Chemlcal Machining CIM Computer-Integrated Manufacturing CL Center Lime CMM Coordinate Measuring Machine CMS Cellular Manufactunng System CNC Computer Numerical Control COBOL Common Business Oriented Language CPR Capacity Resources Planning CPU Central Processing Unit (Computer) CRT Cathode Ray Tube CVD Chemical Vapor Deposition DBM Data-Base Management Direct Data Acquisition System Direct D~grtal Control Digital (or Direct or Distributed) Numerical Control DOS Disk Operating System DP Diametrical Pitch DPRO Digital Position Readout DRO Digital Readout EAROM Electrically-Alterable Read-Only Memory EBCDIC Extended Binary Coded Decimal Interchange Code EBM Electron Beam Machrning (EBW = Weldzng) (EBC = Cutting) ECM Electrochemical Machining EDM Electrodischarge Machining (EDG = Grzndzng) EM1 Electromagnetic Interface EOB End of Block EOP End of Program (workpiece) EOT End of Tape EROM Eraseable Read-Only Memorv ESW Electroslag Welding FCAW Flux Cored Arc Welding FEM Emite-Element Method FMC Flexible Manufacturing Cell FMS Flexible Manufacturing System FORTRAN Formula Translation F'RN Feed Rate Number GMAW Gas Metal Arc Weldrng GT Group Technology GTAW Gas Tungsten Arc Weldrng HAZ Heat Affected Zone HEW High Energy Rate Forming DDAS DDC DNC

HGVS HIP IGES IMPSs I10 IOCS JIT LAN LASER LBM L-CMS LED LP LSI MAP MCU MDI MIG MPS mrp MRPII MSD NC NDT OCR OM OPM OS O'IT

Human-Guided Vehicle System (fork-lift with driver) Hot Isostatic Pressing Initial Graphics Exchange System Integrated Manufacturing Production Systems InpuVOutput InpuVOutput Control System Just-In-Time Local Area Network Light Amplification by Stimulated Emission of Radiation Laser Beam Machining (LBW = Welding) (LBC = Cuttzng) Linked-Cell Manufacturing System Light Emitting Diode Lean Production Large Scale Integration Manufacturing Automation Protocol Machine Control Unit Manual Data Input Metal-Inert Gas Manufacturing Production System Material Reqnlrements Planning Manufacturing Resources Planning Manufacturing System Design Numerically Control NonDestructive Testing (NDE = Evaluation) (NDI = Inspection) Optical Character Recognition Orthogonal Machining Orthogonal Plate Machining Operating System Orthogonal Thbe Turning

PAW PCB PD PDES PLC POK PROM PS P/M PVD QC QMS RAM RIM ROM SAW SCA SMAW SPC SPF SQC TCM TIR TPS TQC USM VA WAN WIP WJM WLK YAG

Plasma Arc Welding (PAC = Cuttmg) (PAM = Machming) Printed Circuit Board Pitch Diameter Product Design Exchange Specification Programmable Logic Controller Production Ordering Kanhan Programmable Read-Only Memory Production System Powder Metallurgy Physical Vapor Deposition Quallty Control Quality Management System Random Access Memory Reaction Injection Molding Read-Only Memory Submerged Arc Welding Single Cycle Automatic Shielded Metal Arc Welding Statistical Process Control Slngle Piece Flow Statistical Quality Control Thermochemcal Machining Total Indicator Readout Toyota Production System Total Quality Control Ultrasonic Machining (USW = Weldzng) Value Analysis Wide Area Network Work-In-Progress (or Process) Water Jet Machining Withdrawl Kanban Yttrium-Aluminum Garnet

I S B N 778-0-470-05512-0