71% found this document useful (7 votes)
10K views383 pagesAnderson J.D. - Computational Fluid Dynamics. The Basics With Applications PDF
Uploaded by
Ramachandra ReddyCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
71% found this document useful (7 votes)
10K views383 pagesAnderson J.D. - Computational Fluid Dynamics. The Basics With Applications PDF
Uploaded by
Ramachandra ReddyCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 383
JOHN D. ANDERSON, JR.
BASICS WITH APPLICATIONS COMPUTATIONAL FLUID DYNAMICS
The Basics with Applications McGraw-Hill Series in Mechanical Engineering
Consulting Editors
Jack P. Holman, Southern Methodist University
John R. Lloyd, Michigan State University
Anderson: Computational Fluid Dynamics: The Basies with Applications
Anderson: Modern Compressible Flow: With Historical Perspective
Arora: Introduction to Optimum Design
Bray and Stanley: Nondestructive Evaluation: A Tool for Design, Manufacturing,
and Se
Burton: Introduction to Dynamic Systems Analysis
Culp: Principles of Energy Conversion
Dally: Packaging of Electronic Systems: A Mechanical Engineering Approach
Dieter: Engineering Design: A Materials and Processing Approach
Driels: Linear Control Systems Engineering
Eckert and Drake: Analysis of Heat and Mass Transfer
Edwards and McKee: Fundamentals of Mechanical Component Design
Gebhart: Heat Conduction and Mass Diffusion
Gibson: Principles of Composite Material Mechanics
Hamrock: Fundamentals of Fluid Film Lubrication
Heywood: inrernal Combustion Engine Fundamentals
Hinze: Turbulence
Holman: Experimental Methods for Engineers
Howell and Buckius: Fundamentals of Engineering Thermodynamics
Hutton: Applied Mechanical Vibrations
Suvinall: Engineoring Considerations of Stress, Strain, and Strength
Kane and Levinson: Dynamics: Theory and Applications
Kays and Crawford: Convective Heat and Mass Transfer
Kelly: Fundamentals of Mechanicul Vibrations
Kimbrell: Kinematics Analysis and Synthesis:
Kreider and Rabl: Heating and Cooling of Buildings
: Kinematics and Dynamics of Machines
Radiative Heat Transfer
= Design of Machinery
Fundamentals of Mechanical Design
Raven: Aufomatic Control Engineering
Reddy: An Introduction 10 the Finite Element Method
Rosenberg and Karnopp: Introduction to Physical Systems Dynamics
Schlichting: Boundary-Layer Theory
Shames: Mechanics of Fluids
Sherman: Viscous Flow
Shigley: Kinematic Analysis of Mechanisms Shigley and Mischke: Mechanical Engineering Design
Shigley and Uieker: Theory of Machines and Mechanisms
Stiffler: Design with Microprocessors for Mechanical Engineers
Stoecker and Jones: Refrigeration and Air Conditioning
Ullman: The Mechanical Design Process
Vanderplaats: Numerical Optimization: Techniques for Engineering Design
with Applications
Wark: Advanced Thermodynamics for Engineers
White: Viscous Fluid Flow
‘Leid: CAD/CAM Theory and Practice
McGraw-Hill Series in Aeronautical and Aerospace Engineering
Consulting Editor
John D. Anderson, Jr., University of Maryland
Anderson: Computational Fluid Dynamics: The Basics with Applications
Anderson: Fundamentals of Aerodynamics
Anderson: Hypersonic and High Temperature Gas Dynamics
Anderson: Introduction to Flight
Anderson: Modern Compressible Flow: With Historical Perspective
Burton: Jntroduction 10 Dynamic Systems Analysis
D’Azzo and Houpis: Linear Control System Analysis and Design
Donaldson: Analysis of Aircraft Structures: An Introduction
Gibson: Principles of Composite Material Mechanics
Kane, Likins, and Levinson: Spacecraft Dynamics
Katz and Plotkin: Low-Speed Aerodynamics: From Wing Theory to Panel Methods
Nelson: Flight Stability and Automatic Control
Peery and Azar: Aircraft Structures
Rivello: Theory and Analysis of Flight Structures
Schlichting: Boundary Layer Theory
White: Viscous Fluid Flow
Wiesel: Spaceflight Dynamics Also Available from McGraw-Hill
Schaum’s Outline Series i
Mechanical Engineering
Most outlines include basic theory, definitions and hundreds of example problems solved
in step-by-step detail, and supplementary problems with answers.
Related titles on the current list inchide:
Acoustics
Continuum Mechanics
Engineering Economies
Engineering Mechanics
Fluid Dy
Fluid Mechanies & Hydraulics
Heat Transfer
Lagrangian Dynamics
Machine Design
Mathematical Handbook of Formulas & Tables
Mechanical Vibrations
Operations Research
Statics & Mechanics of Materials
Strength of Materials .
Theoretical Mechanics
Thermodynamics for Engineers
Thermodynamics with Chemical Applications
Schaum’s Solved Problems Books
Each title in this series is a complete and expert source of solved problems with solutions
worked out in step-by-step detail
Related titles on the current list include:
3000 Solved Problems in Calculus
2500 Solved Problems in Differential Equations
2500 Solved Problems in Fluid Mechanics & Hydraulics
1000 Solved Problems in Heat Transfer
3000 Solved Problems in Linear Algebra
2000 Solved Problems in Mechanical Engineering Thermodynamics
2000 Solved Problems in Numerical Analysis
700 Solved Problems in Vector Mechanics for Engineers: Dynamics
800 Solved Problems in Vector Mechanies for Engineers: Statics
Available at most college bookstores, or for a complete list of titles and prices, write to:
Schaum Division
‘McGraw-llill, Ine.
1221 Avenue of the Americas
New York, NY 10020 COMPUTATIONAL FLUID
DYNAMICS
The Basics with
Applications
John D. Anderson, Jr.
Department of Aerospace Engineering
University of Maryland
McGraw-Hill, Inc.
New York St. Louis San Francisco Auckland Bogoté Caracas
Lisbon London Madrid Mexico City Milan Montreal
New Delhi San Juan Singapore Sydney Tokyo Toronto This book was set in Times Roman.
‘The editors were John J, Corrigan and Eleanor Castellano;
the production supervisor was Denise L. Puryear
The cover was designed by Rafael Hernandez
R.R. Donnelley & Sons Company was printer and binder.
COMPUTATIONAL FLUID DYNAMICS
‘The Basics with Applications
Copyright © 1995 by McGraw-Hill, Inc. All rights reserved. Printed in the United States of
‘America. Except as permitted under the United States Copyright Act of 1976, no part of this
publication may be reproduced or distributed in any form or by any means, or stored in a data
base or retrieval system, without the prior written permission of the publisher.
This book is printed on acid-free paper.
234567890 DOC DOC 9098765
ISBN 0-07-001685-2
Library of Congress C:
alogin;
Anderson, John David
Computational uid dynamics: the basies with applications / John
D. Anderson, sr
p. cm. — (McGraw-Hill series in mechanical enginecring—McGraw-Hill series in acro-
nautical and aerospace engineering)
Includes bibliographical references and index.
ISBN 0-07-001685-2
1. Fluid dynamics—Data processing, I, Title. II. Series.
QASIL.ASS 1995
532/.05'015118 -de20 94-21237 ABOUT THE AUTHOR
John D. Anderson, Jr., was born in Lancaster, Pennsylvania, on October 1, 1937.
He attended the University of Florida, graduating in 1959 with high honors and a
Bachelor of Aeronautical Engineering Degree. From 1959 to 1962, he was a
lieutenant and task scientist at the Aerospace Research Laboratory at Wright-
Patterson Air Force Basc. From 1962 to 1966, he attended the Ohio State University
under the National Science Foundation and NASA Fellowships, graduating with a
Ph.D. in aeronautical and astronautical engineering. In 1966 he joined the U.S.
Naval Urdnance Laboratory as Chiet of the Hypersonic Group. In 14/3, he became
Chairman of the Department of Aerospace Engineering at the University of
Maryland, and since 1980 has been professor of Aerospace Engineering at
Maryland. In 1982, he was designated a Distinguished Scholar/Teacher by the
University, During 1986-1987, while on sabbatical from the university, Dr.
Anderson occupied the Charles Lindbergh chair at the National Air and Space
Museum of the Smithsonian Institution. He continues with the Museum in a part-
time appointment as special assistant for aerodynamics. In addition to his appoint-
ment in aerospace engineering, in 1993 he was elected to the faculty of the
Committee on the History and Philosophy of Science at Maryland,
Dr. Anderson has published five books: Gasdvnamic Lasers: An Introduction,
Academic Press (1976), and with McGraw-Hill, /ntroduction to Flight, 3d edition
(1989), Modern Compressible Flow, 2d Edition (1990), Fundamentals of
Aerodynamics, 2d edition (1991), and Hypersonic and High Temperature Gas
Dynamics (1989). He is the author of over 100 papers on radiative gasdynamics, re-
entry aerothermodynamics, gas dynamic and chemical lasers, computational fluid
dynamics, applied aerodynamics, hypersonic Dow, and the history of aerodynamics.
Dr. Anderson is in Who 8 Who in America, and is a Fellow of the American Institute
of Aeronautics and Astronautics (AIAA). He is also a Fellow of the Washington
‘Academy of Sciences, and a member of Tau Beta Pi, Sigma Tau, Phi Kappa Phi, Phi
Eta Sigma, The American Society for Engineering Education (ASE), The Society
for the History of Technology, and the History of Science Society. He has received
the Lee Atwood Award for excellence in Aerospace Engineering Education from
the AIAA and the ASEE. To SARAH-ALLEN, KATHERINE, AND ELIZABETH
for all their love and understanding CONTENTS
Preface xix
Part I Basic Thoughts and Equations
1 Philosophy of Computational Fluid Dynamics 3
1.1 Computational Fluid Dynamics: Why? 4
1.2 Computational Fluid Dynamics as a Research Too! 6
1,3. Computational Fluid Dynamics as a Design Tool 9
1.4. The Impact of Compntational Fluid Dynamics—Some Other
Examples 13
1.4.1 Automobile and Engine Applications 14
1.4.2 Industrial Manufacturing Applications 17
1.4.3 Civil Engineering Applications 19
14.4 Environmental Engineering Applications 20
1.4.8 Naval Architecture Applications (Submarine Example) 22
1.5 Computational Fluid Dynamics: What Is It? 23
1.6 The Purpose of This Book 32
2 The Governing Equations of Fluid Dynami
Their Derivation, a Discussion of Their
Physical Meaning, and a Presentation of Forms
Particularly Suitable to CFD 37
2.1 Introduction 38
2.2 Models of the Flow 40
2.2.1 Finite Control Volume 41
2.2.2 Infinitesimal Fluid Element 2
2.2.3. Some Comments a
2.3. The Substantial Derivative (Time Rate of Change Following
a Moving Fluid Element B
2.4 The Divergence of the Velocity: Its Physical Meaning 47
2.4.1 A Comment 48
xi xii contents
2.5 The Continuity Equation 49
2.5.1 Model of the Finite Control Volume Fixed in Space 49
2.5.2 Model of the Finite Control Volume Moving with the
Fluid aL
2.5.3 Model of an Infinitesimally Small Element Fixed
in Space 33
2.5.4 Model of an Infinitcsimally Small Pluid Element
Moving with the Flow 38
2.5.5 All the Equations Are One: Some Manipulations 56
2.5.6 Integral versus Differential Form of the Equations:
‘An Important Comment 60
2.6 The Momentum Equation 60
2.7 The Energy Equation 66
2.8 Summary of the Governing Equations for Fluid Dynamics:
With Comments 8
2.8.1 Equations for Viscous Flow (the Navier-Stokes
Equations) 5
2.8.2 Equations for Inviscid Flow (the Euler Equations) 7
2.8.3 Comments on the Governing Equations 78
2.9. Physical Boundary Conditions 80
2.10 Forms of the Governing Equations Particularly Suited for
CED: Comments on the Conservation Form, Shock Fitting,
and Shock Capturing 82
2.11 Summary 92
Problems 8
3 Mathematical Behavior of Partial Differential
Equations: The Impact on CFD 95
3.1 Introduction 95
3.2. Classification of Quasi-Linear Partial Differential Equations 97
3.3 A General Method of Determining the Classification of
Partial Differential Equations: The Eigenvalue Method 102
34 General Rehavior of the Different Classes of Partial
Differential Equations: Impact on Physical and
Computational Fluid Dynamics 105
3.4.1 Hyperholic Equations 106
3.4.2 Parabolic Equations Mi
3.4.3 Elliptic Equations 17
3.4.4 Some Comments: The Supersonic Blunt Body
Problem Revisited 119
3.5. Well-Posed Problems 120
3.6 Summary 121
Problems 121
Part II Basics of the Numerics
4 Basic Aspects of Discretization 125
4.1 Introduction 125
4.2. Introduction to Finite Differences 128 5
‘CONTENTS
43. Difference Equations
4.4 Explicit and Implicit Approaches: Definitions and Contrasts
4.5 Errors and an Analysis of Stability
4.5.1 Stability Analysis: A Broader Perspective
46 Summary
GUIDEPOST
Problems
Grids with Appropriate Transformations
5.1 Introduction
5.2. General Transformation of the Equations
5.2. Metrics and Jacobians
5.4 Form of the Governing Equations Particularly Suited
for CFD Revisited: The Transformed Version
3.5 A Comment
5.6 Stretched (Compressed) Grids
5.7 Boundary-Fitted Coordinate Systems; Elliptic Grid
Generation
GUIDEPOST
5.8 Adaptive Grids
5.9 Some Modern Developments in Grid Generation
5.10 Some Modern Developments in Finite-Volume Mesh
Generation: Unstructured Meshes and a Return to Cartesian
Meshes.
S.11 Summary
Problems
Some Simple CFD Techniques: A Beginning
6.1 Introduction
6.2 The Lax-Wendroff Technique
6.3 MacCormack’s Technique
GUIDEPOST
64 Some Comments: Viscous Flows, Conservation Form,
and Space Marching
6.4.1 Viscous Flows
6.4.2. Conservation Form
643 Space Marching
65 The Relaxation Technique and Its Use with Low-Speed
Inviscid Flow
6.6 Aspects of Numerical Dissipation and Dispersion; Artificial
Viscosity
6.7. The Alternating-Direction-Implicit (ADI) Technique
6.8 The Pressure Correction Technique: Application
to Incompressible Viscous Flow
6.8.1 Some Comments on the Incompressible
Navier-Stokes Equations
xiii
142
145
153
165
165
166
167
168
168
171
178
183
186
186
192
193
200
208
216
216
217
222
223
225
225
225
226
229
232
243
247
248 xiv CONTENTS
68.2 Some Comments on Central Differencing of the
Incompressible Navier-Stokes Equations; The Need
for a Staggered Grid 250
683 The Philosophy of the Pressure Correction Method 253
6.8.4 The Pressure Correction Formula 254
68.5 The Numerical Procedure: The SIMPLE Algorithm 261
6.8.6 Boundary Conditions for the Pressure Correction
Method 262
GUIDEPOST 264
69 Some Computer Graphic Techniques Used in CFD 264
69.1 xy Plots 264
69.2 Contour Plots 205
69.3 Vector and Streamline Plots 270
6.9.4 Scatter Plots 223
6.9.5 Mesh Plots 273
6.9.6 Composite Plots 274
6.9.7 Summary on Computer Graphics 274
6.10 Summary 27
Problems 278
Part III Some Applications
7 Numerical Solutions of Quasi-One-Dimensional
Nozzle Flows 283
7.1 Introduction: The Format for Chapters in Part IIT 283
7.2. Introduction to the Physical Problem: Subsonic-Supersonic
Insentropic Flow 285
7.3 CFD Solution of Subsonic-Supersonic Isentropic Nozzle
Flow: MacCormack’s Technique 288
73.1. The Setup 288
7.3.2 Intermediate Results: The First Few Steps 308
7.3.3. Final Numerical Results: The Steady-State Solution 313
7.4 CFD Solution of Purely Subsonic Isentropic Nozzle Flow 325
7.4.1 The Setup: Boundary and Initial Conditions 327
7.4.2. Final Numerical Results: MacCormack’s Technique 330
7.43 The Anatomy of a Failed Solution 325
7.5 The Subsonic-Supersonic Isentropic Nozzle Solution
Revisited: The Use of the Governing Equations in
Conservation Form 336
7.5.1 The Basic Equations in Conservation Form 337
75.2. The Setup 340
7.5.3. Intermediate Calculations: The First Time Step 345
7.54 Final Numerical Results: The Steady State Solution 351 10
‘CONTENTS
7.6 A Case with Shock Capturing
7.6.1 The Setup
7.6.2 The Intermediate Time-Marching Procedure:
The Need for Artificial Viscosity
7.63 Numerical Results
7.7 Summary
Numerical Solution of a Two-Dimensional
: Prandtl-Meyer Expansion
8.1 Introduction
8.2 Introduction to the Physical Problem: Prandtl-Mever
Expansion Wave—Exact Analytical Solution
8.3 The Numerical Solution of a Prandtl-Meyer Expansion Wave
Flow Ficld
8.3.1 The Governing Equations
83.2. The Setup
83.3 Intermediate Results
83.4 Final Results
84 Summary
Incompressible Couette Flow: Numerical
Solutions by Means of an Implicit Method
and the Pressure Correction Method
9.1 Introduction
9.2 The Physical Problem and Its Exact Analytical Solution
9.3. The Numerical Approach: Implicit Crank-Nicholson
Technique
93.1 The Numerical Formulation
9.3.2. The Setup
9.3.3 Intermediate Results
9.3.4. Final Results
9.4 Another Numerical Approach: The Pressure Correction Method
9.4.1. The Setup
9.4.2 Results
9.5 Summary
Problem
Supersonic Flow over a Flat Plate: Numerical
Solution by Solving the Complete Navier-Stokes
Equations
10.1 Introduction
10.2. The Physical Problem
10.3. The Numerical Approach: Explicit Finite-Difference
Solution of the Two-Dimensional Complete Navier-Stokes
Equations
10.3.1 The Governing Flow Equations
10.3.2. The Setup
xv
356
358
363
364
372
374
314
316
377
377
386
397
407
414
416
416
417
420
421
425
426
430
435
436
442
445
446
447
447
449)
450
450
452 xvi conrents
10.3.3 The Finite-Difference Equations 453
10.3.4 Calculation of Step Sizes in Space and Time 455
10.3.5 Initial and Boundary Conditions 487
10.4 Organization of Your Navier-Stokes Code 459
10.4.1 Overview 459
10.4.2 The Main Program 461
10.4.3. The MacCormack Subroutine 403
104.4 Final Remarks 466
10.5 Final Numerical Results: The Steady State-Solution 466
10.6 Summary ama
Part IV Other ‘opics
11 Some Advanced Topics in Modern CFD:
A Discussion 479
11.1 Introduction 479
11.2. The Conservation Form of the Governing Flow Equations
Revisited: The Jacobians of the System 480
11.2.1 Specialization to One-Dimensional Flow 4g?
11.2.2. Interim Summary 489
11.3 Additional Considerations for Implicit Methods 439
11.3.1 Linearization of the Equations: The Beam and
Warming Method 490
11.3.2. The Multidimensional Problem: Approximate
Factorization 492
11.3.3 Block Tridiagonal Matrices 496
11.3.4. Interim Summary 497
11.4 Upwind Schemes: 497
114.1 Flux-Vector Splitting 300
11.4.2. The Godunov Approach 502
11.4.3 General Comment 507
11.5. Second-Order Upwind Schemes 507
11.6 High-Resolution Schemes: TVD and Flux Limiters 509
11.7 Some Results 510
11.8 Multigrid Method 313
119 Summary sid
Problems S14
12. The Future of CFD sis
12.1 The Importance of CFD Revisited $15
12.2 Computer Graphics in CFD 316
12.3. The Future of CFD: Enhancing the Design Process 317
12.4 The Future of CFD: Enhancing Understanding 526
12.5 Conclusion 533 CONTENTS
Appendix A Thomas’ Algorithm for the
Solution of a Tridiagonal System of Equations
References
Index
xvii
534
539
543 PREFACE
This computational fluid dynamics (CFD) book is truly for beginners. If you have
never studied CFD before, if you have never worked in the area, and if you have no
real idea as to what the discipline is all about, then this book is for you. Absolutely
no prior knowledge of CFD is assumed on your part—only your desire to learn
something about the subject is taken for granted,
The author's single-minded purpose in writing this book is to provide a simple,
satisfying, and motivational approach toward presenting the subject to the reader
who is learning about CFD for the first time. In the workplace, CFD is today a
mathematically sophisticated discipline. In turn, in the universities it is generally
considered to be a graduate-level subject; the existing textbooks and most of the
professional development short courses are pitched at the graduate level. The
present hook is a precursor to these activities, It is intended to “break the ice” for
the reader. This book is unique in that it is intended to be read and mastered before
you go on to any of the other existing textbooks in the field, before you take any
regular short courses in the discipline, and Defure you endeavor to read the existing,
literature. The hallmarks of the present book are simplicity and motivation. It is
intended to prepare you for the more sophisticated presentations elsewhere—to give
you an overall appreciation for the basic philosophy and ideas which will then make
the more sophisticated presentations more meaningful to you later on. The
mathematical level and the prior background in fluid dynamics assumed in this
book are equivalent to those of a college senior in engineering or physical science
Indeed, this book is targeted primarily for use as a one-semester, senior-level course
in CFD; it may also be useful in a preliminary, first-level graduate course,
There are no role models for a book on CFD at the undergraduate level; when
you ask ten different people about what form such a book should take, you get ten
different answers, This book is the author’s answer, as imperfect as it may be,
formulated after many years of thought and teaching experience. Of course, to
achieve the goals stated above, the author has made some hard choices in picking
and arranging the material in this book. It is not a state-of-the-art treatment of the
modem, sophisticated CFD of today. Such a treatment would blow the uninitiated
reader completely out of the water, This author knows; he has seen it happen over
xix XX PREFACE
and over again, where a student who wants to learn about CFD is totally turned off
by the advanced treatments and becomes unmotivated toward continuing further.
Indeed, the purpose of this book is to prepare the reader to benefit from such
advanced treatments at a later date. ‘Ihe present book provides a general
perspective on CFD; its purpose is to turn you, the reader, on to the subject, not to
intimidate you. Therefore. the material in this book is predominately an intuitive,
physically oriented approach to CFD. A CFD expert, when examining this book,
may at first think that some of it is “old-fashioned,” because some of the material
covered here was the state of the art in 1980. But this is the point: the older, tried-
and-proven ideas form a wonderfully intuitive and meaningful leaning experience
for the uninitiated reader. With the background provided by this book, the reader
can then progress to the more sophisticated aspects of CFD in graduate school and
in the workplace. However, to increase the slope of the reader’s learning curve,
state-of-the-art CFD techniques are discussed in Chap. 11, and some very recent
and powerful examples of CFD calculations are reviewed in Chap. 12. In this
fashion, when you finish the last page of this book, you are already well on your
way to the next level of sophistication in the discipline.
This book is in part the product of the author’s experience in teaching a one-
week short course titled “Introduction to Computational Fluid Dynamics,” for the
past ten years at the von Karman Institute for Fluid Dynamics (VKI) in Belgium,
and in recent years also for Rolls-Royce in England. With this experience, this
author has discovered much of what it takes to present the elementary concepts of
CFD in a manner which is acceptable, productive, and motivational to the first-time
student. The present book directly reflects the author’s experience in this regard.
The author gives special thanks to Dr. John Wendt, Director of the VKI, who first
realized the need for such an introductory treatment of CFD, and who a decade ago
galvanized the present author into preparing such a course at VKI. Over the ensuing
years, the demand for this “Introduction to Computational Fluid Dynamics” course
hhas been way beyond our wildest dreams. Recently, a book containing the VKI
course notes has been published; it is Computational Fluid Dynamics: An
Introduction, edited by John F. Wendt, Springer-Verlag, 1992. The present book is a
greatly expanded sequel to this VKI book, aimed at a much more extensive
presentation of CFD pertinent to a one-semester classroom course, but keeping
within the basic spirit of simplicity and motivation.
This book is organized into four major parts. Part I introduces the basic
thoughts and philosophy associated with CFD, along with an extensive discussion
of the governing equations of fluid dynamics. It is vitally important for a student of
CFD to fully understand, and feel comfortable with, the basic physical equations;
they are the lifeblood of CED. ‘The author feels so strongly about this need to fully
understand and appreciate the governing equations that every effort has been made
to thoroughly derive and discuss these equations in Chap. 2. In a sense, Chap. 2
stands independently as a “mini course” in the governing equations. Experience has
shown that students of CFD come from quite varied backgrounds; in turn, their
understanding of the governing equations of fluid dynamics ranges across the presace xxi
spectrum from virtually none w adequate. Students from the whole range of this
spectrum have continually thanked the author for presenting the material in Chap. 2;
those from the “virtually none” extreme are very appreciative of the opportunity to
become comfortable with these equations, and those from the “adequate” extreme
are very happy to have an integrated presentation and comprehensive review that
strips away any mystery ahont the myriad of different forms of the governing
equations. Chapter 2 emphasizes the philosophy that, to be a good computational
fluid dynamicist, you must first be a good fluid dynamicist.
In Part Il, the fundamental aspects of numerical discretization of the governing
equations are developed; the discretization of the partial differential equations
(finite-difference approach) is covered in detail. Here is where the basic numerics
are introduced and where several popular numerical techniques tor solving How
problems are presented. The finite-volume discretization of the integral form of the
equations is covered via several homework problems.
Part III contains applications of CFD to four classic fluid dynamic problems
with well-known, exact analytical solutions, which are used as a basis for
comparison with the numerical CFD results. Clearly, the real-world applications of
CFD.are to problems that do not have known analytical solutions; indeed, CFD is
our mechanism for solving flow problems that cannot be solved in any other way.
However, in the present book, which is intended to introduce the reader to the basic
aspects of CED, nothing is gained by choosing applications where it is difficult to
check the validity of the results; rather everything is gained by choosing simple
flows with analytical solutions so that the reader can fundamentally see the
strengths and weaknesses of a given computational technique against the
background of a known, exact analytical solution. Each application is worked in
great detail so that the reader can see the direct use of much of the CFD
fundamentals which are presented in Parts I and II. The reader is also encouraged to
write his or her own computer programs to solve these same problems, and to check
the results given in Chaps. 7 to 10. In a real sense, although the subject of this book
is computational fluid dynamics, it is also a vehicle for the reader to become more
thoroughly acquainted with fluid dynamics per se. This author has intentionally
emphasized the physical aspects of various flow problems in order to enhance the
reader's overall understanding. In some respect, this is an example of the adage that
a student really learns the material of course N when he or she takes course N + 1.
In terms of some aspects of basic fluid dynamics, the present book represents
course N + 1.
Part IV deals with some topics which are more advanced than those discussed
earlier in the book but which constitute the essence of modern state-of-the-art
algorithms and applications in CFD. It is well beyond the scope of this book to
present the details of such advanced topics—they await your attention in your
future studies. Instead, such aspects are simply discussed in Chap. 11 just to give
you a preview of coming attractions in your future studies. The purpose of Chap. 11
is just to acquaint you with some of the ideas and vocabulary of the most modern
CFD techniques being developed today, Also, Chap. 12 examines the future of xxii prerace
CFD, giving some very recent examples of pioneering applications; Chap. 12
somewhat closes the loop of this book by extending some of the motivational ideas
first discussed in Chap. 1
The matter of computer programing per se was another hard choice faced by
the author. Should detailed computer listings be included in this book as an aid to
the reader’s computer programing and as a recognition of the importance of efficient
and modular programing for CFD? The decision was no, with the exception of a
‘computer listing for Thomas’ algorithm contained in the solution for Couette flow
and listed in App. A. There are good and bad programming techniques, and it
behooves the reader to become familiar and adept with efficient programming.
However, this is not the role of the present book, Rather, you are encouraged to
tackle the applications in Part Ill by writing your own programs as you see fit, and
not following any prescribed listing provided by the author. This is assumed to be
part of your learning process. The author wants you to get your own hands “dirty”
with CFD by writing your own programs; it is a vital part of the learning process at
this stage of your CFD education. On the other hand, detailed computer listings for
all the applications discussed in Part III are listed in the Solutions Manual for this
book, This is done as a service to classroom instructors. In turn, the instructors are
free to release to their students any or all of these listings as deemed appropriate.
Something needs to be said about computer graphics. It was suggested by one
reviewer that some aspects of computer graphics be mentioned in the present book.
It is a good suggestion. Therefore, in Chap. 6 an entire section is devoted to
explaining and illustrating the different computer graphic techniques commonly
used in CFD, Also, examples of results presented in standard computer graphic
format are sprinkled throughout the book.
Something also needs to be said about the role of homework problems in an
introductory, senior-level CFD course, and therefore about homework problems in
the present book. This is a serious consideration, and one over which the author has
mulled for a considerable time. The actual applications of CFD—even the simplest
techniques as addressed in this hook—require a substantial learning period hefore
the reader can actually do a reasonable calculation. Therefore, in the early chapters
of this book, there is not much opportunity for the reader to practice making
calculations via homework exercises. This is a departure from the more typical
undergraduate engineering course, where the student is usually immersed in the
“learning by doing” process through the immediate assignment of homework
problems. Insead, the reader of this book is immersed in first learning the basic
vocabulary, philosophy, ideas, and concepts of CFD before he or she finally
encounters applications—the subject of Part III. Indeed, in these applications the
reader is finally encouraged to set up calculations and to get the experience of doing
some CFD work himself or herself. Even here, these applications are more on the
scale of small computer projects rather than homework problems per se. Even the
reviewers of this book are divided as to whether or not homework problems should
be included; exactly half the reviewers said yes, but the others implied that such
problems are not necessary. This author has taken some middle ground. There are
homework problems in this book, but not very many. They are included in several prerace Xxiii
chapters to help the reader think about the details of some of the concepts being
discussed in the text. Because there are no established role models for a book in
CED at the undergraduate level for which the present book is aimed, the author
prefers to leave the generation of large numbers of appropriate homework problems
to the ingenuity of the readers and instructors—you will want to exercise your own
creativity in this regard
This book is in keeping with the author’s earlier books in that every effort has
been made to discuss the material in an easy-to-understand writing style. This book
will ‘alk (o you in a conversational style in order to expedite your understanding of
material that sometimes is not all that easy to understand.
As stated earlier, a unique aspect of this book is its intended use in
undergraduate programs in engineering and physical science. Since the seventeenth
century, science and engineering have developed along two parallel tracks: one
dealing with pure experiment and the other dealing with pure theory. Indeed, today’s
undergraduate engineering and science curricula reflect this tradition; they give the
student a solid background in both experimental and theoretical techniques.
However, in the technical world of today, computational mechanics has emerged as
a new third approach, along with those of experiment and theory. Every graduate
will in some form or another be touched by computational mechanics in the future.
Therefore, in terms of fluid dynamics, it is essential that CFD be added to the
curriculum at the undergraduate level in order to round out the three-approach
world of today. This book is intended to expedite the teaching of CFD at the
undergraduate level and, it is hoped, to make it as pleasant and painless as possible
to both student and teacher.
A word about the flavor of this book. The author is an aerodynamicist, and
there is some natural tendency to discuss aeronautically related problems. However,
CED is interdisciplinary, cutting across the fields of aerospace, mechanical, civil,
chemical, and even electrical engineering, as well as physics and chemistry. While
writing this book, the author had readers from all these areas in mind. Indeed, in the
CED short courses taught by this author, students from all the above disciplines
have attended and enjoyed the experience. Therefore, this book contains material
related to other disciplines well beyond that of aerospace engineering. In particular,
mechanical and civil engineers will find numerous familiar applications discussed
in Chap. | and will find the ADI and pressure correction techniques discussed in
Chap. 6 to be of particular interest. Indeed, the application of the pressure
correction technique for the solution of a viscous incompressible flow in Chap. 9 is
aimed squarely at mechanical and civil engineers. However, no matter what the
application may be, please keep in mind that the material in this book is generic and
that readers from many fields are welcome.
What about the sequence of material presented in this book? Can the reader
hop around and cut out some material he or she may not have time to cover, say ina
given one-semester course? The answer is essentially yes. Although the author has
composed this book such that consecutive reading of all the material in sequence
will result in the broadest understanding of CKD at the introductory level, he
recognizes that many times the reader and/or instructor does not have that luxury. XXIV | PREFACE
Therefore, at strategic locations throughout the book, specifically highlighted
GUIDEPOSTS appear which instruct the reader where to go in the book and what
to do in order to specifically tailor the material as he or she so desires. The location
of these GUIDEPOSTS is also shown in the table of contents, tor ready reference.
The author wishes to give special thanks to Col. Wayne Halgren, professor of
aeronautics at the US. Air Force Academy. Colonel Halgren took the time to study
the manuscript of this book, to organize it for a one-semester senior course at the
Academy, and to field-test it in the classroom during the spring of 1993. Then he
graciously donated his time to visit with the author at College Park in order to share
his experiences during this field test. Such information coming from an
independent source was invaluable, and a number of features contained in this
book came out of this interaction. The fact that Wayne was one of this author's
doctoral students several years ago served to strengthen this interaction. This author
is proud to have been blessed with such quality students.
The author also wishes to thank all his colleagues in the CFD community for
many invigorating discussions on what constitutes an elementary presentation of
CFD, and especially the following reviewers of this manuscript: Ahmed Busnaina,
Clarkson University; Chien-Pin Chen, University of Alabama-Huntsville; George
S. Dulikravich, Pennsylania State University; Ira Jacobson, University of Virginia;
Osama A. Kandil, Old Dominion University; James McDonough, University of
Kentucky; Thomas J. Mueller, University of Notre Dame; Richard Pletcher, lowa
State University: Paavo Repri. Florida Institute of Technology: P. L. Roe, University
of Michigan—Ann Arbor; Christopher Rutland, University of Wisconsin; Joe F.
Thompson, Mississippi State University; and Susan Ying, Florida State University.
This book is, in part, a product of those discussions. Also, special thanks go to
‘Ms. Susan Cunningham, who was the author's personal word processor for the
detailed preparation of this manuscript. Sue loves to type equations—she should
have had a lot of fun with this book. Of course, special appreciation goes to two
important institutions in the author's life—the University of Maryland for providing
the necessary intellectual atmosphere for producing such a book, and my wife,
Sarah-Allen, for providing the necessary atmosphere of understanding and support
during the untold amount of hours at home required for writing this book. To all of
you, I say a most heartfelt thank you.
So, let’s get on with it! I wish you a productive trail of happy reading and happy
‘computing. Have fun (and I really mean that).
John D. Anderson, Jr. COMPUTATIONAL FLUID DYNAMICS
The Basics with Applications PART
|
BASIC
THOUGHTS
AND
EQUATIONS
I n Part |, we introduce some of the basic philosophy and ideas of computational
fluid dynamics to serve as a springboard for the rest of the book. We also derive
and discuss the basic governing cquations of fluid dynamics under the premise that
these equations are the physical foundation stones upon which all computational
fluid dynamics is based. Before we can understand and apply any aspect of
computational fluid dynamics, we must fully appreciate the governing equations —
the mathematical form and what physics they are describing. All this is the essence
of Part I. CHAPTER
1
PHILOSOPHY
OF
COMPUTATIONAL
FLUID
DYNAMICS
All the mathematical sciences are founded on
relations between physical laws and laws of
numbers, so that the aim of exact science is to
reduce the problems of nature to the determination
of quantities by operations with numbers.
James Clerk Maxwell, 1856
In the late 1970, this approach (the use of
supercomputers to solve aerodynamic problems)
began to pay off. One early success was the
experimental NASA aircraft called HiMAT (Highly
Maneuverable Aircraft Technology), designed to
test concepts of high maneuverability for the
next generation of fighter planes. Wind tunnel
tests of a preliminary design for HiMAT
showed that it would have unacceptable drag
at speeds near the speed of sound; if built that
way the plane would be unable to provide any
useful data, The cost of redesigning it in further
wind tunnel tesis would have been around
$150,000 and would have unacceptably delayed
the project. Instead, the wing was redesigned by a
computer at a cost of $6,000.
Paul E, Ceruzzi, Curator, National Air and Space
Museum, in Beyond the Limits, The MIT Press, 1989
3 4 mosony oF COMPUTATIONAL FLUID DYNAMICS
1.1. COMPUTATIONAL FLUID DYNAMIC:
WHY?
The time: early in the twenty-first century. The place: a major airport anywhere in
the world. The event: a sleek and beautiful aircraft roles down the runway, takes off,
and rapidly climbs out of sight. Within minutes, this same aircraft has accelerated to
hypersonic speed, still within the atmosphere, its powerful supersonic combustion
ramjet engines* continue to propel the aircraft to a velocity near 26,000 fs—
orbital velocity—and the vehicle simply coasts into low earth orbit. Is this the stuff
of dreams? Not really; indeed, this is the concept of a transatmospheric vehicle,
which has been the subject of study in several countries during the 1980s and 1990s.
artist's
parti
ar,
concept for the National Aerospace Plane (NASP), the subject of an intensive study
project in the United States since the mid-1980s. Anyone steeped in the history of
aeronautics, where the major thrust has always been to fly faster and higher, knows
that such vehicles will someday be a reality. But they will be made a reality only
when computational fluid dynamics has developed to the point where the complete
three-dimensional flowfield over the vehicle and through the engines can be
computed expeditiously with accuracy and reliability. Unfortunately, ground test
facilities—wind tunnels—do not exist in all the flight regimes covered by such
hypersonic flight. We have no wind tunnels that can simultaneously simulate the
higher Mach numbers and high flowfield temperatures to be encountered by
transaumospheric vehicles, and the prospects for such wind tunnels in the
twenty-first century are not encouraging. Hence, the major player in the design
of such vehicles is computational fluid dynamics. It is for this reason, as well as
many others, why computational fluid dynamics—the subject of this book-~is so
important in the modern practice of fluid dynamics.+
Computational fluid dynamics constitutes a new “third approach” in the
philosophical study and development of the whole discipline of fluid dynamics. In
the seventeenth century, the foundations for experimental fluid dynamics were laid
in France and England, The cightcenth and nineteenth centuries saw’ the gradual
development of theoretical fluid dynamics, again primarily in Europe. (See Refs. 3—
5 for presentations of the historical evolution of fluid dynamics and aerodynamics.)
As a result, throughout most of the twentieth century the study and practice of fluid
dynamics (indeed, all of physical science and engineering) involved the use of pure
theory on the one hand and pure experiment on the other hand. If you were learning
1, wl
* A supersonic combustion ramjet engine, SCRAMJET for short, is an air-breathing ramjet engine
where the flow through the engine remains above Mach 1 in all sections of the engine, including the
combustor. Fuel is injected into the supersonic airstream in the combustor, and combustion takes place
in the supersonic flow. his is in contrast to a conventional ramjet or gas turbine engine, where the flow
in the combustor is ut a iow subsonic Mach number.
+ Fora basic introduction to the principles of hypersonic flight, see chap. 10 of Ref. 1. For an in-depth
presentation of these principles, see Ref. 2. ff an aerospace plane. (NASA)
fluid dynamics as recently as, say, 1960, you would have been operating in the
“two-approach world” of theory and experiment. However, the advent of the high-
speed digital computer combined with the development of accurate numerical
algorithms for solving physical problems on these computers has revolutionized the
way we study and practice fluid dynamics today. It has introduced a fundamentally
important new third approach in fluid dynamics—the approach of computational
fluid dynamics. As sketched in Fig. 1.2, computational fluid dynamics is today an
equal partner with pure theory and pure experiment in the analysis and solution of
fluid dynamic problems. And this is no flash in the pan—computational fluid
dynamics will continue to play this role indefinitely, for as long as our advanced
human civilization exists. Therefore, by studying computational fluid dynamics
today, you are participating in an awesome and historic revolution, truly a measure
of the importance of the subject matter of this book
However, to keep things in perspective, computational fluid dynamics
provides a new third approach—but nothing more than that, It nicely and
synergistically complements the other two approaches of pure theory and pure
experiment, but it will never replace either of these approaches (as sometimes
suggested). There will always be a need for theory and experiment, The future
advancement of fluid dynamics will rest upon a proper balance of all three 6 PricosopHy OF COMPUTATIONAL FLUID DYNAMICS
Pure Pure
‘experiment | tery
‘Computational
fluid
dynami
\
The “three dimensions” of fluid dynamics.
approaches, with computational fluid dynamics helping to interpret and understand
the results of theory and experiment, and vice versa.
Finally, we note that computational Nuid dynamics is commonplace enough
today that the acronym CFD is universally accepted for the phrase “computational
fluid dynamics.” We will use this acronym throughout the remainder of this book.
1.2, COMPUTATIONAL FLUID DYNAMICS
AS A RESEARCH TOOL
Computational fluid dynamic results are directly analogous to wind tunnel results
obtained in a laboratory—they both represent sets of data for given flow con-
figurations at different Mach numbers, Reynolds numbers, etc. However, unlike a
wind tunnel, which is generally a heavy, unwieldy device. a computer program (say
in the form of floppy disks) is something you can carry around in your hand. Or
better yet, a source program in the memory of a given computer can be accessed
remotcly by people on terminals that can be thousands of miles away from the
computer itself. A computer program is, therefore, a teadily transportable tool, a
“transportable wind tunnel.”
Carrying this analogy further, a computer program 1s a tool with which you
can carry out numerical experiments. For example, assume that you have a program
which calculates the viscous, subsonic, compressible flow over an airfoil, such as
that shown in Fig. 1.3. Such a computer program was developed by Kothari and
Anderson (Ref. 6); this program solves the complete two-dimensional Navier-
Stokes equations for viscous flow by means of a finite-difference numerical
technique. The Navier-Stokes equations, as well as other governing equations
for the physical aspects of fluid flow, are developed in Chap. 2. The computational
techniques employed in the solution by Kothari and Anderson in Ref. 6 are standard
approaches-—all of which are covered in subsequent chapters of this book.
Therefore, by the time you finish this book, you will have all the knowledge
necessary to construct, among many other examples, s
equations for compressible flows over airfoils, just as described in Ref. 6. Now,
assuming that you have such a program, you can carry out some interesting
experiments with it—experiments which in every sense of the word are analogous to
those you could carry out (in principle) in a wind tunnel, except the experiments you COMPUTATIONAL FLUID DYNAMICS AS 4 RESEARCH TOOL 7
(b) Turbutent flow
FIG. 1.3
Example of a CFD numerical experiment, (a) Instantaneous streamlines ever a Wortmann airfoil
(FX63-137) for laminar flow. Re = 100,000; M,. = 0.5; zero angle of attack. The laminar flow is
unsteady; this picture corresponds to only one instant in time. (b) Streamlines over the same airfoil for
the same conditions except that the flow is mirhulent
perform with the computer program are numerical experiments. To provide a more
conerete understanding of this philosophy, Iet us examine one of these numerical
experiments, gleaned from the work of Ref. 6
This is an example of a numerical experiment that can elucidate physical
aspects of a flow field in a manner not achievable in a real laboratory experiment.
For example, consider the subsonic compressible flow over the Wortmann airfoil
shown in Fig, 1.3. Question: What arc the differences between laminar and turbulent
flow over this airfoil for Re = 100,000? For the computer program, this is a
straightforward question—it is just a matter of making one run with the turbulence
model switched off (laminar flow), another run with the turbulence model switched
on (turbulent flow), and then comparing the two sets of results. In this fashion you
can dabble with Mother Nature simply by tuming a switch in the computer
program~-something you cannot do quite as readily (if at all) in the wind tunnel.
For example, in Fig. 1.3a the flow is completely laminar. Note that the calculated
flow is separated over both the top and bottom surfaces of the airfoil, even though
the angle of attack is zero. Such separated flow is characteristic of the low Reynolds
number regime considered here (Re = 100,000), as discussed in Refs. 6 and 7.
‘Moreover, the CFD calculations show that this laminar, separated flow is unsteady.
The numerical technique used to calculate these flows is a time-marching method,
using a time-accurate finite-difference solution of the unsteady Navier-Stokes
equations. (The philosophy and numerical details associated with time-marching
solutions will be discussed in subsequent chapters.) The streamlines shown in Fig. 8 puiLosopy OF COMPUTATIONAL FLUID DYNAMICS
1.3a are simply a “snapshot” of this unsteady flow at a given instant in time. In
contrast, Fig. 1.34 illustrates the calculated streamlines when a turbulence model is
“turned on” within the computer program. Note that the calculated turbulent flow is
attached flow; moreover, the resulting flow is steady. Comparing Fig. 1.3a and b, we
see that the laminar and turbulent flows are quite different; moreover, this CFD
numerical experiment allows us to study in detail the physical differences hetween
the laminar and turbulent flows, all other parameters being equal, in a fashion
impossible to obtain in an actual laboratory experiment.
Numerical experiments, carried out in parallel with physical experiments in
the laboratory, can sometimes be used to help interpret such physical experiments,
and even to ascertain a basic phenomenological aspect of the experiments which is
not apparent from the laboratory data. The laminar/turbulent comparison reflected
in part in Fig. 13a and b is such a case. This comparison has even more
implications, as follows. Consider Fig. 1.4, which is a plot of experimental wind
tunnel data (the open symbols) for lift coefficient c; as a function of angle of attack
for the same Wortman airfoil. The experimental data were obtained by Dr. Thomas
Mueller and his colleages at Notre Dame University. (See Ref. 7.) The solid
symbols in Fig. 1.4 pertain to the CFD results at zero angle of attack, as described in
Ref. 6. Note that there are two distinct sets of CFD results shown here. The solid
circle represents a mean of the laminar flow results, with the brackets representing
the amplitude of the unsteady fluctuations in c, due to the unsteady separated flow,
as previously illustrated in Fig. 1.3a. Note that the laminar flow value of ¢; is not
even close to the experimental measurements at « = 0. On the other hand, the solid
square gives the turbulent flow result, which corresponds to steady flow as
previously illustrated in Fig, 1.3. The turbulent value of c7 is in close agreement
FIG. 4
Example of a CFD numerical experi-
ment, Lift coefficient versus angle of
attack for a Wortmann airfoil
ed Laminar Re = 100,000; Mg. = 0.5 COMPUTATIONAL, FLUID DYNAMICS AS A DESIGN TOOL 9
os:
04
f
C1 ¢
wm |
poet tp
m4 4 Zo ms
‘© Experimental results deg
Present calculations:
1m Turbulent
Heat Laminar
FIG. 15
Example of a CFD numerical
experiment. Drag coefficient ver-
sus angle of attack for a Wort-
mann airfoil. Re = 100,000:
My = 05;
with the experimental data. This comparison is reinforced by the results shown in
Fig. 1.5, which is a plot of the airfoil drag coefficient versus angle of attack. The
open symbols are Mueller’s experimental data, and the solid symbols at « = 0 are
the CFD results. The fluctuating laminar values of the computed cy are given by the
solid circle and the amplitude bars; the agreement with experiment is poor. On the
other hand, the solid square represents the steady turbulent result; the agreement
with experiment is excellent for this case. The importance of this result goes heyond
just a simple comparison between experiment and computation. During the course
of the wind tunnel experiments, there was some uncertainty, based on the
experimental observations themselves, as to whether of not the flow was laminar
or turbulent. However, examing the comparisons with the CFD data shown in Figs
14and 1.5, we have to conclude that the flow over the airfoil in the wind tunnel was
indeed turbulent, because the turbulent CFD results agreed with experiment
whereas the laminar CFD results were far off. Here is a beautiful example of
how CED can work harmoniously with experiment—not just providing a quanti-
tative comparison, but also in this case providing a means to interpret a basic
phenomenological aspect of the experimental conditions. Here is a graphic example
of the value of numerical experiments carried out within the framework of CFD.
1.3 COMPUTATIONAL FLUID DYNAMICS
AS A DESIGN TOOL
In 1950, there was no CFD in the way that we think of it today. In 1970, there was
CED, but the type of computers and algorithms that existed at that time limited all
practical solutions essentially to two-dimensional flows. The real world of fluid 10 paivosopty oF comPUTATIONAL FLUID DYNAMICS
dynamic machines—compressors, turbines, flow ducts, airplanes, ete.—is mainly a
three-dimensional world. In 1970, the storage and speed capacity of digital
computers were not sufficient to allow CFD to operate in any practical fashion
in this three-dimensional world. By 1990, however, this story had changed
substantially. In today’s CFD, three-dimensional fiow field solutions are abundant;
they may not be routine in the sense that a great deal of human and computer
resources are still frequently needed to successfully carry out such three-dimen-
sional solutions for applications like the flow over a complete airplane config-
uration, but such solutions are becoming more and more prevalent within industry
and government facilities. Indeed, some computer programs for the calculation of
three-dimensional flows have become industry standards, resulting in their use as a
tool in the design process, In this section, we will examine one such example, just to
emphasize the point.
Modern high-speed aircraft, such as the Northrop F-20 shown in Fig. 1.6, with
their complicated transonic aerodynamic flow patterns, are fertile ground for the use
of CFD as a design tool. Figure 1.6 illustrates the detailed pressure coefficient
variation over the surface of the F-20 at a nearly sonic freestream Mach number Mx,
of 0.95 and an angle of attack « of 8°. These are CFD results obtained by Bush,
Jager, and Bergman (Ref. 9), using a finite-volume explicit numerical scheme
developed by Jameson et al. (Ref. 10). In Fig. 1.6a, the contours of pressure
coefficient are shown over the planform of the F-20; a contour line represents a
locus of constant pressure, and hence regions where the contour lines cluster
together are regions of large pressure gradients. In particular, the heavily clustered
band that appears at the wing trailing edge and wraps around the fuselage just
downstream of the trailing edge connotes a transonic shock wave at that location.
Other regions involving local shock waves and expansions are clearly shown in Fig.
1.6. In addition, the local chordwise variations of the pressure coefficient over the
top and bottom of the wing section are shown for five different spanwise stations in
Fig. 1.6b tof. Here, the CFD calculations, which involve the solution of the Euler
equations (see Chap. 2), are given by the solid curves and are compared with
experimental data denoted by the solid squares and circles. Note that there is
reasonable agreement between the calculations and experiment. However, the major
point made by the results in Fig. 1.6 relative to our discussion is this: CFD provides
a means to calculate the detailed flow field around a complete airplane config-
uration, including the pressure distribution over the three-dimensional surface. This
knowledge of the pressure distribution is necessary for structural engineers, who
need to know the detailed distribution of aerodynamic loads over the aircraft in
order to properly design the structure of the airframe. This knowledge is also
necessary for aerodynamicists, who obtain the lift and pressure drag by integrating
the pressure distribution over the surface (see Ref. 8 for details of such an
integration). Moreover, the CFD results provide information about the vortices
which are formed at the juncture of the fuselage strakes and the wing leading edge,
such as shown in Fig. 1.7, also obtained from Ref. 9. Here, the values of M,, and «
are 0.26 and 25°, respectively. Knowing where these vortices go and how they
interact with other parts of the airplane is essential to the overall aerodynamic
design of the airplane. FIG. L6
An example of the calculation of the aerodynamic flow field for « cou ay figuration, the
Northrop F-20 fighter. (2) Contours of pressure coefficie 1 entire upper surface of the airplane
are shown. (b)~( f°) The graphs give the pressure coefficient variation over the top and bottom of the
wing at various spanwise locations, denoted by ich is the spanwise location referenced to the
semispan of the wing. (From Ref. 9.
In short, CED is playing a strong role as a design tool. Along with its role as a
research tool as described in Sec. 1.2, CFD has become a powerful influence on the
way fluid dynamicists and aerodynamicists do business. Of course, one of the
purposes of this book is to introduce you developing and using this
to means 0
power, 0
24 24
29) 224]
“16 16
me
cP Fs 08
04 oa
0.0] 00
04
08 sem
(9
"Qo°0i 02 03 04 05 06 07 08 09 10
we
(6) = 0.406
"Qo OT 02 03 04 05 06 07 08 09 10
MC
0531
©
MC
(d) n=0.656
0 0.1 02 03 04 05 06 07 08 09 10
"Qo 07 02 03 04 05 06 07 08 09 10
Me
(n= 0.769
2.0
Test, upper surface
© Test, lower surface
— Euler
o.9|
08|
135
12
0.1 02 03 04 05 06 0.7 08 09 10
Zeal
o
B91 14 THE IMPACT OF COMPUTATIONAL FLUID
DYNAMICS—SOME OTHER EXAMPLES
Historically, the early development of CFD in the 1960s and 1970s was driven by
the needs of the aerospace community, Indeed
described in Sec. 1.1 to 1.3 are from
d, the examples of CFD applications
community, However, modern CFD cuts
across alll dis where the flow of a fluid is important. The purpose of this
section is to highlight some of these other, nonaerospace, applications of CFD. 14 paivosopiy oF COMPUTATIONAL FLUID DYNAMICS
1.4.1 Automobile and Engine Applications
To improve the performance of modern cars and trucks (environmental quality, fuel
economy, ete.), the automobile industry has accelerated its use of high-technology
research and design tools. One of these tools is CFD. Whether it is the study of the
extemal flow over the body of a vehicle, or the internal flow through the engine,
CFD is helping automotive engineers to bettct understand the physival flow
processes, and in turn to design improved vehicles. Let us examine several such
examples.
The calculation of the external airflow over a car is exemplified by the paths of
air particles shown in Fig. 1.8. The outline of the left half of the car is shown by the
mesh distrinted over its surface, and the white streaks are the calculated paths of
various air particles moving over the car from left to right. These particle paths were
calculated by means of a finite-volume CFD algorithm. The calculations were made
over a discrete thtee-dimensional mesh distributed in the space around the car; that
portion of the mesh on the center plane of symmetry of the car is illustrated in Fig.
1.9. Note that one of the coordinate lines of the mesh is fitted to the body surface, a
so-called boundary-fitted coordinate system. (Such coordinate systems are dis-
cussed in Sec. 5.7.) Figures 1.8 and 1.9 are taken from a study by C. T. Shaw of
Jaguar Cars Limited (Ref. 58). Another example of the calculation of the external
flow over a car is the work of Matsunaga et al. (Ref. 59). Figure 1.10 shows contours
of vorticity in the flow field over a car, obtained from the finite-difference
calculations described in Ref. 59.* (Aspects of finitc-difference methods are
discussed throughout this book, beginning with Chap. 4.) Here, the calculations
are made on a three-dimensional rectangular grid, a portion of which is shown in
Fig, 1.11. ‘The fundamentals of grid generation—an important aspect of CFD—are
discussed in Chap. 5, and special mention of cartesian, or rectangular, grids
wrapped around complex three-dimensional bodies is made in Sec. 5.10.
The calculation of the internal flow inside an internal combustion engine such
as that used in automobiles is exemplified by the work of Griffin et al. (Ref. 60)
Here, the unsteady flow ficld inside the cylinder of a four-stroke Otto-cycle engine
was calculated by means of a time-marching finite-difference method. (Time-
marching methods are discussed in various chapters of this book.) The finite-
difference grid for the cylinder is shown in Fig. 1.12. The piston crosshatched at the
bottom of Fig. 1.12 moves up and down inside the cylinder during the intake,
compression, power, and exhaust strokes; the intake valves open and close
appropriately; and an unsteady, recirculating flow field is established inside the
cylinder. A calculated velocity pattern in the valve plane when the piston is near the
bottom of its stroke (bottom dead center) during the intake stroke is shown in Fig,
1.13. These carly calculations were the first application of CFD to the study of flow
* Recall that vorticity is defined in fiuid dynamics as the vector quantity Vx V. which is equal to twice
the instantaneous angular velocity of a fluid element. Contours of the x component of vorticity (in the
flow direction) are shown in Fig. 1.10 16. reosorr MPUTATIONAL FLUID DYNAM
FIG. 1.11
A portion of a rectangul
1.10, (From Ref, 59. Reprin
resian) grid wrapped around a car, used
the caculations shown in Fi
SAE SP-908, 1992. © 1992 Society of Automotive
tons valve Exhaust vl
ore cloned
docierese arog f
LLL 5 es
LLZZZAZZZZ| MG.
portion ofl ¢ plane in cylindrical coordinates fea
piston-ylinde in Ret. 60. Only about half the ‘THE IMPACT OF COMPUTATIONAL FLUID DYNAMICS.-SOME OTHER EXAMPLES 17
ythd
er dy
hie
yee
ee
VA
‘
Ge!
‘Velocity pattem in the valve plane near bottom dead center of the intake stroke for a piston-cylinder
arrangement in an internal combustion engine. (From Ref. 60.)
fields inside internal combustion engines. Today, the massive power of modem CFD
is being applied by automotive engineers to study all aspects of the details of
internal combustion engine flow fields, including combustion, turbulence, and
coupling with the manifold and exhaust pipes.
‘As an example of the sophistication of modem CFD applications to a gas
turbine engine, Fig. 1.14 illustrates a finite-volume mesh which is wrapped around
both the external region outside the engine and the inernal passages through the
compressor, the combustor, the turbine, etc. (Grids and meshes are discussed in Sec.
5.10.) This complex mesh is generated by researchers at the Center for Computa-
tonal Field Simulation at Mississippi State University and is a precursor to a
coupled external-internal CFD calculation of the complete flow process associated
with a gas turbine. In the author's opinion, this is one of the most complex and
interesting CFD grids generated to date, and it clearly underscores the importance of
CFD to the automotive and the gas turbine industry.
1.4.2, Industrial Manufacturing Applications
Here we will give just two examples of the myriad CFD applications in manu-
facturing,
Figure 1.15 shows a mold being filled with liquid modular cast iron. The
liquid iron flow field is calculated as a function of time. The liquid iron is
introduced into the cavity through two side gates at the right, one at the center and
the other at the bottom of the mold. Shown in Fig. 1.15 are CFD results for the
velocity field calculated from a finite-volume algorithm; results are illustrated for
three values of time during the filling process: an early time just after the two gates
are opened (top figure), a slightly later time as the two streams surge into the cavity
(center figure). and yet a later time when the two streams are impinging on each 18 prmLosopity OF COMPUTATIONAL FLUID DYNAMICS
FIG. 1.14
A zonal mesh which simultaneously covers the external region around a jet engine and the internal
passages through the engine. (Courtesy of the Center for Computational Fluid Simulation, Mississippi
other (bottom figure). These calculations were made by Mampaey and
WTCM Foundary Research Center in Belgium (Ref. 61). Such CFD calculations
give a more detailed understanding of the real flow behavior of the liquid metal
during mold filling and contribute to the design of improved casting techniques.
‘A second example of CFD in manufacturing processes is that pertaining to the
manufacture of ceramic composite materials. One method of production involves
the chemical vapor infiltration technique wherein a gaseous material flows through a
porous substrate, depositing material on the substrate fibers and eventually forming
a continuous matrix for the composite. Of particular interest is the rate and manner
in which the compound silicon carbide, SiC, is deposited within the space around
the fibers. Recently, Steijsiger et al. (Ref, 62) have use CFD to model SiC deposition
in a chemical vapor deposition reactor. The computational mesh distribution within
the reactor is shown in Fig, 1,16. The computed streamline pattern inside the reactor
is shown in Fig. 1.17. Here, 2 gaseous mixture of CH,SiCls and Hp flows into the
reactor from a pipe at the bottom. The ensuing chemical reaction produces SiC,
which then deposits on the walls of the reactor. The calculations shown in Fig. 1.17 ‘THE IMPACT OF COMPUTATIONAL FLUID DYNAMICS—SOME OTHER EXAMPLES 19
@ )
©
FIG. 1.15
‘Computed results at three different times for the velocity field set up by liquid iron flowing into a mold
from two gates on the right side of the mold. (After Ref. 61.)
are from a finite-volume solution of the governing flow equations, and they
Tepresent an application of CFD as a research tool, contributing information
of direct application to manufacturing.
1.4.3 Civil Engineering Applications
Problems involving the rheology of rivers, lakes, estuaries, etc., are also the subject
of investigations using CFD. One such example is the pumping of mud from an
underwater mud capture reservoir, as sketched in Fig. 1.18. Here, a layer of water
sits on top of a layer of mud, and a portion of the mud is trapped and is being sucked
away at the bottom left. This is only half the figure, the other half being a mirror
image, forming in total a symmetrical mud reservoir. The vertical line of symmetry 20 pitosorty oF COMPUTATIONAL FLUID DYNAMICS:
1.00 Aaa 7 Tr
T
| |
x(n) FE } t =
075 =} ne : +4
050
+
025 F
0.020 0.010 0.000 0010 0.020
rom
FIG. 1.16
Finite-volume mesh for the calculation of the flow in a chemical vapor deposition reactor.
(After Ref. 62.)
is the vertical line at the left of Fig. 1.18. As the mud is sucked away at the bottom
left, a crater is formed in the mud layer which fills with water. The only mation of
the water is caused by the filling of this crater. The computed velocity field in both
the water and mud at a certain instant in time is shown in Fig. 1.19, where the
magnitude of the velocity vectors are scaled against the arrow designated as 1 cvs.
These results are from the calculations of Toorman and Berlamont as given in Ref.
63. These results contribute to the design of underwater dredging operations, such
as the major offshore dredging and beach reclamation project carried out at Ocean.
City, Maryland, in the early 1990s,
1.4.4 Environmental Engineering Applications
The discipline of heating, air conditioning, and general air circulation through
buildings have all come under the spell of CFD. For example, consider the propane- ‘THE IMPACT OF COMPUTATIONAL FLUID DYNAMICS—SOME OTHER EXAMPLES 21
0.020 0.010.
0010 0.020
rom)
FIG. 1.17
Computed streamline pattern for the flow of CH;SiCl; and Hp into a chemical vapor deposition reactor.
(fer Ref. 62.)
burning furnace sketched in Fig. 1.20, taken from Ref. 64. The calculated velocity
field through this furnace is shown in Fig. 1.21; the velocity vectors emanating from
grid points in a perpendicular vertical plane through the furnace are shown. These
results are from the finite-difference calculations made by Bai and Fuchs (Ref. 64).
Such CFD applications provide information for the design of furnaces with in-
creased thermal efficiency and reduced emissions of pollutants.
A calculation of the flow from an air conditioner is illustrated in Fig. 1.22 and
1.23. A schematic of a room module with the air supply forced through a supply slot
in the middle of the ceiling and return exhaust ducts at both corners of the ceiling is
given in Fig. 1.22. A finite-volume CFD calculation of the velocity field showing
the air circulation pattern in the room is given in Fig. 1.23. These calculations were
made by McGuirk and Whittle (Ref. 65).
An interesting application of CFD for the calculation of air currents
throughout a building was made by Alamdari et al. (Ref. 66). Figure 1.24 shows
the cross section of an office building with two symmetrical halves connected by a 22. PiiLosopiy oF COMPUTATIONAL FLUID DYNAMICS,
Remeshing
zone |
Remeshing '
zone 2 \
|e
nat Density profile
T.=0
wv: velocity profile
‘ea 03 ass |
ge _ ,
Latte -
FIG. 1.18
‘Mud and water layers in a model ofa mud trap, with mud being suctioned away at the bottom left. (Afier
Ref 63)
passageway. Each half has a large. glazed atrium, in keeping with a popular trend in
architectural design. These atria, in connection with suitable locations for air inlet
and exhausts, provide a natural ventilation system which is cost- and energy-
efficient. A typical wintertime simulation of the velocity field in a cross section
through the entrance hall, as calculated from a finite-volume CFD algorithm, is
shown in Fig. 1.25.
1.4.5 Naval Architecture Applications (Submarine
Example)
Computational fluid dynamics is a major tool in solving hydrodynamic problems
associated with ships, submarincs, torpedos, ct. An example of a CFD application
to submarines is illustrated in Fig. 1.26 and 1.27. These calculations were made by
the Science Applications International Corporation and were provided to the author
by Dr. Nils Salveson of SAIC. Figure 1.26 shows the multizonal grid used for the
flow calculations over a generic submarine hull, (Such zonal grids are discussed in
Sec. 5.9.) The three-dimensional Navier-Stokes equations for an incompressible
flow are solved, including a turbulence model, for the flow over this submarine,
Some results for the local streamline pattern at the stern of the submarine are given
in Fig. 1.27. Flow is moving from left to right. Here we see an example of a
numerical experiment, following the philosophy set forth in Sec. 1.2. The upper half
of the figure shows the streamline with a propeller, and the lower half shows the COMPUTATIONAL FLUID DyAaaucs: wir Is ter 23
FIG, 1.19
Computed velocity field for the two-
layer water and mud model shown in
Fig. 1.18: results after 240 s of sue-
tioning, (fier Ref. 63.)
N
sis
KK
\\\V
Wi
AN
ANA
ANNI
Ant
\
WA
ANANt
VY)
\\\
Mi
WA\\h
WA
ne WA
a\\\\
\\\\\
ency propane furnace model.
(From Ref. 64.)
streamlines without a propeller. In the latter case, flow separation is observed at the
first corner, whereas with the propeller no flow separation takes place.
1.5 COMPUTATIONAL FLUID DYNAMICS:
WHAT IS IT?
Question: What is CFD? To answer this question, we note that the physical aspects
of any fluid flow are governed by three fundamental principles: (1) mass is
conserved; (2) Newton’s second law (force = mass x acceleration); and (3) energy 24 PHILOSOPHY OF COMPUTATIONAL FLUID DYNAMICS
FIG. 1.21
Computed velocity field for the flow through the furnace in Fig. 1.20. Velocity vectors are shown
emanating from points in a vertical perpendicular plane through the furnace. (From Ref. 64.)
is conserved, These fundamental physical principles can be expressed in terms of
basic mathematical equations, which in their most general form are either integral
equations or partial differential equations. These equations and their derivation are
the subject of Chap. 2. Computational fluid dynamics is the art of replacing the
integrals or the partial derivatives (as the case may be) in these equations with
discretized algebraic forms, which in turn are solved to obtain numbers for the flow
Exhaust
Supply slot
Exhaust
HIG. 1.22
Schematic of a room module with air supply and
exhaust ducts in the ceiling. (Affer Ref. 65.) COMPUTATIONAL FLUID DYNaMcs: waar Is ry 25
uy
PtP
TPT
8
x
x
‘
1
'
'
'
'
,
So aaa,
Taan
