Comparison of Austempering and Quench-And-Tempering Processes For

University of Windsor
Scholarship at UWindsor
Electronic Theses and Dissertations Theses, Dissertations, and Major Papers
2013
Comparison of Austempering and Quench-and-Tempering

Processes for Carburized Automotive Steels
Andrew Clark
Follow this and additional works at: https://scholar.uwindsor.ca/etd
Recommended Citation
Clark, Andrew, "Comparison of Austempering and Quench-and-Tempering Processes for Carburized
Automotive Steels" (2013). Electronic Theses and Dissertations. 4919.
https://scholar.uwindsor.ca/etd/4919
This online database contains the full-text of PhD dissertations and Masters’ theses of University of Windsor
students from 1954 forward. These documents are made available for personal study and research purposes only,
in accordance with the Canadian Copyright Act and the Creative Commons license—CC BY-NC-ND (Attribution,
Non-Commercial, No Derivative Works). Under this license, works must always be attributed to the copyright holder
(original author), cannot be used for any commercial purposes, and may not be altered. Any other use would
require the permission of the copyright holder. Students may inquire about withdrawing their dissertation and/or
thesis from this database. For additional inquiries, please contact the repository administrator via email
([email protected]) or by telephone at 519-253-3000ext. 3208.
by
Andrew Clark
A Thesis
Submitted to the Faculty of Graduate Studies
through Engineering Materials
in Partial Fulfilment of the Requirements for
the Degree of Master of Applied Science
at the University of Windsor
Windsor, Ontario, Canada
2013
© 2013 Andrew Clark

by
Andrew Clark
APPROVED BY:
Dr. W. Abdul-Kader
Department of Industrial and Manufacturing Systems Engineering
Dr. X. Sun
Chrysler LLC.
Dr. J. Sokolowski
Department of Mechanical, Automotive, and Materials Engineering
Dr. D.O. Northwood, Co-Advisor

Dr. R.J. Bowers, Co-Advisor

13 June 2013
DECLARATION OF CO-AUTHORSHIP / PREVIOUS PUBLICATION
1. Co-Authorship Declaration
I hereby declare that this thesis incorporates material that is the result of joint
research with Xichen Sun and Peter Bauerle of the Chrysler Headquarters and Test
Center, Michigan USA, and Derek O. Northwood and Randy J. Bowers of the University
of Windsor, Ontario. The research collaboration is covered in Chapter 4 of the thesis. In
all cases, the key ideas, primary contributions, data analysis, and interpretation were
performed by the author, and the contribution of co-authors was in the capacity of the
research in the form of technical advice and suggestions.
I am aware of the University of Windsor Senate Policy on Authorship and I

certify that I have properly acknowledged the contribution of other researchers to my
thesis, and have obtained permission from each of the co-authors to include the above
materials in my thesis.
I certify that, with the above qualification, this thesis, and the research to which it
refers, is the product of my own work.
2. Declaration of Previous Publication
This thesis includes one original paper that has previously been published in peer
reviewed conference proceedings, as follows:
Thesis Publication title/full citation Publication

Chapter status
Chapter 3,4 Clark, A. D., D. O. Northwood, R. J. Bowers, X. Published

Sun, and P. Bauerle. "Comparison of
Austempering and Quench-and-Tempering
Processes for Carburized Automotive Steels." SAE
International Journal of Materials and
Manufacturing 6 (2) (2013): 146-153.
iii
I certify that I have obtained permission from the copyright owners to include the
above published material in my thesis, and that the above material describes work
completed during my registration as a graduate student at the University of Windsor.
I declare that, to the best of my knowledge, my thesis does not infringe upon
anyone’s copyright, nor violate any proprietary rights and that any ideas, techniques,
quotations, or any other material from the work of other people included in my thesis,
published or otherwise, are fully acknowledged in accordance with the standard
referencing practices. To the extent that I have included copyrighted material that
surpasses the bounds of fair dealing within the meaning of the Canada Copyright Act, I
certify that I have obtained permission from the copyright owners to include such
materials in my thesis.
I declare that this is a true copy of my thesis, including any final revisions, as
approved by my thesis committee and the Graduate Studies office, and that this thesis has
not been submitted for a higher degree to any other university or institution.
iv
ABSTRACT
Austempering was examined as a replacement for the current quench-and-
tempering process as a method of heat treating carburized low alloy steel automotive
components. Three carburizing grade steels, SAE 8620, 4320, and 8822, were carburized
and heat treated by both processes. Twelve austempering and three quench-and-
tempering parameters were used. The effect of heat treatment on the case and core
microstructures was examined. Distortion was characterized using Navy C-ring samples,
which were measured before and after the carburizing and heat treatment process. X-ray
diffraction was used to measure residual stress and retained austenite. Charpy impact and
Rockwell C hardness testing were performed. Austempering produced improved
distortion and residual stress characteristics over quench-and-tempering, while
maintaining similar or improved mechanical properties. Full data sets for distortion and
mechanical properties were developed. Wear and fatigue testing are identified as
necessary next steps to fully examine the viability replacing the quench-and-tempering
process with austempering.
v
This thesis is dedicated to
George Henry Clark
vi
ACKNOWLEDGEMENTS
First, I would like to thank my advisors Dr. Derek O. Northwood and Dr. Randy
Bowers for their patience, support, and encouragement during the past two years. It is
through their invaluable guidance that I was able to learn the skills I need to succeed in
the workplace.
I would also like to thank Dr. Xichen Sun and Peter Bauerle from Chrysler LLC,
who provided both technical suggestions and coordinated the testing which took place at
the Chrysler Headquarters and Technology Center in Auburn Hills, Michigan. Special
thanks are also extended to Proto Manufacturing Ltd. in Oldcastle, Ontario, whose X-ray
diffraction services were generously provided.
Graciously acknowledged is the financial support of the Natural Sciences and
Engineering Research Council of Canada (NSERC) and the Ontario College-University
Consortium Council.
vii
TABLE OF CONTENTS
DECLARATION OF CO-AUTHORSHIP / PREVIOUS PUBLICATION .............. iii

ABSTRACT ................................................................................................................... v
DEDICATION..................................................................................................................vi
ACKNOWLEDGEMENTS .........................................................................................vii
LIST OF TABLES ......................................................................................................... x
LIST OF FIGURES ...................................................................................................... xi
LIST OF ABBREVIATIONS .................................................................................... xiii
I. INTRODUCTION ...................................................................................................... 1
1.1 Driving Force for Research .................................................................................... 2
II. LITERATURE REVIEW ......................................................................................... 3
2.1 Steel Microstructure ............................................................................................... 3
2.2 Alloying Effect on the TTT Curve ........................................................................ 10
2.3 Case Hardening .................................................................................................... 11
2.4 Heat Treatment Processes ..................................................................................... 12
2.5 Distortion Effects ................................................................................................. 15
2.5.1 The Navy C-Ring ........................................................................................... 16
2.6 Residual Stress ..................................................................................................... 17
2.7 Retained Austenite ............................................................................................... 21
2.8 Mechanical Properties .......................................................................................... 22
2.8.1 Hardness ........................................................................................................ 22
2.8.2 Toughness...................................................................................................... 23
III. EXPERIMENTAL DETAILS .............................................................................. 24
3.1 Material Selection ................................................................................................ 24
3.2 Heat treatment ...................................................................................................... 25
3.3 Optical Microscopy .............................................................................................. 26
3.4 Distortion Analysis ............................................................................................... 28
3.5 Residual Stress ..................................................................................................... 34
viii
3.7.1 Hardness ........................................................................................................ 36
3.7.2 Charpy Impact Testing ................................................................................... 37
IV. RESULTS & DISCUSSION.................................................................................. 38
4.1 Microstructure ...................................................................................................... 38
4.2 Distortion ............................................................................................................. 44
4.2.1 Size Distortion ............................................................................................... 44
Inner Diameter ................................................................................................... 45
Outer Diameter ................................................................................................... 48
Thickness ............................................................................................................ 51
Gap Width........................................................................................................... 53
Summary of Size Distortion ................................................................................. 56
4.2.2 Shape Distortion ............................................................................................ 57
Flatness .............................................................................................................. 57
Roundness ........................................................................................................... 59
Cylindricity ......................................................................................................... 61
Summary of Shape Distortion .............................................................................. 62
4.3 Residual Stress ..................................................................................................... 63
4.5.1 Hardness ........................................................................................................ 69
4.5.2 Toughness...................................................................................................... 72
4.6 Economic Discussion ........................................................................................... 76
V. CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK .......... 77
5.1 Conclusions .......................................................................................................... 77
5.2 Summary of Conclusions...................................................................................... 79
5.3 Recommendations for Future Work ...................................................................... 80
5.4 Unique Contribution of the Work ......................................................................... 81
REFERENCES ............................................................................................................ 82
PUBLICATIONS AND PRESENTATIONS .............................................................. 88
VITA AUCTORIS ....................................................................................................... 89
ix
LIST OF TABLES
Table 2.1 - Volume change associated with phase transformations [39]. ........................ 17
Table 3.1 - Compositions of Tested Steels ..................................................................... 25
Table 3.2 - Initial Heat Treatment Conditions ................................................................ 26
Table 3.3 - Heat Treatment Parameters Selected for Further Study................................. 27
Table 3.4 - Identification numbers of Navy C-rings ....................................................... 28
Table 3.5 - Residual Stress Measurement Depths for 4320 Steel .................................... 35
Table 3.6 - Retained Austenite Measurement Depths for 4320 Steel .............................. 36
Table 4.1 - Inner Diameter Distortion for 8620 Steel...................................................... 46
Table 4.2 - Outer Distortion for 8620 Steel .................................................................... 49
Table 4.3 - Thickness Distortion for 8620 Steel ............................................................. 51
Table 4.4 - Gap Width Distortion for 8620 Steel ............................................................ 54
Table 4.5 - Flatness Distortion ....................................................................................... 58
Table 4.6 - Roundness Distortion ................................................................................... 60
Table 4.7 - Cylindricity Distortion ................................................................................. 61
Table 4.8 - Surface Residual Stress Results for 4320 Steel ............................................. 64
Table 4.9 - Residual Stress Depth Measurements for 4320 Steel .................................... 65
Table 4.10 - Retained Austenite Measurement ............................................................... 68
Table 4.11 - Core Hardness............................................................................................ 70
Table 4.12 - Case Hardness............................................................................................ 71
Table 4.13 - Charpy Impact Toughness for 8620 Steel ................................................... 72
x
LIST OF FIGURES
Figure 2.1: The Iron-Carbon Phase Diagram [3]. ............................................................. 5
Figure 2.2: (A) Coarse pearlite and (B) Fine pearlite. [4]. ................................................ 6
Figure 2.3: TTT Curve for a eutectoid steel [4] ................................................................ 6
Figure 2.4: Martensite formed in core of 4320 steel carburized and austempered at
500 °F for 2 hours............................................................................................................ 8
Figure 2.5: (A) Lower bainite, with a needle like appearance. (B) Upper bainite,
with a more feathery appearance .................................................................................... 10
Figure 2.6: TTT curve for 4340 steel showing the bainite shelf [4] ................................ 11
Figure 2.7: Quench-and-temper process on a TTT curve [4]. ......................................... 13
Figure 2.8: TTT Curve showing three austempering heat treatments to produce
different amounts of bainite [4]. ..................................................................................... 15
Figure 2.9: A Modified Navy C-ring [31, 35]................................................................. 16
Figure 2.10: Typical residual stress depth profile found in carburized steel.
Reproduced from [44].................................................................................................... 18
Figure 2.11: Bragg's law demonstrated visually [52].......................................................19
Figure 2.12: Axes and Directions Used in X-Ray Diffraction [40, 53]............................ 21
Figure 3.1: 4320 steel carburized then austempered at 500 °F for (A) 30 minutes
and (B) 120 minutes. ..................................................................................................... 27
Figure 3.2: C-Ring Dimensioning Diagram [31, 35]....................................................... 31
Figure 3.3: Flatness Test Plot Produced by CMM .......................................................... 32
Figure 3.4: Inner Cylindricity Test Plot Produced by CMM ........................................... 32
Figure 3.5: Roundness Test Plot Produced by CMM. ..................................................... 33
Figure 3.6: C-Ring positioned for surface residual stress measurement (left) and
after electropolishing (right). ......................................................................................... 35
Figure 3.7: Dimensions of Type A Charpy Test Specimen from ASTM E23 [61]. ......... 37
xi
Figure 4.1: Austempered 8620 steel case microstructures. ............................................. 40
Figure 4.2: Austempered 4320 steel case microstructures. ............................................. 41
Figure 4.3: Austempered 8822 steel case microstructures .............................................. 42
Figure 4.4: Case and core microstructures for 4320 steel................................................ 43
Figure 4.5: Inner diameter distortion. ............................................................................. 48
Figure 4.6: Outer diameter distortion. ............................................................................ 50
Figure 4.7: Thickness distortion. .................................................................................... 53
Figure 4.8: Gap width distortion. ................................................................................... 56
Figure 4.9: Flatness distortion. ....................................................................................... 59
Figure 4.10: Roundness distortion. ................................................................................. 61
Figure 4.11: Cylindricity distortion. ............................................................................... 62
Figure 4.12: Surface residual stress for 4320 steel. ......................................................... 64
Figure 4.13: Surface residual stress and outer diameter distortion. ................................. 65
Figure 4.14: Residual stress and depth for austempered and quench-and-tempered
4320 steel. ..................................................................................................................... 66
Figure 4.15: Retained austenite with depth. .................................................................... 69
Figure 4.16: Core hardness. ........................................................................................... 70
Figure 4.17: Case hardness. ........................................................................................... 71
Figure 4.18: Charpy toughness results............................................................................ 74
Figure 4.19: SEM fracture surface images. .................................................................... 75
xii
LIST OF ABBREVIATIONS
NSERC Natural Sciences and Engineering Research Council of Canada
SAE Society of Automotive Engineers
CMM Coordinate Measurement Machine
TTT Time-Temperature Transformation
CCT Continuous Cooling Transformation
Q&T Quench and Tempering
ASTM American Society for Testing and Materials
DBTT Ductile-to-Brittle Transformation Temperature
ID Inner Diameter
OD Outer Diameter
SEM Scanning Electron Microscope
xiii
I. INTRODUCTION
Carburizing is a method of case-hardening low carbon steel to improve its wear
and friction characteristics. In this process low carbon steel is placed in a high carbon
atmosphere, or packed in a carbon rich material; the temperature is raised into the
austenitic phase region; and carbon diffuses into the surface layer of the material. This
carburized surface layer influences the microstructural transformations that occur upon
cooling. Typically, carburized steels are quenched to room temperature, forming a hard,
high carbon martensitic surface while maintaining a more ductile, low carbon content
core. The steel is then commonly tempered at intermediate temperature to increase the
toughness of the material.
Associated with the carburization and subsequent heat treatment processes,
dimensional changes occur in the part due to both phase transformations, as well as
thermal expansion and contraction. This distortion can cause parts to deviate from the
desired dimensions, which necessitates corrective measures, and hence another step in the
manufacturing process. Distortion is closely related to residual stresses, which are also
formed during transformation and non-uniform cooling. Typically compressive residual
stresses are formed in the case of carburized steel, which increase the fatigue life of a
component.
In the present study, austempering was considered as a possible replacement for
the quench-and-tempering of carburized steels. It has the potential to lower distortion,
which could eliminate the costly reworking step associated with traditional quench-and-
temper processes for carburized components.
1
1.1 Driving Force for Research
The purpose of this research project was to examine the viability of austempering
replacing quench-and-tempering as a method of producing carburized automotive parts. It
examined three common carburizing-grade steels currently used for roller bearings, gears,
and camshafts as well as the heat treatments used to produce them.
The effects of heat treatment on the case and core microstructures of carburized
SAE 8620, 8822, and 4320 steels were examined. A total of 12 austempering and 3
quench-and-tempering conditions were examined. Hardness for both case and core were
measured. Charpy V-notch testing was used to determine the toughness of the heat
treated steels.
Navy C-rings were machined and heat treated to examine distortion, residual
stress, and retained austenite levels during heat treatment. Distortion was measured using
a coordinate measurement machine (CMM). X-ray diffraction was used to measure the
residual stress and retained austenite levels in the heat treated samples.
The ability to remove processing steps such as distortion correction and
tempering provides an opportunity to reduce production cost for parts. For the
austempering process to be considered, it would be necessary for the mechanical
properties, distortion levels, and microstructural characteristics of the austempered
conditions to be comparable to those found for the current quench-and-tempered process.
2
II. LITERATURE REVIEW
The following literature review addresses the important aspects of this thesis
topic. Included are discussions on microstructure, heat treatment, and mechanical
properties of carburized heat treated steels, as well as a discussion of the austempering
and quench-and-tempering processes. Distortion, residual stress, retained austenite, and
the measurement of these factors, is also presented.
2.1 Steel Microstructure
Steel is a high strength iron-carbon alloy. It sees a wide variety of uses in the
automotive industry, where it is used for panels, structural members and driveline parts
[1]. The mechanical properties of steel can be affected by both its chemistry and the heat
treatments which are applied during manufacturing. Figure 2.1 shows the iron-carbon
phase diagram; where the weight percentage of carbon in the steel is on the horizontal
axis, ranging from 0% to 6.67%, while temperature on the vertical axis. Steels typically
have below 1% carbon, although steels with higher carbon contents up to 2% do see use
[2].
There are 4 phases which are shown on the phase diagram: α-ferrite, body-
centered-cubic structured iron which has low solubility for carbon; δ-ferrite, a high
temperature ferrite phase which has slightly higher solubility for carbon; austenite, a
face-centered-cubic structure that has much greater solubility carbon than ferrite; and
cementite, an intermetallic compound which is 6.67 weight% carbon and 93.3 weight%
iron. The eutectoid temperature, noted on the diagram at 727 °C, is the lowest
temperature at which it is possible to maintain austenite. Heating above this temperature
3
and holding will cause the steel microstructure to begin transform to austenite. The
eutectoid temperature is associated with the eutectoid composition, 0.77 weight% carbon.
Steels with greater than eutectoid carbon content are referred to as hyper-eutectoid, while
those with less than eutectoid carbon content are hypo-eutectoid.
At equilibrium cooling rates, the austenite formed at high temperature will
transform to α ferrite and cementite. For non-eutectoid steels, the austenite will begin to
transform above the eutectoid temperature. A hyper-eutectoid steel will begin forming
proeutectoid cementite during cooling, and will have a resultant structure of pearlite and
cementite. Hypo-eutectoid steels will form pro-eutectoid ferrite, and result in a structure
of ferrite and pearlite. As cooling rates increase the time for diffusion will decrease,
resulting in finer pearlite, and because of this the hardness, toughness, and strength of the
material will rise. Fine and coarse pearlite can be seen in Figure 2.2. With increased
cooling rates there is a method to predict the resultant microstructure. Time-Temperature
Transformation (TTT) curves and Continuous Cooling Transformation (CCT) curves
show how holding at temperature and cooling rate can affect final microstructure. Figure
2.3 shows the TTT curve for eutectoid steel. The lines on the graph represent
microstructural transformation points. There are start lines for the transformation to
pearlite, bainite, and martensite, as well as lines which show when complete
transformation of austenite to another phase can be expected.
4
Figure 2.1: The Iron-Carbon Phase Diagram [3].
5
Figure 2.2: (A) Coarse pearlite and (B) Fine pearlite. The darker portions of the diagram
are cementite, the lighter portions are ferrite [4].
Figure 2.3: TTT Curve for a eutectoid steel [4].
6
When the steel crosses a start line on the TTT graph the austenite formed at the
high temperature will begin to transform. Depending on the cooling rate, any of the start
lines can be crossed, and the respective microstructures will begin to form. Quenching
quickly below the martensite start temperature will result in the transformation
martensite, resulting in a strong and hard structure that is brittle [5]. Figure 2.4 shows a
martensitic structure. The martensitic transformation involves the changing of the crystal
structure to body centered tetragonal, which stretches the vertical axis of the unit cell in a
shear process, going from a face-centered cubic structure to a body-centered tetragonal
structure [6]. This stretching effect results in stresses formed in the material, and in
severe cases can lead to what is known as “quench cracking” [2, 5]. Two main types of
martensite form, based on carbon content. Lath martensite forms in steels of composition
up to 0.6 weight% carbon, while plate martensite forms in steels over 1 weight% carbon.
It is possible to form mixes of lath and plate martensite in steels between 0.6 wt% and 1
wt% carbon [2].
Several equations have been produced to calculate the martensite start
temperatures based on the specific effect that each alloying element has on the start
temperature [7, 8]. Equation 2.1 is the formula for martensite start temperature, first
presented in 1965 by Andrews [9]. It is of particular use for steels which have low carbon
content. Other equations have been determined for use with steels with lower carbon
contents, or alloyed with different alloying elements [7, 8]. One such equation, for carbon
contents up to 0.02 wt%, proposed by Liu et al, is given in Equation 2.2 [8].
Equation 2.1
7
Equation 2.2
Figure 2.4: Martensite formed in core of 4320 steel carburized and austempered at 500
°F for 2 hours
Bainite is a non-lamellar mix of ferrite and cementite formed at intermediate
cooling rates. There are two types of bainite, upper and lower bainite, which form
depending on the temperature ranges used. They can be identified by their different
appearances, Figure 2.5. Lower bainite has a more needle like appearance, and forms
with a shearing mechanism similar to martensite, although there are some diffusion
effects present as well [2, 10]. Upper bainite has a more feathery appearance, and is
produced predominantly by diffusion mechanisms. Lower bainite is formed upon cooling
to a temperature just above the Ms temperature and holding, while upper bainite can be
formed by cooling past the pearlite shelf of the TTT curve and holding at temperature.
8
Lower bainite tends to have a higher tensile strength and hardness than upper bainite
[11].
Much like martensite there have been attempts to characterize the bainite start
temperature. Work by Zhao et al. produced Equation 2.3, a bainite start temperature
equation for low alloy steels [12, 13]. It is assumed that the bainite start temperature
marks the “knee” of the cooling curve that has to be passed before bainite can be formed.
This equation does not take into effect the necessary cooling rate required to avoid
pearlite formation. Using bainite start temperature equations as a guide it is possible to
approximate the TTT curve for a given steel using empirical testing to find the times
required for transformation [14]. Upon crossing the bainite start line on the TTT curve,
bainite will begin to form. If the transformation process is interrupted by a quench it is
possible to only partially transform the austenite to bainite. The austenite which has yet to
transform will transform to martensite upon quenching; any untransformed austenite will
remain trapped in the bainite matrix as retained austenite at room temperature.
Equation 2.3
9
Figure 2.5: (A) Lower bainite, with a needle like appearance. (B) Upper bainite, with a
more feathery appearance
2.2 Alloying Effect on the TTT Curve
It is not possible with a plain carbon steel to design a heat-treatment schedule to
form bainite and martensite from austenite. The proeutectoid and pearlite start lines on
the TTT curve do not allow enough time to quench to an austempering temperature or the
martensite start temperature before the proeutectoid and eutectoid reactions begin. Hence
any microstructure would contain a finite amount of pearlite and a proeutectoid phase.
The hardenability, or ease of forming martensite, of steel can be increased by adding
alloying elements to the steel [15, 16]. These alloying elements shift the pearlite start line
to longer times, allowing sufficient cooling time to allow the formation of bainite and
martensite. Figure 2.6 shows a TTT curve for 4340 steel, an alloy steel of nickel and
chrome.
10
Figure 2.6: TTT curve for 4340 steel showing the bainite shelf [4]
2.3 Case Hardening
Case hardening describes a series of processes where in the outer layer of a
component is provided increased hardness. It is desirable where components will be
subject to friction and wear. There are several methods of case hardening steel; such as
carburizing, carbonitriding, nitriding, laser surface hardening, and induction hardening
[17, 18]. Carburizing is most often performed using a two stage gas carburizing process
known as boost diffuse processing; it requires less time than the traditional single stage
carburizing [19]. The first stage, the boost, is carried out by placing the component into a
high temperature furnace, typically between 1550 °F and 1750 °F [2, 5, 20], with a high
carbon potential atmosphere. The carbon potential is higher than the final desired carbon
11
content of the material, which “boosts” the carbon content in the surface layer of the
material to close to the desired amount. In the second stage, the carbon potential of the
atmosphere is lowered to the desired carbon potential and held at a slightly lower
temperature [21, 22]. The higher carbon content at the surface then diffuses farther into
the lower carbon content areas of the case [2, 20]. The carburized layer offers additional
hardness on the surface, while maintaining for a tougher, lower-carbon, core [16].
Of all the alloying elements in steel, carbon has the largest effect on hardenability.
Hence it is possible to form a different microstructure at the surface of a case carburized
component compared to the microstructure found in the core. Hardenability can also be
affected by prior austenite grain size. Grain size can be increased by using a higher
temperature for austenitizing, or, in case hardening, a higher carburizing temperature
[23].
2.4 Heat Treatment Processes
The heat treatment process used to bring the component from the high furnace
temperature to room temperature can affect the final properties of the material. After
carburizing, a steel can be slow cooled, then reheated to austenitize it before performing a
final heat treatment [24]. Quench and tempering (Q&T) is currently the most common
method of heat treating case carburized parts [25]. In the quench and temper process the
parts are directly oil quenched from the carburizing process to form martensite in both the
case and core, and then reheated to temper the martensite, thereby increasing the ductility
and toughness. Typical tempering temperatures are in the range of 300 to 400 °F (149 to
205 °C), although these can be considerably higher depending on what type of tempering
12
is being undertaken [17, 26]. A TTT curve showing the quench and tempering process is
shown in Figure 2.7. Upon full transformation to martensite, any subsequent tempering
will not result in the transformation to bainite or pearlite. The tempering reaction allows
carbon from the super saturated martensite lattice to form cementite particles at the prior
austenite grain boundaries or between the martensite laths and plates. It is the lowering of
the carbon content in the martensite that results in the increase in ductility and toughness.
Figure 2.7: Quench-and-temper process on a TTT curve [4].
The tempering process generally allows for an increase of ductility and toughness.
However, depending on the temperature ranges selected, it is possible to make the
material more brittle. Tempered martensite embrittlement is a phenomenon where at
certain ranges of temperature, generally between 500 and 700 °F. It is possible for
13
alloying elements to not only reduce the effectiveness of a temper, but make the material
more brittle [2].
Austempering is a different heat treatment process. After carburizing, the part is
quenched to an intermediate temperature and held to allow the formation of bainite
within the carburized case of the sample [27]. Bainite is able to form in the carburized
case because its high carbon content has lowered the temperature to below the hold
temperature. Typically, salt baths are employed as a quenchant for austempering, and the
part is washed with water afterwards. Reclamation of both salt and water can allow for
austempering to produce nearly no waste, although there are costs associated with the
reclamation process [28]. As well, austempering can remove the need for additional
distortion-correcting treatments, associated with quenched parts, to prepare the material
for service; as well it reduces the chance of hydrogen embrittlement [29, 30]. Figure 2.8
shows an austempering process on a TTT curve. Line A shows an interrupted
transformation and would result in a mostly martensitic and retained austenite structure
with some bainite, line B shows an interrupted transformation which would result in a
mostly bainitic structure with some martensite and retained austenite, and line C shows a
transformation which would yield an entirely bainitic microstructure.
14
Figure 2.8: TTT Curve showing three austempering heat treatments to produce different
amounts of bainite [4].
2.5 Distortion Effects
Distortion, or non uniform volume change, occurs during the quenching of parts
as a result of both phase transformations, as well as thermal expansion and contraction.
Distortion is divided into two main categories, size and shape distortion. Size distortion is
most commonly associated with thermal expansion and contraction, while shape
distortions tend to arise due to local temperature differences and differences in section
size during cooling [31]. Although it is possible to predict with some accuracy the size
distortion that will take place during heat treatment, it is difficult to predict the shape
distortions which will occur, because they depend on local variations in cooling rate.
Distortion results in a need for post-heat-treatment grinding to restore the part
15
dimensions. If distortion is severe enough, parts have to be scrapped [32]. Dimensional
correction can cause weakening of a case carburized material, as well as increase
production times and costs [33, 34]. As such reduction in distortion becomes an
important aspect of part design. By changing the heat treatment method, distortion can be
reduced. In this study, distortion is measured with a coordinate measurement machine
(CMM) by comparing dimensions of a Navy C-ring sample both before and after case
carburizing and heat treatment.
2.5.1 The Navy C-Ring
The Navy C-ring has both thick and thin sections, which allows for the shape of
the sample to change during heat treatment. It is a standard shape used to measure the
effects of distortion during processing; its use has been well documented to measure the
distortion effects associated with heat treatments [31, 35, 36, 37, 38]. A modified Navy
C-ring is shown in Figure 2.9.
Figure 2.9: A Modified Navy C-ring [31, 35].
16
There are seven standard measurements made on the Navy C-ring, divided into
size and shape distortions. Size distortions relate to the changing of major dimensions.
They include the inner and outer diameters, thickness of the sample, and are most often
caused by volume changes associated with phase transformations. The quenching of
austenite to form martensite results in a positive volume change of 4.63% for a 0.02 wt%
C steel [39]. Table 2.1 shows volume change resulting from steel phase transformations
as a function of carbon content. Shape transformations are most often associated with
both heat treatment variables, as well as thermal expansion and contraction through parts
of varying thickness. The result is a deviation from the desired angles and shapes of the
part. Shape distortions are quantified on the Navy C-ring by using the flatness, roundness,
cylindricity, and gap width measurements.
Table 2.1 - Volume change associated with phase transformations [39].
Transformation Volume Change %

Pearlite -> Austenite -4.64 + 2.21x(C%)
Austenite -> Martensite 4.64 - 0.53x(C%)
Austenite -> Lower Bainite 4.64 - 1.43x(C%)
Austenite -> Upper Bainite 4.64 - 2.21x(C%)
2.6 Residual Stress
Similar to distortion, residual stress is caused by both phase transformations, as
well as thermal expansion and contraction; it can be affected by the heat treatment
conditions placed upon a material [35, 40]. As steel is cooled, different sections of the
material will transform at different temperatures, most often due to variances in
composition [41]. Residual stresses are formed as sections of the material change size at
17
different rates. Typically with carburized steels, the residual stress is compressive at the
surface, and increases in magnitude immediately below the surface. Farther into the
material, the stress returns to surface levels, becomes tensile, then approaches zero [39,
42, 43]. A typical distribution of residual stress into the carburized case of a steel is
shown in Figure 2.10 [44].
Figure 2.10: Typical residual stress depth profile found in carburized steel. Reproduced
from [44].
Residual stresses can be beneficial or detrimental to performance depending on
their nature. Compressive residual stresses are preferred over tensile ones, because they
can help improve wear and fatigue resistance [45, 46]. Should any cracks form in the
material, the compressive stresses have the ability to relax and prevent crack growth.
Attempts have been made to correlate the exact effect of residual stress on fatigue life;
however, no studies have been able to do so conclusively [47]. Fatigue life is of particular
importance in the production of gears, and austempering has been shown to increase the
fatigue life of a material over traditional quench-and-tempering processes [48].
18
There are several methods of measuring residual stress: destructive methods such
as inferometric strain rosettes with hole drilling; and non-destructive methods, including
x-ray diffraction [49]. X-ray diffraction measures the stresses within a material by
measuring the crystallographic lattice spacings and comparing them to an unstressed
sample [50]. Typically, this unstressed sample is a powder which has been furnace
annealed to allow the stresses present in the material to relax [40]. The strain is calculated
using Equation 2.4, where is the lattice spacing of the unstressed powder.
Equation 2.4
The lattice planes diffract the x-ray beam when the beam hits the material, which
causes a change in the peak intensity of the diffracted beam [40, 51]. It is through the use
of the wavelength, angle of refraction, and a constant near unity that the lattice spacing,
d, is calculated. This is shown in Equation 2.5, known as Bragg’s Law [51]. It is
demonstrated visually in Figure 2.11, where the X-ray beams are the dashed lines, and
the horizontal dotted lines are the lattice.
Equation 2.5
Figure 2.11: Bragg’s law demonstrated visually [52].
19
The method involves taking a larger number of d spacing measurements
and accounting for the angle of the hkl Miller index planes found in the lattice. In
Equation 2.6 the and terms are x-ray elastic parameters of the material, is the
stress in the direction of measurement, ψ is the angle subtended by the bisector of the
incident and diffracted beam and the surface normal, and is the strain associated with
a particular ψ angle [53].
Equation 2.6
This equation allows for the calculation of stress within a material without having
to use the annealed powder method to get the zero strain value [40]. Figure 2.12 shows
the orientation of the various stress components. The method allows for more
accuracy in measurement with increased collection time and more ψ angles. For
examination of residual stresses at greater depths, material must be removed, because x-
ray diffraction is not able to penetrate deeply into a material [54]. One method used to
remove material is electro polishing; however, corrections must be made to the
measurements to account for the material removed by the electropolishing process [55].
20
Figure 2.12: Axes and Directions Used in X-Ray Diffraction [40, 53].
2.7 Retained Austenite
Retained austenite occurs when there is an incomplete transformation to a new
structure during the quenching of a steel. It is a common occurrence in case hardened
steels, as the additional carbon content in the case of the part lowers the martensite finish
temperature to such a point that complete transformation is difficult to achieve [46, 56].
The low temperature, combined with a rapid quench can result in even more retained
austenite at the end of the quenching process. It is often seen that there can be between
20% and 30% retained austenite in carburized steel at the surface [24]. Retained austenite
levels increase with carbon content [10].
Retained austenite levels are also commonly measured using X-ray diffraction
techniques, although other measurement techniques exist [57]. When using X-ray
diffraction to measure retained austenite levels the “four peak method” is most commonly
used. This method measures the intensities of the {211} and {200} families of planes in
21
the martensite, and compares them to the {200} and {220} families of planes in the
austenite [58].
Retained austenite can be further transformed by the application of either strain or
heat treatment. Strain application allows the shear process to take place, and results in a
martensitic microstructure. Heat treatment, such as the tempering process used in the
production of quench-and-tempered parts, will result in the retained austenite
decomposing into bainite, either upper or lower bainite depending on the temperature.
Retained austenite can also allow for benefits in the fatigue life of a carburized steel, as
the retained austenite is associated with compressive residual stress, and in higher strain
cases, the transformation of retained austenite through deformation accommodates some
of the strain [59].
2.8 Mechanical Properties
2.8.1 Hardness
Hardness is measured using a variety of scales depending on material and
expected hardness. It can be measured on both the micro and macro level, with separate
scales existing for each. Case hardness can be used to get a rough estimate of both
strength and wear resistance of a case hardened material [29]. Core hardness can give an
indication of the toughness that can be expected for a material. Hardness testing on the
Rockwell scale is outlined in ASTM [60].
22
2.8.2 Toughness
Toughness is a measure of the energy absorbed by a material before fracture.
ASTM Standard E23-12c, Standard Test Methods for Notched Bar Impact Testing of
Metallic Materials, outlines two types of impact tests, the Charpy and Izod tests. Most
commonly used in North America is the Charpy V-notch test [61], which uses a notched
specimen. Other notch geometries are available, as outlined in the ASTM standard.
Charpy testing is performed by using a weighted pendulum to break notched sample.
When the pendulum swings and breaks the sample, the height to which it swings on the
opposite side is recorded, and the energy absorbed by the sample is measured.
For carburized parts, Charpy toughness is often recognized as a function of the
core microstructure. In steels that had not been carburized, it was demonstrated that
austempering to form a lower bainitic structure produced better toughness than quench-
and-tempering [25].
Impact energy can be heavily influenced by temperature as well as heat treatment.
The ductile-to-brittle transformation temperature (DBTT) is a temperature at which a
steel will go from having lower-energy fracture to higher-energy fracture [2, 61]. Ideally,
a material will have a lower DBTT than the temperature at which it will be in service.
Nickel and molybdenum have the ability to increase toughness and lower the DBTT. An
increase in carbon content can greatly lower toughness and increase the DBTT [62]. With
Charpy impact tests, it is possible to examine the method of failure, similar to the way a
tensile test failure mode can be determined. Ductile fracture is associated with increased
toughness; brittle fracture is associated with decreased toughness.
23
III. EXPERIMENTAL DETAILS
Austempering and quench-and-tempering processes were compared in three
different carburizing grades of steel by means of microstructural, mechanical properties,
and distortion properties testing. Twelve initial austempering parameters were reduced to
7 after microstructural examination of the case. Further micrographs were produced for
both the case and core of the quench-and-tempered conditions, as well as the remaining
austempering conditions. Hardness measurements were recorded for both the case and the
core of both conditions. Charpy impact testing was used to determine the fracture energy
of the materials. Distortion resulting from heat treatment was measured using a
coordinate measurement machine (CMM). Residual stress and retained austenite
measurements were made using X-ray diffraction.
3.1 Material Selection
Three steels were examined in this study, SAE 8620, 8822, and 4320. Their
chemical compositions are shown in Table 3.1. They were selected for testing based on
their current, frequent use in the carburized state. All three steels are common carburizing
grades, varying in hardenability, strength, and toughness, and seeing use for gears, roller
bearings, and piston pins. 8620 steel features good hardenability, medium strength, and,
when case hardened, good wear resistance. It is the most commonly used of the three
steels examined. 8822 steel has good case hardenability and core hardness, as well as
higher strength than 8620 steel due to its higher carbon content [1]. 4320 steel is a nickel-
chrome-molybdenum steel; it has a higher nickel content compared to the other two
materials, and generally higher toughness. Martensite start temperatures were calculated
using the formula of K.W. Andrews, Equation 3.1, and converted into Fahrenheit for both
24
the case and core. temperatures for the carburized cases were calculated based on 0.9
wt% carbon [9].
Table 3.1 - Compositions of Tested Steels
Steel wt% C wt% Mn wt% Si wt% Cr wt% Ni wt% Mo

Core Case
8822 0.22 0.96 0.17 0.56 0.34 0.33 754 °F 236 °F
8620 0.2 0.91 0.23 0.5 0.56 0.16 769 °F 236 °F
4320 0.19 0.64 0.2 0.47 1.74 0.22 753 °F 213 °F
Equation 3.1
3.2 Heat treatment
Carburization was performed using gas carburizing at a temperature of 1650 °F
for five hours, with a carbon potential of 0.9. Salt baths were used to austemper the
samples, at one of three temperatures: 500, 550, or 580 °F; for one of four times: 30, 60,
120, or 240 minutes. There were twelve time-temperature combinations. Quench and
tempered samples were oil quenched to room temperature, then reheated and tempered at
either 650 or 750 °F for one hour. These temperatures were selected to demonstrate the
upper range of tempering. Table 3.2 provides an overview of the heat treatment
parameters.
25
Table 3.2 - Initial Heat Treatment Conditions
Treatment
Temperature (°F) Time (min)
Method
30
60
500
120
240
30
60
Austemper 550
120
240
30
60
580
120
240
Quench & 650 60

Temper 750 60
3.3 Optical Microscopy
Optical microscopy was used to examine the amount of bainite in the case of the
austempered samples. Samples were mounted in diallyl pthalate using a Buehler
Simplimet mounting press. Grinding was performed using a Buehler Handimet 2 with
silicon carbide papers of 240, 320, 400, and 600 grit. After grinding, the samples were
rough polished with a 9-micrometer diamond paste, followed by fine polishing with 1.0-
and 0.05- micrometer aluminium suspensions. Etching was accomplished using a 2%
nital solution. Austempered microstructures showed a mixture of bainite and martensite
in the case and martensite in the core. Quench-and-tempered samples show martensite in
both the case and core. A minimum limit of 50% bainite in the case was set to determine
26
the austempering conditions for further study. Figure 3.1 shows an example of sufficient
and insufficient bainite in a microstructure. Table 3.3 identifies the final selected heat
treatments.
Figure 3.1: 4320 steel carburized then austempered at 500 °F for (A) 30 minutes and (B)
120 minutes. The darker phase is identified as bainite. The lighter phase is martensite.
There was less than 50% bainite in (A) so further testing was not warranted.
Table 3.3 - Heat Treatment Parameters Selected for Further Study

Treatment Parameter
Temperature (°F) Time (min) ID
Method
120 500/2
500
240 500/4
120 550/2
550
Austemper 240 550/4
60 580/1
580 120 580/2
240 580/4
Quench & 650 60 QT650

Temper 750 60 QT750
27
3.4 Distortion Analysis
Navy C-ring samples were employed to characterize distortion. Two C-rings were
machined for each austempering condition as well as for each quench-and-temper
condition. Identification numbers can be found in Table 3.4.
Table 3.4 - Identification numbers of Navy C-Rings

Material 8620 8822 4320
Code A B C
Temperature Parameter
Heat Treatment Time (h) Number
(F) ID
13
500 2 500/2
14
3
500 4 500/4
4
11
550 2 550/2
12
5
Austemper 550 4 550/4
6
7
580 1 580/1
8
9
580 2 580/2
10
1
580 4 580/4
2
17
650 1 QT650
Quench & 18
Temper 15
750 1 QT750
16
Distortion analysis was performed using a coordinate measurement machine
(CMM). Measurements of the C-ring specimens were made both before the carburizing +
heat treating process, as well as after. The equation used to calculate the distortion
amount is shown in Equation 3.2, where is the measured value before carburizing+
heat treating, and is the measured value after carburizing + heat treatment. Averages
28
were calculated, and standard deviation was calculated using Equation 3.3 where σ is the
standard deviation.
Equation 3.2
Equation 3.3
Seven types of distortion were measured, four size and three shape distortions.
Size distortions are inner diameter (ID), outer diameter (OD), thickness, and gap width.
Shape distortions are flatness, roundness, and cylindricity. A diagram of the Navy C-ring
showing nominal dimensioning values, as well as measurement locations for the different
distortion types, is given in Figure 3.2.
Size distortion measurements were taken as follows. Diameters were measured
using diametrically opposed points on the inner and outer circumferences at both of the
flat surfaces of the C-ring, as well as on the plane midway between the two. Thickness
was measured at five points on the outer diameter, as the distance between the two flat
surfaces of the C-ring. Each of the five measurements is reported. Gap width is a
measurement of the distance between the two surfaces at the gap cut at the thinnest
section; it was measured along both vertical edges as well as at the centre of the gap.
Shape distortion measurements were taken as follows. Approximately 2400
flatness measurements were performed along the edges of the surface of the C-ring which
was marked with the specimen identification code. The difference between the highest
and lowest points was the reported flatness value. A sample test plot from the CMM
29
machine is shown in Figure 3.3. The small circles on the test plot indicate the highest and
lowest points on the edges of the surface
Cylindricity measures both the inner and outer diameters for their consistency in
the axial direction. 2800 points were used to measure the ID; 4500 points were used to
measure the OD. ID and OD were measured at both flat surfaces as well as the sample
mid-height. The difference between the innermost and outermost points gives the
cylindricity measurement. Figure 3.4 shows the output plot from a CMM machine for
inner cylindricity. The small circles on the plot indicate the innermost and outermost
points relative to the black dot, which marks the central axis of the circle being measured.
Roundness measurements were made on the inner and outer diameters at the flat
surfaces and mid-height. The largest and smallest radii were recorded, and then the
difference reported as the roundness value for that plane of measurement. As a result, 6
measurements are reported, one for each plane measured for both the inner and outer
diameter. A test plot from the CMM is shown in Figure 3.5. The plot for roundness
shows the three measurements, scaled so as to not overlap. The blue circles indicate the
actual measurement, with the red marks on them noting the innermost and outermost
points.
30
Figure 3.2: C-Ring Dimensioning Diagram [31, 35].
31
Figure 3.3: Flatness Test Plot Produced by CMM – Circles indicate the highest and
lowest points of deviation from the ideal plane.
Figure 3.4: Inner Cylindricity Test Plot Produced by CMM – Circles indicate the
innermost and outermost points relative to the central axis.
32
Figure 3.5: Roundness Test Plot Produced by CMM – Plot displays all three planes of
measurement. Measurements are scaled to display all 3 measurements without
overlapping.
33
3.5 Residual Stress
Residual stress measurement was carried out using X-ray diffraction techniques.
CrKα radiation of wavelength 2.291 Angstroms was used, at a power of 30.00 kV, 25.00
mA. The 211 family of planes were measured, with the Bragg angle held at 156.4°.
Eleven psi angles (ψ) were used, at 0, ±25.00, ±20.59, ±15.83, ±11.84, and ±3.71
degrees. An aperture of 1 mm was used. Calibration was performed in accordance with
ASTM E915, using iron powder for both surface and depth testing [50].
Figure 3.6 shows a C-ring positioned for residual stress measurement on the
surface, and subsurface at the “R” point identified in Figure 3.2. Residual stress
measurements were made on one C-ring from each austempered condition, and one from
each quench-and-tempered condition for the 4320 steel. 4320 steel was selected for
testing based on its distortion and mechanical properties performance, the results of
which are covered in chapter 4. Additionally, residual stress depth profile testing was
performed on two of the 4320 steel C-rings; one austempered at 500 °F for 4 hours; the
other quench-and-tempered at 750 °F for one hour. For depth testing, electro polishing
was used to remove surface layers to reach the desired depths. Measurements were taken
at 10 depths, with a maximum depth of 0.1 inches. Exact measurement depths for both C-
rings are shown in Table 3.5.
34
Figure 3.6: C-Ring positioned for surface residual stress measurement (left) and after
electropolishing (right).
Table 3.5 - Residual Stress Measurement Depths for 4320 Steel
Condition 500/4 QT750

0.0000 0.0000
0.0006 0.0060
0.0011 0.0012
0.0030 0.0028
Measurement 0.0052 0.0048
Depths (in) 0.0103 0.0107
0.0200 0.0235
0.0454 0.0403
0.0666 0.0686
0.1016 0.0997
Similar to the residual stress profile, a depth profile of retained austenite levels
was produced for one of the two 4320 steel 500/4 samples, and one of the two QT750
samples. Samples were chosen based on their distortion and residual stress results. X-ray
diffraction was used to determine the amount of retained austenite to a depth of 1/16 th of
an inch. Exact measurement depths are shown in Table 3.6. A beam of CrKα radiation
was used with a wavelength of 2.291 Angstroms. Power levels were 30.00 kV, 25.00 mA.
35
The four peak method was used to measure the retained austenite. Electro polishing was
used to remove surface material to expose the surface underneath. Retained austenite
measurements were carried out at certain depths in conjunction with residual stress
measurement.
Table 3.6 - Retained Austenite Measurement Depths for 4320 Steel
Condition 500/4 QT750

0.0000 0.0000
0.0011 0.0012
Measurement 0.0052 0.0048
Depth (in) 0.0103 0.0107
0.0200 0.0235
0.0666 0.0686
Mechanical properties testing consisted of hardness testing and Charpy impact
testing.
3.7.1 Hardness
Case and core hardness were tested on all samples using a Wilson Rockwell
hardness tester, verified in accordance with ASTM E18, Standard Test Methods for
Rockwell Hardness of Metallic Materials [60]. Five measurements were taken for both
the case and core using a diamond cone with a 150 kilogram load. For each measurement,
the highest and lowest results were eliminated, and the average of the remaining three
measurements was recorded.
36
3.7.2 Charpy Impact Testing
Charpy impact testing was performed on all samples using Type A Charpy
(simple beam) impact test specimens, machined in accordance with the dimensions in
ASTM E23. Specimens were machined, then carburized and heat treated. A diagram of
the Type A Charpy test specimen is given in Figure 3.7 [61]. Three Charpy test
specimens were produced for each austempering condition, as well as for the two quench
and tempered conditions. Additionally two Charpy specimens were produced which were
carburized, oil quenched, then tempered at 350 °F for one hour to simulate the currently
used carburization with quench-and-temper process. The fracture surfaces of selected
samples were examined using a scanning electron microscope (SEM).
Figure 3.7: Dimensions of Type A Charpy Test Specimen from ASTM E23 [61].
37
IV. RESULTS & DISCUSSION
A comparison of the austempering process with the quench and tempering process
was accomplished for three carburized steel compositions. Microstructural analysis of the
effects of tempering time and temperature was undertaken to explain the effect on
mechanical properties. Distortion is quantified by an examination of Navy C-Ring
dimensions taken before carburizing and after the subsequent carburizing and heat
treatment processes. X-ray diffraction was employed to quantify residual stress and
retained austenite both at as well as beneath the surface of the Navy C-Rings. While the
austempering process was generally found to be equivalent or better than the traditional
quench-and-tempering process, the issues of wear performance and a more quantitative
focus on the economics of process substitution remain.
4.1 Microstructure
The case microstructures found in 8620 steel after the initial carburizing and
austempering heat treatment processes are documented in Figure 4.1. For each
temperature-time combination, Table 3.2, bainite was noted to form. For the 30-minute
processes, complete transformation to bainite did not occur; the remainder of the
microstructure was martensite.
A level of 50% bainite was identified as being necessary to obtain the similar or
improved mechanical properties compared to the current quench-and-temper condition.
As such, the initial austempering trials were able to eliminate all of the 30-minute
austempering conditions for all three steels: 8620, 4320, and 8822, as well as the 60-
minute conditions for 500 and 550 °F. Figures 4.2 and 4.3 document the microstructures
38
for the 4320 and 8822 steels, respectively, that were noted to contain a minimum 50%
bainite.
For the 8620 steel, the type of bainite formed depended on the temperature. The
more feathery appearing upper bainite was noted at 580°F; acicular lower bainite was
noted at 500°F. Hence, the TTT curve for this steel would exhibit a bainite curve with its
nose between 580 and 500°F. Longer times at lower temperatures resulted in carbide
precipitation at the edges of the bainite needles, Figures 4.1.G and 4.1.J.
The microstructures of the austempered 4320 steel, Figure 4.2, exhibited upper
bainite across the temperature range. As such, the TTT curve for this steel would have its
bainite nose situated below 500°F. For the 8822 steel, the situation was similar to the
8620 steel, with upper bainite present at 580°F and lower bainite at 500°F. The effect of
microstructure on mechanical properties is discussed in section 4.4.
39
Figure 4.1: Austempered 8620 steel case microstructures.
40
Figure 4.2: Austempered 4320 steel case microstructures.
41
Figure 4.3: Austempered 8822 steel case microstructures
Both case and core microstructures are documented in Figure 4.4 for the quench-
and-temper processes in 4320 steel. At both the 650 and 750°F tempering temperatures,
the case and core microstructures exhibit tempered martensite. The 0.19 weight %
carbon core shows lath martensite, which is formed at compositions below 0.6 weight %
carbon [2]. The transition between plate and lath martensite occurs between 0.6 and 1-
weight % carbon. At 0.9 weight % carbon, the case microstructure exhibits a mix of
both lath and plate martensite. The lower carbon content martensite at the core is
42
expected to be less brittle than the case martensite. For toughness testing, the core
microstructure represents the majority of the fracture surface and most affects the
absorbed energy value. Included in Figure 4.4 are the case and core microstructures for
4320 steel austempered at 500°F for four hours. The case microstructure shows nearly
100% bainite. The core microstructure is martensite. Lath martensite is expected due to
the low carbon content. The 500°F temperature is below the temperature; therefore
any marteniste present at 500°F would be tempered for 4 hours at 500°F. No data was
found on the temperature for 8822 steel. As such, any remaining austenite at 500°F
could either decompose to bainite or quench to martensite at the end of the 4-hour
process. The resulting hardness values of the core microstructure are presented in section
4.5.1.
Figure 4.4: Case and core microstructures for 4320 steel.
43
4.2 Distortion
The results of the Navy C-ring distortion testing are given in section 4.2.1 for size
distortion, and 4.2.2 for shape distortion. For each measurement parameter the
dimensions used to calculate the distortion are given. All distortion measurements are
given as a percentage change regardless of whether the change was positive or negative,
in order to highlight the magnitude of change in dimension. The small number of C-rings,
two per heat treatment condition, limited any statistical analysis of the data. Nonetheless,
an average of the two C-rings is provided for the distortion results; and a standard
deviation is calculated using Equation 3.3. The small sample size is based on the
preliminary nature of this study, which is meant to provide a proof of concept for future
work. Multiple measurement points on each C-ring for the different distortion parameters
helped increase the statistical significance of the results.
4.2.1 Size Distortion
C-ring size distortion results are given in Tables 4.1 - 4.4, and Figures 4.5 - 4.8.
The distortion values were calculated by averaging the values of distortion measured for
the two C-ring samples heat treated by the same condition. For the inner (ID) and outer
(OD) diameter measurements the percentage change of the maximum values is presented,
along with the average change of the three measured values. Thickness distortion is
presented as a single value for distortion, calculated by averaging the change in thickness
at the 5 measurement locations on the C-ring. Gap width distortion is presented as an
average of the change in gap at the three measurement locations.
44
Inner Diameter:
Inner diameter results for the 4320 show that the austempering heat treatments
resulted in lower values that the quench-and-temper heat treatment. The same is true of
the 8822 results; indeed the 8822 steel exhibited the lowest inner diameter distortion
among the steels tested. For the 8620 steel, the inner diameter distortion values varied
within the austempering conditions.
The most consistent results for inner diameter distortion occurred for the 580°F
heat treatments. At this temperature, the distortion decreases monotonically with time for
all three steels. For all three steels, the microstructure exhibited was upper bainite. This
consistency is not noted for the other temperatures. In relation to the metallurgy of steels,
upper bainite is formed more by a diffusion process than a shear transformation. As such,
the greater amount of diffusional transformation product with increasing time would
correlate to lower distortion. For all materials, the middle-location measurement for inner
diameter exhibited less distortion than the top and bottom surface measurements. This
lower measurement may demonstrate a possible lack of freedom for expansion and
contraction compared to the free surfaces.
45
Table 4.1a - Inner Diameter Distortion for 8620 Steel
Inner Diameter Distortion % Condition

Condition Sample #
Top Middle Bottom Average Distortion
A13 0.2898 0.2119 0.2927 0.2648
500/2 0.2855
A14 0.3342 0.2575 0.3270 0.3063
A3 0.3127 0.2309 0.3026 0.2821
500/4 0.2727
A4 0.2792 0.2237 0.2869 0.2633
A11 0.1809 0.1048 0.1623 0.1493
550/2 0.1591
A12 0.1861 0.1243 0.1962 0.1689
A5 0.2529 0.1948 0.2696 0.2391
550/4 0.2443
A6 0.2609 0.2084 0.2791 0.2495
A7 0.3251 0.2614 0.3301 0.3055
580/1 0.2861
A8 0.2953 0.2221 0.2827 0.2667
A9 0.1823 0.1431 0.1895 0.1716
580/2 0.2040
A10 0.2645 0.1938 0.2510 0.2364
A1 0.2240 0.1575 0.2294 0.2036
580/4 0.1863
A2 0.1940 0.1318 0.1810 0.1689
A17 0.1401 0.0356 0.1568 0.1108
QT650 0.1362
A18 0.2362 0.1022 0.1464 0.1616
A15 0.2321 0.1234 0.2358 0.1971
QT750 0.2741
A16 0.3473 0.2889 0.4173 0.3512
46
Table 4.1b - Inner Distortion for 8822 Steel

Condition Sample #
B13 0.1927 0.1552 0.1971 0.1817
500/2 0.1970
B14 0.2213 0.1816 0.2339 0.2123
B3 0.1807 0.1508 0.1822 0.1712
500/4 0.1603
B4 0.1543 0.1303 0.1634 0.1493
B11 0.2405 0.1873 0.2289 0.2189
550/2 0.2087
B12 0.2148 0.1710 0.2094 0.1984
B5 0.1942 0.1587 0.2084 0.1871
550/4 0.1921
B6 0.1952 0.1719 0.2239 0.1970
B7 0.2317 0.1898 0.2294 0.2170
580/1 0.2570
B8 0.2062 0.1854 0.4992 0.2970
B9 0.2059 0.1659 0.1930 0.1883
580/2 0.1849
B10 0.1898 0.1587 0.1961 0.1816
B1 0.1376 0.1020 0.1615 0.1337
580/4 0.1350
B2 0.1558 0.1092 0.1438 0.1363
B17 0.4530 0.2464 0.3959 0.3651
QT650 0.3377
B18 0.3967 0.2371 0.2969 0.3102
B15 0.2889 0.1904 0.3169 0.2654
QT750 0.3078
B16 0.3772 0.2673 0.4058 0.3501
Table 4.1c - Inner Distortion for 4320 Steel

Condition Sample #
C13 0.0545 0.0009 0.0648 0.0401
500/2 0.0951
C14 0.1746 0.1145 0.1611 0.1501
C3 0.1130 0.0690 0.1153 0.0991
500/4 0.0867
C4 0.0986 0.0384 0.0860 0.0744
C11 0.0867 0.0589 0.0889 0.0782
550/2 0.0881
C12 0.1170 0.0668 0.1103 0.0980
C5 0.1100 0.0920 0.1333 0.1118
550/4 0.0916
C6 0.0929 0.0400 0.0813 0.0714
C7 0.1383 0.1012 0.1404 0.1266
580/1 0.0846
C8 0.0674 0.0148 0.0457 0.0426
C9 0.0812 0.0415 0.0941 0.0723
580/2 0.0727
C10 0.0949 0.0343 0.0902 0.0731
C1 0.0833 0.0139 0.0731 0.0568
580/4 0.0464
C2 0.0434 0.0003 0.0642 0.0360
C17 0.2835 0.1355 0.1900 0.2030
QT650 0.1589
C18 0.1587 0.0236 0.1622 0.1149
C15 0.2328 0.0762 0.2558 0.1882
QT750 0.2106
C16 0.3290 0.1267 0.2435 0.2330
47
0.45
0.4
0.35
ID % Change
0.3
0.25
0.2
0.15 8620
0.1 4320
0.05 8822
0
-0.05
Condition
Figure 4.5: Inner diameter distortion.
Outer Diameter:
Outer diameter distortion measurements for both 4320 and 8822 steel showed
austempering resulted in less distortion than quench-and-tempering. At 550 °F, 8620 steel
had austempered distortion values which were similar to those found in quench and
tempering. For other temperatures, the 8620 distortion values for austempering were
lower than those of the quench-and-tempering processes. For all temperatures, 8620 steel
exhibited less distortion with increased time. This trend was not seen with the other two
materials. As such, the argument for diffusional versus shear transformation would
appear incomplete. The change from a 650 °F to 750 °F temper decreased the outer
diameter distortion in the 4320 and 8822 steels; it increased the distortion in the 8620
steel.
48
Table 4.2a - Outer Distortion for 8620 Steel
Outer Diameter Distortion % Condition
Condition Sample #
A13 0.1813 0.1909 0.1835 0.1852
500/2 0.1909
A14 0.1991 0.1940 0.1963 0.1965
A3 0.1848 0.1747 0.1810 0.1802
500/4 0.1746
A4 0.1689 0.1676 0.1707 0.1690
A11 0.1890 0.1793 0.1743 0.1809
550/2 0.2420
A12 0.2051 0.3414 0.3631 0.3032
A5 0.2076 0.4133 0.2330 0.2847
550/4 0.2332
A6 0.1860 0.1794 0.1799 0.1818
A7 0.1927 0.1925 0.2030 0.1961
580/1 0.2048
A8 0.2092 0.2171 0.2143 0.2135
A9 0.1550 0.1556 0.1853 0.1653
580/2 0.1829
A10 0.1994 0.1925 0.2096 0.2005
A1 0.1531 0.1521 0.1670 0.1574
580/4 0.1551
A2 0.1524 0.1524 0.1536 0.1528
A17 0.1831 0.2031 0.2123 0.1995
QT650 0.2073
A18 0.2111 0.2126 0.2215 0.2150
A15 0.2341 0.2390 0.2394 0.2375
QT750 0.2563
A16 0.2516 0.2718 0.3018 0.2751
Table 4.2b - Outer Distortion for 8822 Steel

Condition Sample #
B13 0.1626 0.1669 0.1445 0.1580
500/2 0.1627
B14 0.1600 0.1781 0.1643 0.1675
B3 0.1401 0.1464 0.1425 0.1430
500/4 0.1435
B4 0.1340 0.1502 0.1476 0.1439
B11 0.1678 0.1755 0.1556 0.1663
550/2 0.1652
B12 0.1566 0.1767 0.1588 0.1641
B5 0.1474 0.1570 0.1476 0.1507
550/4 0.1658
B6 0.1810 0.1773 0.1846 0.1810
B7 0.2188 0.2161 0.2098 0.2149
580/1 0.1868
B8 0.1641 0.1650 0.1471 0.1587
B9 0.1423 0.1419 0.1287 0.1376
580/2 0.1343
B10 0.1289 0.1360 0.1279 0.1309
B1 0.1242 0.1442 0.1269 0.1318
580/4 0.1350
B2 0.1381 0.1511 0.1259 0.1383
B17 0.2471 0.2348 0.2682 0.2500
QT650 0.2301
B18 0.2104 0.2068 0.2135 0.2102
B15 0.1701 0.1797 0.1622 0.1707
QT750 0.2026
B16 0.2367 0.2194 0.2476 0.2346
49
Table 4.2c - Outer Distortion for 4320 Steel

Condition Sample #
C13 0.0811 0.0849 0.0880 0.0847
500/2 0.1085
C14 0.1303 0.1391 0.1277 0.1324
C3 0.1161 0.1151 0.1165 0.1159
500/4 0.1096
C4 0.0792 0.0882 0.1423 0.1032
C11 0.1257 0.1475 0.1275 0.1336
550/2 0.1267
C12 0.1156 0.1272 0.1168 0.1199
C5 0.1682 0.1736 0.1829 0.1749
550/4 0.1810
C6 0.1959 0.1787 0.1869 0.1872
C7 0.1365 0.1403 0.1311 0.1360
580/1 0.1093
C8 0.0865 0.0860 0.0754 0.0826
C9 0.1087 0.1138 0.1058 0.1094
580/2 0.1003
C10 0.0885 0.0945 0.0905 0.0912
C1 0.1097 0.1125 0.1133 0.1118
580/4 0.1078
C2 0.0949 0.1046 0.1121 0.1039
C17 0.1884 0.2005 0.1744 0.1878
QT650 0.2037
C18 0.2172 0.2366 0.2052 0.2197
C15 0.1574 0.1944 0.2154 0.1891
QT750 0.1921
C16 0.1975 0.2024 0.1853 0.1951
0.35
0.3
OD % Change
0.25
0.2
0.15 8620
0.1 4320
0.05 8822
Condition
Figure 4.6: Outer diameter distortion.
50
Thickness:
Thickness results for all three materials revealed that austempering produced less
distortion than quench-and-tempering. For all three steels the 500 °F and 550 °F
austempering temperatures exhibited decreased the distortion levels with increasing time.
The opposite was true with the 580 °F temperature, where distortion increased with
increased austempering time regardless of material.
Table 4.3a - Thickness Distortion for 8620 Steel
Distortion
Sample # Average σ/2
%
A13 0.0471
500/2 0.0721 0.0177
A14 0.097
A3 0.0203
500/4 0.0314 0.0078
A4 0.0425
A11 0.1295
550/2 0.1220 0.0053
A12 0.1144
A5 0.073
550/4 0.0560 0.0120
A6 0.039
A7 0.0712
580/1 0.0395 0.0224
A8 0.0079
A9 0.099
580/2 0.0732 0.0182
A10 0.0474
A1 0.1102
580/4 0.0803 0.0212
A2 0.0503
A17 0.1099
QT650 0.1557 0.0324
A18 0.2015
A15 0.094
QT750 0.1841 0.0637
A16 0.2742
51
Table 4.3b - Thickness Distortion for 8822 Steel
Distortion
%
B13 0.0499
500/2 0.0480 0.0171
B14 0.0462
B3 0.0098
500/4 0.0326 0.0160
B4 0.0554
B11 0.0663
550/2 0.0701 0.0272
B12 0.0740
B5 0.0461
550/4 0.0534 0.0188
B6 0.0607
B7 0.0331
580/1 0.0372 0.0431
B8 0.0413
B9 0.0504
580/2 0.0431 0.0233
B10 0.0358
B1 0.0882
580/4 0.0576 0.0089
B2 0.0271
B17 0.1267
QT650 0.1727 0.0246
B18 0.2188
B15 0.1195
QT750 0.1712 0.0261
B16 0.2229
Table 4.3c - Thickness Distortion for 4320 Steel

Distortion
%
A13 0.0964
500/2 0.0922 0.0030
A14 0.0879
A3 0.0793
500/4 0.0641 0.0107
A4 0.0490
A11 0.1015
550/2 0.1022 0.0005
A12 0.1028
A5 0.0685
550/4 0.0717 0.0023
A6 0.0750
A7 0.0078
580/1 0.0267 0.0133
A8 0.0455
A9 0.0693
580/2 0.0828 0.0096
A10 0.0963
A1 0.0863
580/4 0.0959 0.0068
A2 0.1056
A17 0.1952
QT650 0.1909 0.0031
A18 0.1865
A15 0.2443
QT750 0.2100 0.0243
A16 0.1757
52
0.3
Thickness % Change 0.25
0.2
0.15
8620
0.1 4320
0.05 8822
Condition
Figure 4.7: Thickness distortion.
Gap Width:
Gap width measurements for all three materials showed that the quench-and-
tempered samples produced larger distortion than the austempered conditions. The 8620
had the largest gap width distortion and 4320 the least for all austempering conditions
except for 500/4, where 8822 steel had the least distortion. The 500/4 condition for 8822,
however, exhibited the largest variation about the mean within the austempered
conditions. All materials exhibited lower distortion with increased austempering time at
500 °F and 580 °F. This trend was not noted at 500 °F.
53
Table 4.4a - Gap Width Distortion for 8620 Steel
Top Middle Bottom Specimen Parameter

Sample # σ/2
Distortion Distortion Distortion %Change %Change
A13 5.2783 5.4663 5.3809 5.3752
500/2 5.4663 0.0617
A14 5.5990 5.5909 5.4825 5.5575
A3 4.9797 4.9888 4.9868 4.9851
500/4 5.0160 0.0855
A4 5.3518 4.8713 4.9176 5.0469
A11 4.1155 4.0753 3.9605 4.0504
550/2 4.0731 0.0318
A12 4.1178 4.0427 4.1267 4.0957
A5 4.7492 4.7920 4.9908 4.8440
550/4 4.7122 0.0883
A6 4.5184 4.5316 4.6913 4.5804
A7 5.8248 5.7672 5.7362 5.7761
580/1 5.6184 0.0880
A8 5.4494 5.4376 5.4953 5.4608
A9 3.4316 3.4220 3.5096 3.4544
580/2 4.0194 0.3099
A10 4.5563 4.5915 4.6054 4.5844
A1 4.2255 4.1898 4.2416 4.2189
580/4 4.0238 0.1114
A2 3.8106 3.7431 3.9321 3.8286
A17 5.1698 5.3723 5.3305 5.2909
QT650 5.8267 0.3137
A18 6.6248 6.4756 5.9873 6.3626
A15 6.4891 6.6753 6.4525 6.5390
QT750 7.1803 0.3675
A16 7.4820 7.8652 8.1178 7.8217
54
Table 4.4b - Gap Width Distortion for 8822 Steel

Sample # σ/2
A13 3.3740 3.3824 3.2721 3.3428
500/2 3.4669 0.0815
A14 3.5088 3.7393 3.5245 3.5909
A3 3.4656 3.0444 3.4249 3.3117
500/4 2.8763 0.6616
A4 1.6535 2.9167 2.7525 2.4409
A11 4.0776 4.0211 3.7667 3.9551
550/2 3.7818 0.1110
A12 3.6207 3.6766 3.5280 3.6084
A5 3.5841 3.6916 3.5783 3.6180
550/4 3.8730 0.1423
A6 4.0896 4.1941 4.1001 4.1279
A7 3.9677 3.9850 3.9196 3.9574
580/1 3.8773 0.0610
A8 3.7772 3.9358 3.6785 3.7971
A9 3.4698 3.5327 3.4199 3.4741
580/2 3.3562 0.0700
A10 3.1895 3.3111 3.2143 3.2383
A1 3.1468 3.0865 2.9720 3.0684
580/4 3.1796 0.0694
A2 3.2485 3.3553 3.2685 3.2908
A17 7.3694 7.4491 7.2680 7.3622
QT650 6.9373 0.2583
A18 6.7679 6.6465 6.1227 6.5124
A15 6.6669 6.8680 6.7232 6.7527
QT750 6.9621 0.1259
A16 7.1716 7.2984 7.0448 7.1716
Table 4.4c - Gap Width Distortion for 4320 Steel

Sample # σ/2
A13 2.3964 2.5279 2.4518 2.4587
500/2 3.0825 0.3433
A14 3.7439 3.7635 3.6115 3.7063
A3 3.0861 3.2283 3.0530 3.1224
500/4 2.7320 0.2219
A4 2.1985 2.3071 2.5190 2.3415
A11 3.2849 3.5859 3.3159 3.3956
550/2 3.3185 0.0701
A12 3.2581 3.2940 3.1722 3.2415
A5 3.3717 3.5120 3.4298 3.4379
550/4 3.3149 0.0731
A6 3.1288 3.2318 3.2153 3.1920
A7 3.5101 3.6569 3.5254 3.5641
580/1 3.0353 0.2930
A8 2.4928 2.6270 2.3997 2.5065
A9 2.5551 2.7220 2.5890 2.6220
580/2 2.7245 0.0666
A10 2.8371 2.8926 2.7509 2.8269
A1 2.5976 2.6812 2.5304 2.6031
580/4 2.4926 0.0699
A2 2.2922 2.4491 2.4051 2.3821
A17 7.4044 7.2238 6.6471 7.0918
QT650 7.0565 0.1292
A18 6.9716 7.1166 6.9758 7.0213
A15 7.4744 7.4564 7.5246 7.4851
QT750 7.5319 0.0974
A16 7.8119 7.6772 7.2467 7.5786
55
9.00
8.00
7.00
Gap Width % Change
6.00
5.00
4.00 8620
3.00 4320
2.00 8822
1.00
0.00
Condition
Figure 4.8: Gap width distortion.
Summary of Size Distortion:
Size distortions results can be summarized as follows. 1) For ID, OD, and gap
width size distortions, 4320 steel had the least distortion for nearly all austempering
conditions. For thickness distortion, it was often associated with the highest distortion
among the three steels. 8822 consistently had less distortion that 8620 when austempered.
2) The austempering heat treatments consistently produced less distortion than the
quench-and-tempering heat treatments. This decreased distortion was attributed to the
phase change to bainite being a smaller volume change than the transformation to
martensite.
56
4.2.2 Shape Distortion
Shape distortion results, flatness, cylindricity, and roundness, are given as a
percentage in Tables 4.5 - 4.7a and Figures 4.9 -4.11, while measurement values are
given in Tables 4.5 – 4.7b. Shape distortions showed a larger percentage change
compared to the size distortion measurements. This increased change occurs because the
initial measurement for shape distortions are small compared to the initial measurements
found in the size distortions. Therefore, any change in shape will be reflected as a larger
percentage change.
Flatness:
The 8822 steel results had flatness values several times lower than the values for
8620 and 4320 steel. These results were consistent for both austemper and quench-and-
temper processes. 8822 steel also had little error distribution relative to the other two
materials. The 8620 and 4320 yielded similar flatness results. The quench-and-tempering
processes produced considerably more than two times the flatness distortion of the
austempering conditions. There was no pattern observed in terms of the effects of time
and temperature on distortion.
57
Table 4.5a - Flatness Distortion
Flatness % Change
Condition 8620 4320 8822
500/2 75.6 ± 15 101.7 ± 45 28.2 ± 2
500/4 52.4 ± 25 83.7 ± 28 26.2 ± 3
550/2 87.2 ± 33 88.0 ± 2 24.7 ± 7
550/4 34.6 ± 18 84.3 ± 8 25.3 ± 8
580/1 82.9 ± 1 72.9 ± 6 13.0 ± 0
580/2 77.6 ± 17 32.0 ± 0 14.7 ± 2
580/4 59.1 ± 4 57.3 ± 7 4.3 ± 1
QT650/1 223.2 ± 69 233.6 ± 12 84.7 ± 12
QT750/1 184.7 ± 17 251.8 ± 46 81.9 ± 34
Table 4.5b - Flatness Distortion
Flatness Change (mm)

Condition 8620 4320 8822
500/2 0.0076 0.0065 0.0064
500/4 0.0053 0.0088 0.0069
550/2 0.0069 0.0061 0.0051
550/4 0.0034 0.0060 0.0047
580/1 0.0080 0.0062 0.0042
580/2 0.0063 0.0040 0.0037
580/4 0.0057 0.0041 0.0015
QT650 0.0207 0.0196 0.0148
QT750 0.0189 0.0214 0.0147
58
350
300
Flatness % Change
250
200
150 8620
4320
100
8822
50
Condition
Figure 4.9: Flatness distortion.
Roundness:
Roundness data exhibits the opposite behaviour of flatness. For each heat
treatment parameter, the 8822 steel had the largest distortion. The quenched and
tempered samples had, interestingly, distortions that fell in the middle range when
compared to the austempered conditions. The 8822 results showed that with increased
time at a given temperature the roundness distortion is reduced. The reported error for
roundness was relatively small compared to the magnitudes of change found.
59
Table 4.6a - Roundness Distortion
Roundness % Change
8620 4320 8822
500/2 4.74 ± 2.0 11.64 ± 1.7 229.89 ± 9.7
500/4 11.44 ± 7.2 26.38 ± 3.0 128.65 ± 13.7
550/2 7.76 ± 0.9 19.74 ± 2.2 212.62 ± 8.5
550/4 1.43 ± 1.8 20.86 ± 5.2 63.53 ± 6.2
580/1 7.86 ± 3.7 7.70 ± 3.0 248.56 ± 7.2
580/2 1.59 ± 2.7 1.72 ± 4.2 144.75 ± 13.1
580/4 12.21 ± 3.0 5.46 ± 7.1 81.58 ± 5.0
QT650 13.59 ± 2.8 19.81 ± 3.6 150.77 ± 10.4
QT750 10.57 ± 3.0 3.35 ± 4.7 200.25 ± 6.8
Table 4.6b - Roundness Distortion
Roundness Change (mm)

Condition 8620 4320 8822
500/2 0.0050 0.0098 0.0284
500/4 0.0113 0.0166 0.0188
550/2 0.0075 0.0105 0.0267
550/4 0.0015 0.0115 0.0070
580/1 0.0081 0.0060 0.0327
580/2 0.0014 0.0014 0.0143
580/4 0.0084 0.0043 0.0132
QT650 0.0125 0.0146 0.0225
QT750 0.0097 0.0028 0.0210
60
300
250
Roundness % Cahnge 200
150 8620
100 4320
8822
50
0
500/2 500/4 550/2 550/4 580/1 580/2 580/4 QT650 QT750
-50
Condition
Figure 4.10: Roundness distortion.
Cylindricity:
As with roundness, the cylindricity results showed that the 8822 again had the
largest distortions. However, the 580/2 and 580/4 conditions for the 8822 steel are in line
with the other two steels. Unlike the roundness measurements, the error associated with
cylindricity could be relatively large. With the exception of the 550/4 condition for 4320
and the 580/2 condition for 8620, the distortion values for austempered conditions were
lower than those found for the quenched-and-tempered conditions.
Table 4.7a - Cylindricity Distortion

Cylindricity % Change
8620 4320 8822
500/2 4.85 8.66 181.66
500/4 0.82 6.96 123.94
550/2 6.09 7.22 145.45
550/4 10.59 48.90 77.45
580/1 5.62 13.98 229.52
580/2 38.67 2.39 17.51
580/4 6.75 3.82 27.23
QT650 27.98 22.05 216.88
QT750 44.42 11.85 141.01
61
Table 4.7b - Cylindricity Distortion
Cylindricity Change (mm)
Condition 8620 4320 8822
500/2 0.0019 0.0005 0.0277

500/4 0.0014 0.0049 0.0201
550/2 0.0014 0.0019 0.0254
550/4 0.0017 0.0057 0.0123
580/1 0.0015 0.0060 0.0341
580/2 0.0086 0.0036 0.0115
580/4 0.0041 0.0024 0.0131
QT650 0.0148 0.0117 0.0281
QT750 0.0171 0.0067 0.0277
350
300
Cylindricity % Change
250
200
8620
150
4320
100 8822
50
0
500/2 500/4 550/2 550/4 580/1 580/2 580/4 QT650 QT750
Condition
Figure 4.11: Cylindricity distortion.
Summary of Shape Distortion:
The primary shape distortion results can be summarized as follows. 1) For both
the cylindricity and roundness measurements, the 8822 steel distortion values were
considerably larger than the other materials for nearly all parameters. 8822 steel exhibits
comparable or superior distortion characteristics for the flatness. Of note is that both
cylindricity and roundness pertain to deviations on the inner and outer curved surfaces, in
62
essence a hoop direction. The flatness measurements, where 8822 steel produced much
less distortion, relates to an axial direction. 2) The quench-and-temper results exhibited
significantly higher distortion values for flatness. The cylindricity and roundness values
were in the mid-range of the austempered conditions.
4.3 Residual Stress
Surface residual stress measurement values are provided in Table 4.8; those
values are presented in graphical form in Figure 4.12. Calibration results, as detailed in
chapter 3, were in accordance with the acceptable limits put forth by ASTM Standard
E915.
The direction of measurement (hoop or axial) did not significantly affect the
magnitude of exhibited surface residual stress. The greatest magnitude change was 4 MPa
for the 580/1 austempered candidate. The austempered samples exhibited higher
compressive residual stress than the quench and tempered samples. This is the expected
result because the tempering process is designed to relax the residual stresses, which
forum due to the transformation associated with the quenching process.
63
Table 4.8 - Surface Residual Stress Results for 4320 Steel
Surface Residual Stress (ksi)

Hoop Axial
500/2 -73 ± 0.9 -75 ± 0.4
500/4 -86 ± 0.9 -86 ± 0.4
550/2 -64 ± 0.7 -67 ± 0.6
550/4 -81 ± 0.6 -83 ± 0.5
580/1 -84 ± 0.9 -80 ± 0.3
580/2 -85 ± 0.9 -85 ± 0.5
580/4 -72 ± 0.9 -71 ± 0.3
QT650/1 -44 ± 0.9 -47 ± 0.5
QT750/1 -59 ± 0.6 -60 ± 0.4
-100
Compressive Hoop Stress (ksi)
-90
-80
-70
-60
-50 Hoop
-40
Axial
-30
-20
-10
0
Condition
Figure 4.12: Surface residual stress for 4320 steel.
Figure 4.13 plots residual stress versus the outer diameter change. No significant
trend is noted for the austempered samples, only that a higher surface compressive
residual stress appears to be associated with smaller OD changes. This association would
hold for most of the size distortion, and with flatness, where the austempered samples
exhibited lower distortion.
64
Austempered Quench and Temper
Residual Compressive Hoop Stress

-100
-90
-80
-70
-60
(ksi)
-50
-40
-30
-20
-10
0
0.05 0.1 0.15 0.2
Outer Diameter % Change
Figure 4.13: Surface residual stress and outer diameter distortion.
The data from residual stress depth profile measurements in one 4320
austempered sample, 500/4 and one QT750 4320 sample are provided in Table 4.9; and
plots of residual stress vs depth can be seen in Figure 4.14a. The initial region is provided
in greater detail in Figure 4.15b.
Table 4.9 - Residual Stress Depth Measurements for 4320 Steel
500/4 QT750
Depth Depth
Axial Hoop Axial Hoop
(inches) (inches)
0.0000 -74 ± 0.7 -76 ± 0.8 0.0000 -59 ± 0.8 -58 ± 0.9
0.0006 -102 ± 1.3 -100 ± 1.0 0.0006 -53 ± 1.2 -52 ± 0.9
0.0011 -116 ± 1.4 -118 ± 1.3 0.0012 -53 ± 1.0 -54 ± 1.0
0.0030 -71 ± 2.0 -69 ± 1.8 0.0028 -30 ± 1.0 -29 ± 0.8
0.0052 -50 ± 1.9 -49 ± 1.5 0.0048 -9 ± 1.0 -5 ± 1.3
0.0103 -59 ± 1.5 -55 ± 2.2 0.0107 -2 ± 1.2 -3 ± 0.9
0.0200 -52 ± 1.4 -51 ± 2.2 0.0235 -8 ± 0.6 -6 ± 0.8
0.0454 +4 ± 1.2 -1 ± 1.0 0.0403 +4 ± 1.2 +3 ± 1.0
0.0666 +22 ± 1.1 +24 ± 0.8 0.0686 +13 ± 0.8 +12 ± 1.1
0.1016 +12 ± 1.1 +12 ± 1.4 0.0997 +6 ± 0.9 +4 ± 1.5
65
40
20
Residual Axial 0
-20
Stress (ksi)
-40
-60
-80
-100
-120 500/4 QT750
-140
0.0000 0.0200 0.0400 0.0600 0.0800 0.1000 0.1200
Depth (Inches)
Figure 4.14a: Residual stress and depth for austempered and quench-and-tempered 4320
steel.
-20
-40
Residual Axial
-60
Stress (ksi)
-80
-100
-120 500/4 QT750

-140
0.0000 0.0020 0.0040 0.0060 0.0080 0.0100 0.0120
Depth (Inches)
Figure 4.14b: Residual stress and depth for austempered and quench-and-tempered 4320
steel.
66
Consistent with the work of other investigators [29, 39, 42], it was noted that both
the austempered and quench-and-tempered conditions had compressive residual stress at
the surface. However, differences were noted in the subsurface residual stress pattern. As
noted in Figure 4.14a, quench-and-tempered sample decreased in its magnitude of
compressive stress to nearly a non-stressed state at 0.012 inches. A small compressive
region continued to about 0.3 inches, after which, it maintained a low tensile residual
stress deeper into the material. The austempered sample increased its magnitude of
compressive stress below the surface, reaching a peak of 116 ksi of compressive stress,
before stress levels dropped off to a magnitude of 50 ksi. Compressive residual stresses
extended to 0.045 inches into the material, before becoming tensile.
Residual stress depth testing showed that the compressive region for the
austempered condition extended deeper into the material than that of the quenched-and-
tempered condition. As well, the largest magnitude of compressive residual stress was
larger for the austempered condition. Although the benefits of these characteristics have
been laid out in the literature [29, 39, 42], wear and fatigue testing should be carried out
to correlate higher levels and deeper penetration of compressive residual stress to wear
and fatigue life.
Retained austenite values are shown in Table 4.10, and are plotted on Figure 4.15.
Retained austenite measurements showed that, for the quench and tempered sample, the
retained austenite amounts were all below 1.5%; the austempered sample peaked at 8%.
The retained austenite levels found in the quench-and-tempered condition can be
67
explained by the high temperature tempering process, which allows for the
decomposition of the retained austenite into bainite. The retained austenite peak found in
the austempered sample is worth noting, primarily, any further decomposition of the
retained austenite during service, could result in additional distortion. While additional
testing would provide a more complete retained austenite profile, the general shape is
identified. There was no correlation between the level of retained austenite found at a
certain depth and the associated residual stress at that depth.
Table 4.10 - Retained Austenite Measurement
500/4 QT750
% %
Depth Depth
Retained Retained
(inches) (inches)
Austenite Austenite
0.0000 1.6 ± 0.9 0.0000 1.1 ± 0.3
0.0011 1.9 ± 0.3 0.0012 1.4 ± 0.9
0.0052 8.0 ± 1.4 0.0048 1.2 ± 0.6
0.0103 2.0 ± 0.4 0.0107 0.8 ± 0.5
0.0200 0.9 ± 0.5 0.0235 1.0 ± 0.4
0.0666 1.0 ± 0.2 0.0686 1.0 ± 0.4
68
10.0
9.0
500/4
Retained Austenite (%)
8.0
7.0
QT750
6.0
5.0
4.0
3.0
2.0
1.0
0.0
0.0000 0.0100 0.0200 0.0300 0.0400 0.0500 0.0600 0.0700 0.0800
Depth (inches)
Figure 4.15: Retained austenite with depth.
4.5.1 Hardness
Hardness measurements for core and case are shown in Tables 4.11 and 4.12, and
Figures 4.16 and 4.17. Core hardness values were found to be fairly similar between the
austempering process and the quench-and-tempering process. Across materials, the 8822
samples had higher hardness compared to the other materials. This is attributed to the
higher carbon content of 8822 steel. 4320 steel tended to be harder than 8620 for the
same condition. It was noted that with additional time and temperature there was a trend
towards softer cores. This is explained in the austempered samples by the fact that with
higher temperatures there is more energy available for diffusion and softening, while
higher temperature allows for tempering of the martensite formed in the core crossing the
line while quenching down to the austempering temperature. With the quench-and-
69
tempered samples and the higher core hardness it is assumed that the elevated
temperature is offset by the short amount of tempering time.
Table 4.11 - Core Hardness
Material
Parameter 8620 4320 8822
500/2 31.2 ± 0.4 38.2 ± 0.4 43.2 ± 0.6
500/4 31.5 ± 0.3 38.4 ± 0.1 42.1 ± 0.2
550/2 33.9 ± 0.5 34.5 ± 0.2 42.9 ± 0.3
550/4 26.3 ± 1.7 34.8 ± 0.2 41.6 ± 0.1
580/1 29.7 ± 0.3 34.8 ± 0.2 36.6 ± 0.2
580/2 26.9 ± 0.1 33.4 ± 0.6 35.4 ± 0.4
580/4 27.3 ± 0.2 32.9 ± 0.2 36.6 ± 0.2
QT650/1 36.3 ± 2.0 41.6 ± 0.9 44.0 ± 0.0
QT750/1 39.8 ± 0.2 39.2 ± 0.1 42.4 ± 0.2
50.0
45.0
40.0
35.0
30.0
HRC
25.0
20.0 8620
15.0 4320
10.0
5.0 8822
0.0
Condition
Figure 4.16: Core hardness.
70
Table 4.12 - Case Hardness
Material
Parameter 8620 4320 8822
500/2 58.1 ± 0.2 59.1 ± 0.6 59.1 ± 0.1
500/4 58.6 ± 0.4 59.0 ± 0.9 58.5 ± 0.2
550/2 58.2 ± 0.4 57.2 ± 0.7 57.7 ± 0.2
550/4 54.5 ± 0.2 55.0 ± 0.1 57.8 ± 0.3
580/1 54.2 ± 0.1 53.5 ± 0.1 54.8 ± 0.6
580/2 52.9 ± 0.4 52.6 ± 0.3 54.7 ± 0.2
580/4 54.4 ± 0.1 53.1 ± 0.2 54.2 ± 0.3
QT650 55.7 ± 0.1 55.2 ± 0.4 56.0 ± 0.1
QT750 52.4 ± 0.1 51.0 ± 0.1 53.1 ± 0.1
64.0
60.0
56.0
HRC
52.0 8620
4320
48.0
8822
44.0
Condition
Figure 4.17: Case hardness.
When comparing the case hardness values, it is first most evident that the case
hardnesses are much higher than those associated with the core. This is expected with the
case hardening process because of the additional carbon content at the surface. Similar to
the core hardness measurements, it was noted that with increased time and temperature
combinations the hardness values decreased. There was not a large difference between
materials in terms of case hardness. All of the samples were carburized at the same
71
carbon potential, and thusly have the same carbon content. The differences in hardness in
the austempered conditions can be associated in part to the type of bainite formed during
austempering. As mentioned, upper bainite is a softer microstructure than that of lower
bainite. Since the bainite formation at 500 °F will be closer to lower bainite, and thusly,
martensite, it will have a higher hardness. The quench-and-temper samples exhibit the
same pattern, where higher temperature results in a lower hardness. This is due to the
higher temperature allowing for more softening of the material.
4.5.2 Toughness
Charpy impact toughness values are shown in Table 4.13. Averages of three
samples are presented for all conditions, along with standard deviation. Full data sets
were not available for all samples. The samples which were quench-and-tempered at 350
°F only had 2 samples prepared for each material. Impact values are compared in Figure
4.18.
Table 4.13a - Charpy Impact Toughness for 8620 Steel
Material Parameter 1 2 3 Average (ft-lbs)

500/2 4.07 4.59 4.44 4.37 ± 0.27
500/4 4.18 3.90 4.23 4.10 ± 0.18
550/2 2.57 2.53 3.21 2.77 ± 0.38
550/4 5.68 6.19 6.70 6.19 ± 0.51
580/1 8.69 7.33 7.50 7.84 ± 0.74
8620
580/2 8.33 7.00 6.97 7.43 ± 0.78
580/4 - - - 9.30
QT350 5.92 5.81 - 5.87 ± 0.04
QT650 2.93 2.81 2.49 2.74 ± 0.23
QT750 7.06 8.55 7.25 7.62 ± 0.81
72
Table 4.13b - Charpy Impact Toughness for 8822 Steel

500/2 2.33 2.52 2.37 2.41 ± 0.10
500/4 2.41 2.21 1.95 2.19 ± 0.23
550/2 2.42 2.27 2.25 2.31 ± 0.09
550/4 2.22 2.24 2.20 2.22 ± 0.02
580/1 3.58 3.06 2.90 3.18 ± 0.36
8822
580/2 3.56 3.43 2.96 3.32 ± 0.32
580/4 - - - 2.68
QT350 1.96 2.27 - 2.12 ± 0.11
QT650 1.79 1.82 1.79 1.80 ± 0.02
QT750 2.35 2.43 2.34 2.37 ± 0.05
Table 4.13c - Charpy Impact Toughness for 4320 Steel

500/2 11.43 12.52 12.12 12.02 ± 0.55
500/4 11.98 11.46 11.81 11.75 ± 0.27
550/2 7.94 9.93 8.54 8.80 ± 1.02
550/4 10.37 11.65 12.91 11.64 ± 1.27
580/1 7.56 5.94 8.31 7.27 ± 1.21
4320
580/2 11.60 10.65 11.18 11.14 ± 0.48
580/4 - - - 11.82
QT350 11.48 10.45 - 10.97 ± 0.36
QT650 11.62 10.87 12.80 11.76 ± 0.97
QT750 19.95 21.58 20.75 20.76 ± 0.82
73
25.00
Energy Absorbed (ft/lbf)

20.00
15.00
10.00 8620
4320
5.00
8822
0.00
Condition
Figure 4.18: Charpy toughness results.
Austempering time and temperature did not seem to have an effect on the impact
toughness of the three steels. As previously noted, toughness is primarily indicative of the
low-carbon content core microstructure, the temperature of which is well above the
austempering temperature, and therefore not as affected by time as the case
microstructures. The 4320 samples which were quench-and-tempered at 750 °F showed
considerably more Charpy toughness than the rest of the samples which were tested. This
highest tempering temperature allows for a softening of the core microstructure.
When the austempered results were compared to those of the 350 °F temper it was
found that austempering was generally able to provide toughness similar to the currently
used process. A comparison of the materials showed that the 4320 steels gave the highest
results for impact toughness overall across all of its austempering and quench-and-
tempering temperatures, which is expected due to its increased nickel content. 8822 steel
74
was not able to achieve any toughness value greater than 3.3 ft/lbs. 8620 had the widest
spread of values within the austempered condition, 2.7 to 9.3 ft-lbs.
Fracture surfaces were examined for selected samples to examine the fracture
mode. Samples were selected based on having high and low impact toughness, as well as
to represent the three materials. Images are shown in Figure 4.19.
Figure 4.19: SEM fracture surface images.
Fracture surfaces for 4320 steel showed a majority of ductile fracture, while 8620
and 8822 steel had mostly brittle and mixed modes of fracture. This correlates with the
general expectation that lower impact toughness samples trend towards brittle rather than
ductile.
75
4.6 Economic Discussion
Comparable or improved mechanical and distortion properties are not by
themselves sufficient to proceed with the substitution of the austempering process for the
traditional quench-and-temper process. As has been noted, assessment of the tribological
behaviour of the austempered carburized case is a vital factor for components subject to
friction and wear. Additionally, the economics of the substitution must be considered. A
primary benefit of the austempering process is its low distortion, which may remove the
manufacturing step associated with distortion correction for the quench-and-temper
process. The time, machining, and materials handling costs associated with distortion
correction need to be compared to the increased time and energy costs associated with the
austempering thermal process.
The microstructure, mechanical properties, and distortion results of this study
indicate that the austempering process shows improvement over the traditional quench-
and-temper process for carburized steel components. A particularly strong argument can
be made for the 4320 steel to continue on to the next logical step of wear testing. The
8620 steel shows improvement in many areas for austempering compared to the quench-
and-temper process. The argument is perhaps not as strong for the 8822 material.
However, each material has a specific purpose with respect to carburizing: 4320 for
toughness with its nickel content; 8822 for higher hardness with its increased carbon
content; 8620 for general purpose carburizing. Should wear testing reveal comparable or
improved behaviour, the following and final step would be a full economic analysis.
76
V. CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK
The effects of austempering were compared to those of quench-and-tempering
when applied to a carburized low alloy steel. Three materials were tested, SAE 8620,
8822, and 4320 steel. Samples were carburized, then either austempered or quench-and-
tempered. Microstructures were observed for all conditions. Distortion due to the
carburizing and heat treatment process was measured. Residual stress and retained
austenite levels were measured for 4320 steel samples. Core and case hardness, as well as
toughness were determined.
5.1 Conclusions
The following conclusions were drawn based on the results of the experiment.
1. Austempering produced bainite in the case of carburized samples. With increased
time it was possible to form a greater percentage of bainite. Higher temperatures
resulted in the formation of more feathery upper bainite, while lower temperatures
resulted in more acicular, needle-like bainite. For the same temperature, 8822
tended to form lower bainite, while 4320 and 8620 would form upper bainite.
Incomplete transformations led to retained austenite and martensite being found in
the case. The case microstructure for all quench-and-tempered samples was
tempered martensite.
2. Core microstructures revealed martensite in the core of all samples. For the
austempered samples the low carbon content of the core, relative to the case,
resulted in the temperature being high enough that martensite transformation
77
occurred during the quench to the austempering temperature. For the quench-and-
tempered samples, the martensite core was subsequently tempered.
3. Size distortion was generally lower for austempered samples than quench-and-
tempered samples. 4320 showed less distortion across most of the size distortions
(ID, OD, gap width) than 8620 and 8822. Austempering produced similar or
better distortion characteristics than quench-and-tempering.
4. Austempering produced less shape distortion than quench and tempering,
although the differences were not significant. Roundness distortion was reduced
with increasing austempering time. 8822 steel exhibited large shape distortions,
and would require more correction than the other two materials.
5. Compressive surface residual stress was found in all tested Navy C-rings.
Austempered samples produced higher compressive residual stress than those
which had been quench-and-tempered. The tempering process allowed for the
relaxation of the residual stresses formed during the initial quenching process.
Residual stresses in the axial and longitudinal were with 4 ksi of each other.
6. Residual stress depth profiles revealed that the compressive layer in the
austempered material extended to a greater depth than that of the quench-and-
tempered material. Austempering peaked at a compressive residual stress of -118
ksi in the hoop direction at a depth of 0.0011 inches, while the quench-and-
tempered sample had its highest compressive stress at the surface. The quench-
and-tempered sample had near-neutral stress found around 0.01 inches
subsurface, before a slight increase in compressive stress. Tensile stresses were
found deeper 0.04 inches and deeper. The austempered sample maintained
78
complete compressive stress until 0.04 inches into the material before tensile
stress became present.
7. Retained austenite was present in greater amounts in the austempered sample;
however, at depths greater than 0.02 inches the retained austenite levels were
similar. Decomposition of retained austenite during tempering led to the lower
retained austenite levels in the quench-and-tempered sample.
8. Case hardness decreased with increased times and temperatures. The highest case
hardness was noted for the lowest austempering temperatures. Core hardness was
higher for quench-and-tempered samples for both 8620 and 4320 steel, and
similar for 8822.
9. 4320 steel was considerably tougher than the other materials, and is designed to
be such. For all steels, austempering produced a sample with toughness similar to
the 350 °F tempered samples. Although the 750 °F temper for 4320 steel
produced a high toughness relative to all the other conditions for all materials, it is
neither commonly used in production, nor is it under consideration for use. For
this study, the 750 °F temper was used as an upper limit of tempering.
5.2 Summary of Conclusions
The conclusions drawn from this study support the potential for austempering to
replace quench-and-tempering as a method of heat treating carburized low-alloy steels.
Austempering was able to reduce shape distortion, and limit size distortion for three low-
alloy steels, while maintaining mechanical properties similar to the currently used
process. Of some concern is the retained austenite level in the case of the austempered
conditions, as it can decompose and result in further distortion during service. Residual
79
stress values indicate that wear and fatigue properties could be improved with
austempering; actual wear and fatigue testing test data is currently lacking.
5.3 Recommendations for Future Work
Recommendations are provided below for further testing to determine the potential
for austempering to replace the quench-and-tempering process for carburized steels.
Suggestions for improving the accuracy of test results are also provided.
1. Larger samples populations are recommended for distortion testing. In this study
only two Navy C-rings were produced for each condition, led to a large statistical
variation in certain cases. Additional samples for each condition would help to
give a more definite value for expected distortion from heat treatment.
2. X-Ray diffraction should be performed on additional C-ring samples. The
increased sample size would limit the effects of localized peaks in stress and
retained austenite in individual specimens on the results. It is also recommended
that additional measurements be taken during depth testing to develop a more
complete profile of residual stress and retained austenite levels.
3. Carbon content depth profiles and case depth measurements would be useful to
relate to residual stress and retained austenite depth profiles, as well as
microstructural observations.
4. Production of quench-and-tempered C-rings tempered at 350 °F for one hour
would allow a full comparison between the current process and the austempering
heat treatment for distortion, residual stress, and retained austenite.
80
5. A more quantitative comparison of the economics of producing austempered parts
and quench-and-tempered parts should be performed. The differences in
production time, material selection, and equipment running costs should be
included. It will provide a more complete picture than just a physical properties
comparison
6. Wear and fatigue testing should be performed to further compare the
austempering and quench-and-tempering processes. Wear and fatigue testing
would allow for prediction of the in-service life of austempered parts compared to
quench-and-tempered ones.
7. A comparison of austempering to austempering-and-tempering would allow the
measurement of the effect of tempering the untempered martensite formed in the
core resulting from austempering alone. In particular, toughness would be
expected to increase.
5.4 Unique Contribution of the Work
The unique contribution of this study to the field was a quantitative examination
of the effects of austempering and quench-and-tempering on carburized low alloy steels.
Previous literature, while extolling the benefits of austempering as a heat treatment for
carburized parts [5, 9, 35, 56], had not made quantitative comparisons of distortion and
residual stress characteristics of the heat treatment processes. This study generated
quantitative measurements of distortion and residual stress for both the austempering and
quench-and-tempering processes.
81
REFERENCES
1. Engineering Properties of Steel. Edited by Philip D. Harvey. Metals Park, Ohio:

American Society For Metals, 1982
2. Smith, William F. Structure and Properties of Engineering Alloys. McGraw-Hill

Inc., 1993
3. “Fe-C Phase Diagram - The Basis of the Heat Treatment of Steel” Retrieved 10
January, 2013, from
http://www.mrl.ucsb.edu/~edkramer/LectureVGsMat100B/99Lecture14VGs/FeC
PhaseDiagramVG.html.
4. Callister Jr., William D. Fundamentals of Materials Science and Engineering / An

Interactive e-Text. New York: John Wiley & Sons, Inc, 2001.
5. Krauss, G. Steels: Heat Treatment and Processing Principles. Metals Park, Ohio;
ASM International, 1990.
6. Nishiyama, Zenji. Martensitic Transformation. New York: Academic Press, Inc,

1978.
7. Capdevila, C., F. G. Caballero, and Garcia De Andres. "Analysis of Effect of

Alloying Elements on Martensite Start Temperature of Steels." Materials Science
and Technology 19, no. 5 (2003): 581-586.
8. Liu, C., Z. Zhao, D. O. Northwood, and Y. Liu. "A New Empirical Formula for
the Calculation of Ms Temperatures in Pure Iron and Super-Low Carbon Alloy
Steels." Elsevier Science BV, P.O. Box 211, Amsterdam, 1000 AE, Netherlands,
2001.
9. Andrews, K.W., “Empirical Formulae for the calculation of some transformation

temperatures” Journal of the Iron and Steel Institute, Vol. 203, 1965, p 721-727.
10. Krauss, G. Principles of Heat Treatment of Steel. Metals Park, Ohio; American
Society for Metals, 1980.
82
11. Herring, Daniel H. "Understanding the Benefits of Austempering." Industrial
Heating 72, no. 12 (2005): 10,12.
12. Zhao, Z., X. Guan, C. Wan, C. Liu, and D. O. Northwood. "A Re-Examination of
the B Sub 0 and B Sub s Temperatures of Steel." Materials and Design (UK) 21,
no. 3 (2000): 207-209.
13. Zhao, Z., C. Liu, Y. Liu, and D. O. Northwood. "A New Empirical Formula for
the Bainite Upper Temperature Limit of Steel." Journal of Materials Science
(USA) 36, no. 20 (2001): 5045-5056.
14. Caballero, Francisca G., Maria Jesus Santofimia, Carlos Garcia-Mateo, and
Carlos Garcia De Andres. "Time-Temperature-Transformation Diagram within
the Bainitic Temperature Range in a Medium Carbon Steel." Materials
Transactions 45, no. 12 (2004): 3272-3281.
15. Lee, J. S., B. H. Song, H. G. Sung, S. Y. Kim, and B. Y. Hur. "The Effect of
Isothermal Heat Treatment on the Rolling Contact Fatigue of Carburized Low
Carbon Microalloyed Steel." Materials Science Forum 544-545, (2007): 151-154.
16. Clark, Donald S., and Varney, Wilbur R., Metallurgy for Engineers. New York,
NY: D. Van Nostrand Company, 1962.
17. Otto, F. J. and D. H. Herring. 2002. "Gear Heat Treatment Part I" Heat Treating
Progress (USA) 2 (5): 27-31.
18. J.R. Davis, Surface Engineering of Carbon and Alloy Steels, Surface Engineering,
Vol. 5, ASM Handbook, ASM International, 1994, p 701–740
19. Guide to practical boost-diffuse carburizing. Industrial Heating 60, no. 1 (1993):
36-37.
20. Réti, T., “Residual Stresses in Carburized, Carbonitrided, and Case-Hardened

Components,” Handbook of Residual Stress and Deformation of Steel. G. Totten,
M. Howes, T. Inoue (Editors), ASM International, Materials Park, OH., 2002.
189-208.
83
21. Child, H. C.; Still, Dr. F. “Boost/diffuse carburizing,” Heat Treatment of Metals
3, no. 2 (1976): 39
22. Pye, David. "Low-Pressure Carburizing." Industrial Heating 76, no. 4 (2009): 82-
82.
23. Larsson, M., B. Jansson, R. Blom, and A. Melander. "Influence of Austenitizing

and Quenching Temperature on Jominy Hardenability of a Case Hardening and a
Boron Steel." Scand.J.Metall. 19, no. 2 (1990): 51-63.
24. Davis, J.R., editor. Surface Hardening of Steels Understanding the Basics.
Materials Park, Ohio: ASM International, 2002
25. Tartaglia, John M.; Hayrynen, Kathy L. “A Comparison of Fatigue Properties of

Austempered Versus Quenched and Tempered 4340 Steel,” Journal of Materials
Engineering and Performance 21 (2012): 1008-1024.
26. Leslie, William C. The Physical Metallurgy of Steels. Marietta, Ohio: Hemisphere
Publishing Corporation, 1981.
27. Keough, John R. Austempered Materials and their Applications to Drive Line and
Suspension Components, 2000.
28. Keough, W. R. Equipment, Process and Properties of Modern Austempering,

edited by Keough, W. R. 1994.
29. Keough, W.R. “Carbo-Austempering.” Carburizing and Nitriding With

Atmospheres; Cleveland, Ohio; USA; 6-8 Dec. 1995(1995) 135-142.
30. Hart, B. J. "Austempering Offers Precision, Consistency, in Steel Quenching."

Heat Treat. 15, no. 10 (1983): 36-38.
31. Boyle, Erin, Randy Bowers, and Derek O. Northwood. "The use of Navy C-ring
Specimens to Investigate the Effects of Initial Microstructure and Heat Treatment
on the Residual Stress, Retained Austenite, and Distortion of Carburized
Automotive Steels." SAE Transactions: Journal of Materials & Manufacturing
116, (2008): 253-261.
84
32. Otto, F. J. and D. H. Herring. "Gear Heat Treatment Part II." Heat Treating
Progress (USA) 2, no. 5 (2002): 27-31.
33. Suliteanu, M. “Minimizing Gear Distortion During Heat Treatment,” Gear

Technology 13, no. 2 (1996): 15-19
34. Demerest, W. L. "Hardening of Industrial Saw Blades without Distortion."

Industrial Heating 48, no. 7 (1981): 14-15.
35. Northwood, Derek O., Lily He, Erin Boyle, and Randy Bowers. 2007. "Retained
Austenite - Residual Stress - Distortion Relationships in Carburized SAE 8620
Steel." Materials Science Forum 539-543: 4464-4469.
36. Hernandez-Morales, B., O. Barba-Mendez, A. Ingalls-Cruz, and J. Barrera-

Godinez. "Mathematical Modelling of Temperature and Stress Evolution during
Cooling of a Stainless Steel Navy C-ring Probe." International Journal of
Materials and Product Technology 24, no. 1-4 (2005): 306-318.
37. Gestwa, Wojciech and Malgorzata Przylecka. "The Modification of Sodium

Polyacrylate Water Solution Cooling Properties by AL2O3." Advances in
Materials Science and Engineering 2010, (2010).
38. Campagna, Victoria, Randy Bowers, Derek O. Northwood, Xichen Sun, and Peter
Bauerle. "Distortion and Residual Stresses in Nitrocarburized and Carbonitrided
SAE 1010 Plain Carbon Steel." SAE International Journal of Materials &
Manufacturing 1, no. 1 (2009): 690-696.
39. Grum, J. “Overview of residual stress after quenching part II: factors affecting
quench residual stresses,” International Journal of Materials and Product
Technology 24, nos. 1-4 (2005): 53-97.
40. Pineault, J.A., Belassel, M., and Brauss, M.E., X-Ray Diffraction Residual Stress
Measurement in Failure Analysis, Failure Analysis and Prevention, Vol. 11, ASM
Metals Handbook, ASM International, 2002, p 484-497
41. Ebert, L. J. 1978. "The Role of Residual Stresses in the Mechanical Performance
of Case Carburized Steels." Metall.Trans.A 9A (11): 1537-1551.
85
42. Ahn, S. G., T. W. Kim, S. C. Jung, and Y. W. Kim. "The Effect of Tempering on
Mechanical and Fatigue Properties in Gas-Carburized Cr-Mo Gear Steel." SAE
Transactions: Journal of Materials & Manufacturing (USA) 106, no. 5 (1997):
671-678.
43. Hildenwall, B. and T. Ericsson. "Residual Stresses in the Soft Pearlite Layer of
Carburized Steel." J.Heat Treat. 1, no. 3 (1980): 3-13.
44. Hua, Jiang, Domenico Umbrello, and Rajiv Shivpuri. "Investigation of Cutting
Conditions and Cutting Edge Preparations for Enhanced Compressive Subsurface
Residual Stress in the Hard Turning of Bearing Steel." Journal of Materials
Processing Technology 171, no. 2 (2006): 180-187.
45. Lohe, D., K-H Lang, and O. Vohringer. Residual Stresses and Fatigue Behavior,
Handbook of Residual Stress and Deformation of Steel, ASM International, 2002,
p 27-53
46. Parrish, Geoffrey, The Influence of Microstructure on the Properties of Case-

Hardened Components. Materials Park, Ohio: American Society for Metals, 1980.
47. Skalli, N. and J. F. Flavenot. "Prise En Compte des Contraintes Résiduelles Dans
un Calcul Prévisionnel de Tenue en Fatigue." CETIM Inf. no. 90 (1985): 35-47.
48. Suffredini, R. L. "Factors Influencing Austempering." Heat Treating. 12, no. 1

(1980): 14-16, 18-19.
49. Ren, W., K. Li, and Y-L, Lee. 2004. "Optical Measurement of Residual Stress at
the Deep-Rolled Crankshaft Fillet." SAE Technical Papers.
50. ASTM Standard E915-10, 2010, “Standard Test Method for Verifying the
Alignment of X-Ray Diffraction Instrumentation for Residual Stress
Measurement” ASTM International, West Conshohocken, PA, 2010, DOI:
10.1520/E0915-10, www.astm.org
51. Kovacs, T. Principles of X-Ray Metallurgy. ILIFFE Books Ltd: London, 1969.
52. “Basic X-Ray Powder Diffraction.” Accessed 31 May 2013.

http://xray.tamu.edu/pdf/notes/intro2xrd.pdf
86
53. Pineault, J., M. Belassel, M. Brauss, and J. Ladouceur. "Mapping Residual Stress
Gradients in Automotive Components Via X-Ray Diffraction." SAE Transactions:
Journal of Materials & Manufacturing 116, (2008): 249-252.
54. Sue, J. A. and G. S. Schajer. Stress Determination for Coatings, ASM

International, Materials Park, OH 44073-0002, USA, 1994.
55. Ruud, C. Measurement of Residual Stresses, Handbook of Residual Stress and

Deformation of Steel, ASM International, 2002, p 100-117
56. Jatczak, C. F. "Retained Austenite and its Measurement by X-Ray Diffraction."

Society of Automotive Engineers, Pp.20, 1980 (1980)
57. Convert, F., Miege, B. “Doser Rapidement l’Austénite Résiduelle Avec

Précision.” Journal de Physique IV. No. 6 (1996): 863-878.
58. Arnell, R. D., K. A. Ridal, and J. Durnin. "Determination of Retained Austenite in

Steel by X- Ray Diffraction." J IRON STEEL INST 206, no. 10 (1968): 1035-
1036.
59. Richman, R. H. and R. W. Landgraf. "Some Effects of Retained Austenite on the

Fatigue Resistance of Carburized Steel." Metallurgical Transactions A 6A, no. 5
(1975): 955-964.
60. ASTM Standard E18-12, 2012, "Standard Test Methods for Rockwell Hardness
of Metallic Materials," ASTM International, West Conshohocken, PA, 2012,
DOI: 10.1520/E0018-12, www.astm.org.
61. ASTM Standard E23-12c, 2012, “Standard Test Methods for Notched Bar Impact
Testing of Metallic Materials,” ASTM International, West Conshohocken, PA,
2012, DOI: 10.1520/E0023-12C, www.astm.org.
62. Lyman, Taylor, editor. Metals Handbook Ninth Edition Volume 1 Properties and
Selection: Irons and Steels. Metals Park, Ohio: American Society for Metals,
1978.
87
PUBLICATIONS AND PRESENTATIONS
Publications
1. Clark, A. D., D. O. Northwood, R. J. Bowers, X. Sun, and P. Bauerle.

"Comparison of Austempering and Quench-and-Tempering Processes for
Carburized Automotive Steels." SAE International Journal of Materials and
Manufacturing 6 (2) (2013): 146-153.
Presentations
1. Clark, A. D., D. O. Northwood, R. J. Bowers, X. Sun, and P. Bauerle.

"Comparison of Austempering and Quench-and-Tempering Processes for
Carburized Automotive Steels." SAE 2013 World Congress, Detroit, MI, U.S.A.,
16-18 April, 2013.
88
VITA AUCTORIS
NAME: Andrew David Clark
PLACE OF BIRTH: Toronto, Ontario
DATE OF BIRTH: December 20, 1989
EDUCATION: Ontario Secondary School Diploma

Thornhill Secondary School
Thornhill, Ontario
2007
Bachelor of Applied Science

Honours Mechanical Engineering with Materials Option
Windsor, Ontario
2011
Master of Applied Science

Engineering Materials
2013
89

Comparison of Austempering and Quench-And-Tempering Processes For

Uploaded by

Copyright:

Available Formats

Comparison of Austempering and Quench-And-Tempering Processes For

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Comparison of Austempering and Quench-And-Tempering Processes For

Uploaded by

Copyright:

Available Formats

University of Windsor

Electronic Theses and Dissertations Theses, Dissertations, and Major Papers

Comparison of Austempering and Quench-and-Tempering

Follow this and additional works at: https://scholar.uwindsor.ca/etd

Submitted to the Faculty of Graduate Studies

through Engineering Materials

in Partial Fulfilment of the Requirements for

the Degree of Master of Applied Science

at the University of Windsor

Windsor, Ontario, Canada

© 2013 Andrew Clark

Dr. D.O. Northwood, Co-Advisor

Dr. R.J. Bowers, Co-Advisor

I am aware of the University of Windsor Senate Policy on Authorship and I

2. Declaration of Previous Publication

Thesis Publication title/full citation Publication

Chapter 3,4 Clark, A. D., D. O. Northwood, R. J. Bowers, X. Published

Austempering was examined as a replacement for the current quench-and-

Rockwell C hardness testing were performed. Austempering produced improved

distortion and residual stress characteristics over quench-and-tempering, while

process with austempering.

diffraction services were generously provided.

Graciously acknowledged is the financial support of the Natural Sciences and

Engineering Research Council of Canada (NSERC) and the Ontario College-University

DECLARATION OF CO-AUTHORSHIP / PREVIOUS PUBLICATION .............. iii

Table 3.1 - Compositions of Tested Steels ..................................................................... 25

Table 3.2 - Initial Heat Treatment Conditions ................................................................ 26

Table 3.3 - Heat Treatment Parameters Selected for Further Study................................. 27

Table 3.4 - Identification numbers of Navy C-rings ....................................................... 28

Table 4.1 - Inner Diameter Distortion for 8620 Steel...................................................... 46

Table 4.2 - Outer Distortion for 8620 Steel .................................................................... 49

Table 4.3 - Thickness Distortion for 8620 Steel ............................................................. 51

Table 4.4 - Gap Width Distortion for 8620 Steel ............................................................ 54

Table 4.5 - Flatness Distortion ....................................................................................... 58

Table 4.6 - Roundness Distortion ................................................................................... 60

Table 4.7 - Cylindricity Distortion ................................................................................. 61

Table 4.10 - Retained Austenite Measurement ............................................................... 68

Table 4.11 - Core Hardness............................................................................................ 70

Table 4.12 - Case Hardness............................................................................................ 71

Table 4.13 - Charpy Impact Toughness for 8620 Steel ................................................... 72

Figure 2.1: The Iron-Carbon Phase Diagram [3]. ............................................................. 5

Figure 2.3: TTT Curve for a eutectoid steel [4] ................................................................ 6

Figure 2.7: Quench-and-temper process on a TTT curve [4]. ......................................... 13

Figure 2.9: A Modified Navy C-ring [31, 35]................................................................. 16

Figure 2.11: Bragg's law demonstrated visually [52].......................................................19

Figure 3.2: C-Ring Dimensioning Diagram [31, 35]....................................................... 31

Figure 3.3: Flatness Test Plot Produced by CMM .......................................................... 32

Figure 3.4: Inner Cylindricity Test Plot Produced by CMM ........................................... 32

Figure 3.5: Roundness Test Plot Produced by CMM. ..................................................... 33

Figure 4.2: Austempered 4320 steel case microstructures. ............................................. 41

Figure 4.3: Austempered 8822 steel case microstructures .............................................. 42

Figure 4.4: Case and core microstructures for 4320 steel................................................ 43

Figure 4.5: Inner diameter distortion. ............................................................................. 48

Figure 4.6: Outer diameter distortion. ............................................................................ 50

Figure 4.7: Thickness distortion. .................................................................................... 53

Figure 4.8: Gap width distortion. ................................................................................... 56

Figure 4.9: Flatness distortion. ....................................................................................... 59

Figure 4.10: Roundness distortion. ................................................................................. 61

Figure 4.11: Cylindricity distortion. ............................................................................... 62

Figure 4.12: Surface residual stress for 4320 steel. ......................................................... 64