Introduction and Perspectives

Carbon case hardening, through natural evolu- tered by subzero treatment after quenching. Cold
tion, commercialism, and economics, has be- working by either peening or rolling can modify
come a process for which the possible number of the surface microstructures and have significant
variables is so large that it is hardly likely that bearing on the life of the component, as too can
any two companies will process exactly the same. surface grinding.
There will always be some difference in choice One must not overlook the value of the
of materials, equipment, or technique, and there microstructure and properties of the core or of
will often be differences in the quality of the the influence of inherent features such as
product. There may even be conflict of opinion microsegregation, cleanliness, and grain size.
regarding what is good practice and what is bad, The aforementioned structural variants are the
and what is a valid test and what is meaningless. subject of this review, and where possible, exam-
For each component treated, there is an optimum ples of their effect in terms of properties are
material and process combination, but who given. Those properties mainly referred to are
knows what this is for any given component? bending-fatigue strength, contact-fatigue resis-
Most conflicts stem from there being too great a tance, hardness, and wear resistance. These
choice of materials or process variables and from properties were chosen because it is to promote
the wide range of components that are required one or more of these properties that the carburiz-
to be case hardened. ing treatment is employed. A gear tooth is a good
Despite all this, what the carburizing processes example in which each of these must be consid-
have in common is that they produce at the sur- ered. Some significance has been placed on the
face of the component a layer of carbon-rich residual stresses developed during carburizing
material that after quenching, by whichever because these are additive to the applied stresses.
technique, should provide a surface that is hard.
Regrettably, this is no indication that the case-
hardening process has been successful. Additional
microstructural features may exist along with, or Why Carburize Case-Harden?
instead of, the aimed-for martensite, and these in-
deed can significantly influence the properties of With some through-hardening steels, it is pos-
the component, thereby affecting its service life. sible to develop hardnesses equal to the surface
The microstructural features referred to are in- hardnesses typical of case-hardening parts; how-
ternal oxidation, decarburization, free carbides, ever, machine parts (for example, gears) would
retained austenite, and microcracks in the not be able to transmit as much load as would
martensite. case-hardened parts. This is because case hard-
Further modifications to the martensite in par- ening produces significant compressive-residual
ticular can be effected by tempering, and the pro- stresses at the surface and within the hard case,
portions of austenite and martensite can be al- whereas with through hardening, the residual
2 / Carburizing: Microstructures and Properties

stresses are much less predictable. Furthermore, shallow (0.3 to 0.6 mm, effective), even with long
high-hardness through-hardened steels tend to processing times, for example, 80 hours. The shal-
lack toughness; therefore, in general, through- lowness of the case limits the range of application
hardened and tempered steels are limited to of nitrided steels. For gears, the limiting tooth
about 40 HRC to develop their best strength-to- size is about 2 mm module (12.7 dp) without
toughness properties. To produce compressive- downgrading. However, within its safe range of
residual stresses to a reasonable depth in a application, the case shallowness provides good
through-hardening steel, one must resort to a lo- bending fatigue, contact fatigue, wear, and scuff-
cal thermal hardening process, such as induction ing resistance.
hardening, or an alternative chemicothermal treat- Carbon case hardening can be employed to
ment, such as nitriding. achieve a wide range of effective case depths (up
When induction hardening is used for gears, to greater than 4 mm) in a wide range of steels
for example, the preferred hardness distribution (limiting core carbon is normally 0.25%) with
is generally to have about 55 HRC at the surface surface carbon contents of approximately 0.9%
and 30 HRC in the core (Ref 1); consequently, and hardnesses of about 60 HRC. The contact-
parts so treated do not have a contact strength or fatigue and bending-fatigue strengths are regarded
wear resistance that are quite as good as in car- as superior to induction-hardened surfaces and to
burized and hardened parts. The induction hard- nitride-hardened surfaces (above a certain size
ening process is useful for large parts that need limit). The drawbacks with carbon case harden-
to be surface hardened but would distort or grow ing are distortion, growth, and costs. Distortion
excessively if carburized and hardened. Typical and growth are controlled as much as possible
gear steels surface hardened by induction are during heat treating (by the use of dies and plugs)
4140 and 4340 (initially in the hardened and and finally corrected by a limited amount of grind-
tempered condition), and typical case depths ing. The costs are justified in the product to ob-
range from 1.0 to 3.0 mm. tain a high power-to-weight ratio and durability.
Nitriding is a means of producing a hard sur- An indication of the advantages of case hard-
face with high surface compressive-residual ening, compared with through hardening, is
stresses. It is a subcritical temperature process, shown in the torque-speed plots of Fig. 1 (Ref 2).
and consequently, it is an essentially distortion- Here, the safe operating zone for case-hardened
and growth-free process. The degree of hardening gear sets is much greater than it is for through-
relates mainly to the chromium content of the hardened steels. This means that to transmit the
steel so that a carbon steel will nitride harden same power at a given speed, a set of case-hardened
only a little. Steel 4140 will harden to about 600 gears can be significantly smaller and/or lighter
to 650 HV, and a 3% Cr-Mo-V steel will than a set of through-hardened gears. Alterna-
achieve more than 800 HV. Unfortunately, the tively, size for size, the case-hardened gear set
cases that can be achieved due to nitriding are will be much more durable.

Precision hobbed shaved

;; ;;
gears, ground gears, all gears
Limited
Limitedby
bywear
wear without involute correction
3.5 3.5
Relative torque capacity of gear set

Relative torque capacity of gear set

Limited Involute
Involutecorrected
correctedprecision
precision
Limitedby
byscuffing
scuffing(precision
(precision ground
3 hobbed
hobbedororshaved
shavedgears)
gears) 3 groundgears
gears
Limited
Limitedpitting
pitting
Limited
Limitedby
bywear
wear
2.5 2.5 Limited
Limitedtooth
toothfracture
fracture
Good
Goodquality
qualitycommerical
commericalhobbing
hobbing
2 or
orshaping

;;
shaping 2
Limited
Limitedtooth
toothfracture
fracture
1.5 Limited 1.5
Limitedpitting
pitting Limited
Limitedbyby
1 1 scuffing
scuffing
Limited
Limitedby
by Safe
Safeoperation
operation
0.5 scuffing
scuffing 0.5
Safe
Safeoperation
operation
0 0
10 102 103 104 105 106 10 102 103 104 105 106
Pitch line velocity, ft/min Pitch line velocity, ft/min

(a) (b)
Failure regions of industrial and automotive spur and helical gears. (a) Through hardened, 180–350
Fig. 1 HB. (b) Precision gears, surface hardened
 Introduction and Perspectives / 3

Variability grade and fastidious for the precision gear. In

most cases, the heat-treatment procedures will be
to the same standard, and the heat treater will
Over the past several decades, the steelmaking perform in the best way possible every time.
industry has moved from basic open-hearth steel
manufacturing to processes such as VIM/VAR;
consequently, the quality and consistency of
steels have improved appreciably. Heat-treatment
Laboratory Tests
furnaces have improved, as have atmosphere and
temperature control systems. Additionally, the
gas-metal reactions, carbon diffusion, and other Laboratory tests to determine the effect of
processes that take place during the carburizing metallurgical variables, for example, carbides,
and hardening of steels have become much better retained austenite, and core strength, are very
understood. Add to these factors the introduc- useful and have contributed appreciably to the
tion of quality systems that favor process and understanding of the influences of metallurgical
product consistency, and, all in all, there has features on material properties. However, there
been considerable improvement (a far cry from are problems associated with laboratory testing
the days of pack carburizing). Having said that, that must be recognized and, where possible, al-
absolute precision is not attained because, lowed for. One problem is that the test specimen
among other reasons, exact steel compositions and method of loading often bear little relation-
are impossible to achieve, and atmosphere con- ship to the machine part and service conditions
trol during carburizing is, at best, often only able they are supposed to represent. Apart from that,
to produce surface carbon contents of ±0.05% of test pieces are often small in section so that the
the target value. Therefore, some metallurgical proportion of case to core can be high, and the
variability must be tolerated. microstructure can be martensitic throughout the
The grade of steel for a given machine compo- test section. The effect of these factors on the re-
nent design, the carburized case depth, and the sidual stress distribution and on the contribution
target values of surface carbon adopted by a of metallurgical features can limit the value of
manufacturer/heat treater are based on experi- the test findings. Another problem is isolating
ence, design procedures, and guidelines provided the metallurgical feature to be studied; generally,
in national or international standards, and per- when conducting a test to determine the effect of
haps on adjustments indicated by laboratory a process variation or metallurgical feature on
test results. It is difficult to determine the opti- some property, the researcher attempts to isolate
mum metallurgical condition for a given situa- that test subject. Sometimes this is easy, for ex-
tion; what is optimum in terms of surface carbon ample, when determining the effects of temper-
or case depth for a gear tooth fillet is different ing or subzero treatment. Other times, it is not so
from what is optimum for a gear tooth flank. In easy. For example, to determine the influence of
fact, even if the optimum condition is known retained austenite on bending-fatigue strength, a
for any given situation (and this can vary from large batch of test pieces are prepared. Half are
situation to situation), the heat treater probably left as carburized and hardened with a high re-
could not provide it due to the variability de- tained austenite content at the surface; the other
scribed in the previous paragraph and the fact half is refrigerated to transform much of the
that most heat treaters are happy to get surface surface retained austenite. This is a common
hardnesses within a fairly wide 58 to 62 HRC method of arriving at two retained austenite lev-
range, and effective case depths within a 0.25 mm els, but what exactly is being studied? Is it the ef-
range. Further, without considering section size, fect of retained austenite, or is it the effect of
the previously mentioned composition variabil- subzero treatment? It is agreed that there are two
ity could give batch-to-batch core-strength varia- austenite levels. Is it the difference in austenite
tions within a 20 ksi band. Hence, the ideal and levels that causes a difference of fatigue strength,
the achievable are often different. Gear standards or is it the effect of the new martensite and its as-
cater to different classes of gears, and these dif- sociated short-range stresses induced by refriger-
ferent classes require different degrees of dimen- ation that are responsible for the difference? The
sional precision and finish, as well as different manufacture of batches of test pieces that are
standards of inspection. It is unlikely, however, identical apart from the presence or absence of
that the heat treater will be lax for the lowest network carbides is another example. One can
4 / Carburizing: Microstructures and Properties

standardize surface carbon content and vary the tigue) have been derived that are somewhat less
heat treatment, or one can standardize heat treat- than the actual failure values. The basic allowables
ment and vary the surface carbon content. Either are published in the gear standards (e.g.,
way, there will be differences other than the car- ANSI/AGMA 2001 or ISO 6336) (Table 1a and
bide network. Nevertheless, laboratory testing b). One should consider that for full-scale gear
provides trends and indicates whether a metallur- testing, the metallurgy of the tested gears might
gical feature will have a small or a large effect on be typical of one heat treater’s quality, which
the property under study. could rate either high or low against other heat
treaters’ qualities. This is another reason for set-
ting the design allowables lower.
Designers also incorporate into a design safety
Design Aspects factors that will account for any adverse effects
of material and manufacturing variability. There-
Laboratory test pieces are designed and loaded fore, there are probably numerous case-hardened parts
to fail. Machine parts, on the other hand, are de- performing satisfactorily in service with surface
signed and loaded not to fail. The basic allow- microstructures that contain adverse metallurgi-
able stresses used by gear designers have been cal features. For example, the high-temperature
conservative in order to acknowledge that design transformation products that accompany internal
procedures are not precise enough to cater to the oxidation tend to be frowned upon, yet there are
very wide range of gear designs, and that mate- numerous case-hardened gears in service with un-
rial variability and process variability do exist. ground roots that, therefore, contain degrees of
These basic allowable stresses are derived from internal oxidation. If the test gears from which
actual gear tests and are set at a lower value than the basic allowable stresses were derived had un-
that of failure stress. For example, in Fig. 2, the ground roots and fillets, then internal oxidation
surface-hardened test gears failed due to tooth will be accounted for anyway. A metallurgical
pitting at contact stresses of 1400 to 1500 MPa. feature might indeed lower the strength of a part
These tests represent nitrided marine and indus- (according to laboratory test results), but the ap-
trial gears that have, in this instance, a design plied service stresses must be high enough for
limit of about 1000 MPa (Ref 3). Comparable that feature to be significant and cause failure. If
gear tests have been conducted for case- hard- the basic allowable stress and the gear designer’s
ened automotive gears and aerospace gears. safety factor together reduce the service stresses
From these tests, appropriate allowable stress val- to, say, half the failure strength of the part, but
ues (for both bending fatigue and contact fa-

Table 1(a) Basic allowable stress numbers for

2000 gears, ISO 6336-5 1996
Contact stress Bending stress
Quality grade limit (σH ), MPa limit (σF ), MPa Hardness, HV

1500 Carburized and hardened

Contact stress, MPa

ME 1650 525 670–745

Contact stress
limit DNV
MQ 1500 452–500(a) 645–745
ML 1300 315 615–800
1000 Induction hardened
ME 1275–1330 375–405 515–620
Nitrided gears MQ 1160–1220 360–270 515–620
Failed, full-scale gear test ML 960–1090 225–275 490–655
500 Not failed, full-scale gear test
Not failed, industrial gears Gas nitrided, through hardened and tempered
(4–15 years service) ME 1210 435 500–650
Not failed, marine gears MQ 1000 360 500–650
(1.5–15 years service)
ML 785 255 450–650
0
0 0.005 0.01 0.015 0.02 Gas nitrided, nitriding steels
Ratio of effective case depth (500 HV) ME 1450 470 700–850
to relative radius of curvature MQ 1250 420 700–850
ML 1125 270 650–850
Fig. 2 The results of full-scale gear tests (failure by
tooth pitting) and the typical design stresses used for in- Stresses are shown in MPa, and all hardness values are converted to
dustrial and marine gears. DNV, Det Norske Veritas. HV. Designers should refer to the appropriate standard. (a) Varies with
core hardness and/or core strength
Source: Ref 3
 Introduction and Perspectives / 5

the heat treatment has induced a serious adverse ratios and, hopefully, reduced costs. Therefore,
metallurgical feature with a strength reduction the metallurgists and heat treaters must continue
potential of, say, 30%, there still might not be a to contribute to the cause.
problem (Fig. 3). However, if something should Currently, it is believed that the limitations of
go wrong, for example, if a bearing begins to de- the conventional case-hardening steels are fairly
teriorate or the gear is slightly misaligned, in- well understood. Any other gains must be made
creasing the tooth stress, then failure is more through design and process refinements (consis-
likely to occur. tency and accuracy) sufficient to enable revision
It is not suggested here that one should ignore of the design allowables.
the metallurgical condition, or that quality con- The future might never provide a case-
trol should be relaxed because design, to some hardening steel that is superior in all respects to
extent, accommodates metallurgical variability. the conventional grades. Even if it did, the cost
On the contrary. It could be that on many occa- of the steel might limit its use to very specialized
sions the designer’s generosity has, in effect, applications. However, it is possible to design a
“saved face” for those responsible for the metal- steel that is superior with respect to one property.
lurgical quality. If the metallurgical variability The newer grades of special-purpose aerospace
could be reduced across the board, and improved gear steels for use at high operating speeds and
quality and quality consistency could be guaran- temperatures exemplify this designing for pur-
teed, then perhaps the basic allowable stresses pose. Examples of such steel are Pyrowear Alloy
could be increased a little. If nothing else, prod- 53 (Carpenter Technology Corp., Wyomissing,
uct reliability would be improved. Designers PA), CBS-1000M VIM-VAR (Timken Latrobe
strive to improve their design procedures, manu- Steel Co., Latrobe, PA), CBS-600 (Timken Co.,
facturers aim to produce levels of accuracy and Canton, OH), Vasco X2-M, and Latrobe
finish the designer specifies, and lubrication en- CFSS-42L, for which the steel compositions and
gineers seek to improve their products. Together heat-treatment operations depart sufficiently
these efforts will lead to better power-to-weight from the conventional. Previously, SAE 9310
steel was preferred by the aerospace industry for

Table 1(b) Basic allowable stress numbers for

gears, AGMA 2001-C95 160
Quality Contact stress limit Bending stress limit
grade (SAC), MPa (SAT), MPa Hardness, HV Example of a bending-fatigue
140 curve for case-hardened gears
Carburized and hardened
3 1900 520 650–800
2 1550 450 or 480(a) 650–800 120
1 1240 380 600–800
Induction hardened

;;
100 30% loss of fatigue
2 1310 152 515 minimum limit due to an adverse
1345 152 580 metallurgical feature
Stress, ksi

1 1172 152 515

1210 152 580 80
Gas nitrided, hardened and tempered, 4140 and 4340 Allowable stresses for case-hardened
3 1210 … 460 minimum 60 steel gears (AGMA glass 3)
1240 … 485
2 1125 317–372(b) 460
1160 317–372(b) 485
40
1 1030 234–276(b) 460
1070 234–276(b) 485
Gas nitrided, 21 2 Cr steel Probable main design
20 range for infinite life
3 1300 420–440(b) 580 minimum
1490 420–440(b) 690
2 1190 395–400(b) 580 0
1350 395–400(b) 690 103 104 105 106 107 108 109
1 1070 280–310(b) 580 Stress cycles
1210 280–310(b) 690
Fig. 3 Theoretically a “safe” gear design can accom-
Stresses are shown in MPa, and all hardness values are converted to modate the presence of an adverse metallurgical feature;
HV. Designers should refer to the appropriate standard. This table is
however, there may be other adverse factors involved
for spur and helical gears. (a) Depends on bainite content. (b) Varies
that also erode the difference between the fracture stress
with core hardness and/or strength
and the allowable stress.
6 / Carburizing: Microstructures and Properties

gears, but its limitations (questionable hot which have a high potential to occur in
strength, for example) inhibited design progress. high-speed, high-temperature gearing (Fig. 1).
The high-temperature limitations of lubricants for This resistance may, to some extent, make up for
high-speed, high-temperature gearing is another the limitations of the lubrication.
factor to consider. The new grades of steel are Metallurgy is only one factor in a bigger
designed to maintain their strength at operating picture that includes machine and component
temperatures and resist scoring and scuffing, design, manufacturing accuracy, machine

2530 26 367

2431 24 352.5

2327 22 337.5

2220 20 321.8

2106 18 305.5

1985 16 288
Compressive stress, MPa

Fatigue limit, SH × 1000

Compressive stress, ksi

1857 14 ( ) 269

1718 12 249

1569 10 227.6

1404 8 203.6
( )

1216 6 ( ) 176
Through-hardened steels
Flame-hardened steel
993 4 Induction-hardened steel 144
Gas-nitrided and salt-bath nitrided steel
Sulphinuz-treated steel
702 2 Gas-nitrided (80 h) steel 102
( ), ( ) Maraging steel
Gas-carburized and hardened
Gas-carburized, hardened and tempered
0 0
0 0 20 40 60 80 100 120 140 160 180
Core strength, tsi

Steel Effective case depth, mm (in.)

Through hardened (various) …
Flame hardened (PCS) …
Induction hardened (4340) 3.75 (0.15)
Gas nitrided and salt-bath nitrided 0.14 (0.005)
Sulphinuz treated 0.17 (0.007)
Gas nitrided (80 h) (3%Cr-Mo) 0.35 (0.015)
Maraging (x) 0.14 (0.005)
Gas carburized, hardened, and tempered (Ni-Cr) 1–1.5 (0.04–0.06)
Gas carburized and hardened (Ni-Cr) 1.5 (0.06)

Fig. 4 Effect of core strength and case depth on the rolling-contact fatigue limit of gear steels. Tests in-
volved two 4 in. disks driven by a 2 in. roller. Test piece may have been either one of the disks or the roller.
Relative radius of curvature, 2/3. SH units = lb/in. of face width divided by the relative radius of curvature.
 Introduction and Perspectives / 7

assembly, lubrication, application, machine use chosen case depth must be adequate to resist the
or abuse, and maintenance (or lack of it). This stress at the dedendum-pitch line area.
book considers some of the current knowledge The contact stress increases with transmitted
regarding the metallurgy of case-hardened steel load so, strictly speaking, the case depth should
parts and what effects or trends the various met- be determined by the load. Using the shear-fatigue
allurgical features have on the properties of such strength (ultimate tensile strength × 0.34) of the
parts. However, it focuses on conventional material as opposed to shear stresses due to load-
case-hardening steels and processing and, there- ing appears to give some conflicting results;
fore, might not be as helpful to designers and us- therefore, it is not clear on which shear stresses
ers of new alloy grades. the case depth requirement should be based. For
example, if the 45° shear stresses (yz) are con-
sidered in conjunction with the test results shown
Case-Depth Specifications in Fig. 4, it is found that, for the 80 hour-nitrided
surfaces, the predicted fatigue limit is about half
At the dedendum-pitch line area of a gear of the value determined by testing. On the other
tooth, there is a smaller radius of curvature than hand, the fatigue limits for the carburized, hard-
at locations above the pitch line. Consequently, ened, and tempered surfaces (100 to 200 °C) and
the contact band there tends to be narrower than for induction-hardened surfaces are better pre-
at the addendum so that for a given load, the con- dicted (Fig. 5). The orthogonal shear stresses
tact stresses will be higher. For that reason, the (ortho), however, predict fairly well the fatigue

120

100 Induction
hardened
Carburized hardened, and
tempered at 0, and and 150 °C
100,100,
80
Tempered at 200 °C
Stress (τyz), ksi

60
Tempered at 250 °C
28,000 SH

24,000 SH
40
Nitrided 20,000 SH

16,000 SH
215 ksi core 12,000 SH 14,000 SH
20
Nitrided steels Maraging 8,000 SH 10,000 SH
(see also Fig. 6) 145 ksi core
110 ksi core
0
0 0.010 0.020 0.030 0.040 0.050 0.060 0.070 0.080 0.090 0.100 0.110 0.120
Distance from surface, in.

Process Predicted fatigue limit, SH Actual fatigue limit, SH

Carbon case hardened, untempered 24,000 18,000
Tempered at 100 °C 24,000 22,000
Tempered at 150 °C 24,000 23,000
Tempered at 200 °C 22,000 25,000
Tempered at 250 °C 20,000 26,800
Induction hardened 18,000 ~18,000
Nitrided steels Prediction equals about one half of actual

Fig. 5 Plots of shear-fatigue strength (from hardness) against plots of shear stresses, yz, in rolling-contact
tests. Predicted and actual fatigue limit values are in close agreement for carburized steels but not for the
four nitrided steels. Relative radius of curvature, 2/3. SH units = lb/in. of face width divided by the relative ra-
dius of curvature.
8 / Carburizing: Microstructures and Properties

limits for the nitrided surfaces but overestimate sidual stresses were not measured in either in-
the fatigue limits for the case-hardened and the stance, it is nevertheless likely that the roller
induction-hardened surfaces (Fig. 6). From these tempered at 250 °C had the lowest compres-
apparently conflicting results, it is difficult to sive-residual stress in the case, and the untem-
draw any meaningful conclusions that would pered roller had the highest (see Fig. 7.12). This
help determine the appropriate hardness profile implies that compressive-residual stresses
and case depth for a given application. might not be beneficial where rolling contact is
The relationship of residual stresses to rolling involved—where the fatiguing actions are
contact fatigue is also unclear. The table in Fig. 4 subsurface but still in the case. Therefore, this
shows that for the case-hardened tests, the un- further complicates arriving at a theoretical solu-
tempered roller produced the lowest fatigue tion for determining adequate hardness profiles
limit, and the roller that had been tempered at and case depths. Fortunately, there is still the
250 °C produced the highest value. Although re- well used case depth-to-tooth diametrical pitch
relationship to fall back on, even if it is not
strictly correct (Fig. 7).
Interestingly, with rolling-contact fatigue tests
120 of shallow-cased surfaces (i.e., when the depth of
maximum hertzian shear stress is deeper than the
SH
SH
32,000
32,000
effective case depth), there is no work hardening
Case
100 hardened at the case-core junction up to the fatigue limit.
28,000 Induction At stresses above the fatigue limit, work harden-
hardened ing does occur, and the extent of the working
24,000 (hardness and depth) increases with the contact
80 Nitrided 215
ksi core
stresses.
Stress (τortho), ksi

20,000
16,000 Nitrided
maraging steel
60 14,000
8,000 Nitrided 145 30
ksi core hemin = 0.119935 × Pnd−0.86105
20 Normal case depth
6,000
40
Normal diametral pitch, Pnd

10
9
Nitrided 110 8 hemin = 0.264693 ×
20 ksi core 7 Pnd−1.12481
6
5 Heavy case depth

4
0 3
0 0.010 0.020 0.030 0.040 0.050
Distance from surface, in. 2

Predicted Actual
Process fatigue limit, SH fatigue limit, SH 1
Carbon case hardened >31,000 20,000–24,000 10−3 2 3 5 10−2 2 3 5 10−1 2 3 1
Induction hardened 27,000 18,000 Minimum effective case depth, (he min), in.
(a)Nitrided for 80 h, 25,000 25,000
215 ksi core
(b)Nitrided for 80 h, 16,500 14,000 Fig. 7 Minimum effective case depth for carbu-
maraging steel rized gears, he min. The values and ranges shown on
(c)Nitrided for 80 h, 14,500 15,000 the case-depth curves are to be used as guides. For
145 ksi core gearing in which maximum performance is re-
(d)Nitrided for 80 h, 7,000 7,000–9,000 quired, detailed studies must be made of the appli-
110 ksi core cation, loading, and manufacturing procedures to
obtain desirable gradients of both hardness and in-
Fig. 6 Plots of shear-fatigue strength against plots ternal stress. Furthermore, the method of measuring
of shear stresses, τortho, in rolling-contact tests. In the case, as well as the allowable tolerance in case
contrast to Fig. 5, predicted and actual fatigue limit depth, may be a matter of agreement between the
values are in good agreement for the four nitrided customer and the manufacturer. Effective case
steels but not the other steels. Relative radius of cur- depth is defined as depth of case with a minimum
vature, 2/3. Shear fatigue strength is ultimate tensile hardness of 50 HRC; total case depth to core carbon
strength × 0.34. SH units = lb/in. of face width di- is approximately 1.5 × effective case depth. See
vided by the relative radius of curvature. ANSI/AGMA 2001-C 95.
 Introduction and Perspectives / 9

Eutectoid Carbon Content martensite and in any retained austenite. Consider

then: is a eutectoid carbon martensite the best to
The requirements and information in any stan- provide all the properties sought for a given appli-
dard are, in general, readily understandable and cation? Or is it the best carbon content for holding
realistic, as they should be. Unfortunately, there the retained austenite to a low value or for develop-
are exceptions. For example, the surface carbon ing a better case toughness? Would a case-hard-
requirement for carburized gears as set out in ened 9310 steel gear with a surface carbon content
ISO 6336-5 1996 is “Eutectoid carbon % of 0.55% be regarded as fit for service even
+0.20%, –0.1%.” The standard does not justify though it might satisfy the case carbon require-
the use of the term eutectoid. It does not provide ments of ISO 6336-5 (1996)?
a list of case-hardening steels along with a repre- To establish where the eutectoid carbon con-
sentative value of eutectoid carbon for each steel, tent figures in deliberation regarding property
nor does it provide an empirical formula for de- optimization for case-hardened parts (and indeed
termining the eutectoid carbon. It is, therefore, it may have a place), there is little alternative but
unhelpful and unworkable as it stands. However, to establish eutectoid carbon data for each steel.
it is understood that the standard is to be revised For this, it may not be necessary to go through
to correct the problem. the complex procedure of determining accurate
The term eutectoid carbon content refers to equilibrium diagrams. Instead, a set procedure
the carbon content that produces only a pearlitic could be devised in which, for example, a 30 mm
matrix microstructure as a result of an extremely bar is carburized to, say, greater than 1% surface
slow cool through the Ac3 or Accm to Ac1 tem- carbon content and cooled, or heat treated to pre-
perature range. A steel with less than the cipitate the excess carbon as carbides. The bar is
eutectoid carbon content (hypoeutectoid) con- then cut into two: one half is used to determine
tains pearlite with some ferrite, whereas a steel the carbon gradient and the other is used as a
with more carbon than the eutectoid carbon con- metallographic sample to determine the depth of
tent (hypereutectoid) contains some carbide carbide penetration. The two sets of data are then
along with pearlite, again due to very slow cool- brought together to give a value of carbon at
ing. Each steel grade has its own eutectoid car- which, under the set conditions, carbides just ap-
bon content, and considering the whole range of pear. This could then be referred to as the “ap-
conventional case-hardening steels, the eutectoid parent eutectoid.” Only with such information
carbon contents could easily vary between 0.45 could the merits of the case carbon requirement
and 0.8%. In case-hardening practice, the cool- of the ISO 6336 standard be assessed.
ing rates employed, even slow cooling from car-
burizing, are much faster than the cooling rates
researchers would use to determine the eutectoid REFERENCES
carbon for an equilibrium diagram. Rapid cool-
ing, typical of commercial quenching, can sup- 1. G. Parrish, D.W. Ingham, and J.M. Chaney, The
press the formation of ferrite in lean-alloy steels Submerged Induction Hardening of Gears, Parts 1
and 2, Heat Treat. Met., Vol 25 (No. 1) 1998, p 1–8,
within about 0.2% C less than the eutectoid and and Vol 25 (No. 2), p 43–50
suppress the carbide formation in that steel when 2. M. Jacobson, Gear Design: Lessons from Failures,
the carbon is up to about 0.2% above the Automot. Des. Eng., Aug 1969
eutectoid. Suppression of ferrite or of carbide 3. I.T. Young, The Load Carrying Capacity of Nitrided
means that the carbon will be in solution in the Gears, BGMA, London, 1982
 ASM International is the society for materials engineers and scientists, a
worldwide network dedicated to advancing industry, technology, and
applications of metals and materials.

ASM International, Materials Park, Ohio, USA

www.asminternational.org

This publication is copyright © ASM International®. All rights reserved.

Publication title Product code

Carburizing: Microstructures and Properties 06677G

To order products from ASM International:

Online Visit www.asminternational.org/bookstore

Telephone 1-800-336-5152 (US) or 1-440-338-5151 (Outside US)

Fax 1-440-338-4634

Customer Service, ASM International

Mail
9639 Kinsman Rd, Materials Park, Ohio 44073, USA

Email [email protected]

American Technical Publishers Ltd.

27-29 Knowl Piece, Wilbury Way, Hitchin Hertfordshire SG4 0SX, United Kingdom
In Europe
Telephone: 01462 437933 (account holders), 01462 431525 (credit card)
www.ameritech.co.uk

Neutrino Inc.
In Japan Takahashi Bldg., 44-3 Fuda 1-chome, Chofu-Shi, Tokyo 182 Japan
Telephone: 81 (0) 424 84 5550

Terms of Use. This publication is being made available in PDF format as a benefit to members and customers of ASM
International. You may download and print a copy of this publication for your personal use only. Other use and distribution is
prohibited without the express written permission of ASM International.

No warranties, express or implied, including, without limitation, warranties of merchantability or fitness for a particular
purpose, are given in connection with this publication. Although this information is believed to be accurate by ASM, ASM
cannot guarantee that favorable results will be obtained from the use of this publication alone. This publication is intended for
use by persons having technical skill, at their sole discretion and risk. Since the conditions of product or material use are
outside of ASM's control, ASM assumes no liability or obligation in connection with any use of this information. As with any
material, evaluation of the material under end-use conditions prior to specification is essential. Therefore, specific testing
under actual conditions is recommended.

Nothing contained in this publication shall be construed as a grant of any right of manufacture, sale, use, or reproduction, in
connection with any method, process, apparatus, product, composition, or system, whether or not covered by letters patent,
copyright, or trademark, and nothing contained in this publication shall be construed as a defense against any alleged
infringement of letters patent, copyright, or trademark, or as a defense against liability for such infringement.