Canale L.C.F., Mesquita R.A., Totten G.E. Failure Analysis of Heat Treated Steel Components

Подождите немного. Документ загружается.

Figure 38 shows elongated microvoids,

obtained by SEM with secondary electrons, in

the vicinity of the tip of one of the cracks that

propagated in the fractured component. The

alignment of the discontinuities generate a

favorable path for material cracking. The voids

that are already interconnected by material

cracking are shown by the arrows in Fig. 38.

In Fig. 39, the 52100 steel microstructure in

the central axle regi on, cut in the longitudinal

Fig. 33

Cross section of a catastrophically fractured axle.

The arrow at left shows the main fracture plane

(longitudinal), and the arrow at right shows the starting point of

the fracture in the circular cross section.

Fig. 34

Detail of the starting point of brittle fracture in the

circular cross section of the component (arrow at

bottom). The clustered arrows show the brittle crack tip front.

Fig. 35

Fractured axle microstructure at the center of the

component thickness. Etched with nital. Original

magniﬁcation: 400·

Fig. 36

Inclusion-like microdefects detected in the vicinity

of a crack in the fractured axle, located at the center

of the component

Fig. 37

Slim, cracklike inclusions in the 52100 steel. The

inclusions are oriented in the longitudinal direction

of the component. No etch

174 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_151-176.pdf/Chap_05/ 18/8/2008 3:11PM Plate # 0 pg 174

plane, is shown in greater detail. An essentially

pearlitic matrix developed with cementite pre-

cipitates (Fe

C) in the globular form (solid

arrows). Free cementite exists in the pearlitic

colonies contour in the form of veins or platelets

(white arrows) that offer an easy path for brittle

crack propagation in the material.

The absorption spectra obtained by EDX of

the 52100 steel conﬁrm that the plate pre-

cipitates, shown in Fig. 39, are mad e of iron

carbide or cementite in the free form (Fe

C).

However, the absorption spectra obtained in

microanalyses of the grayish material inside the

elongated microcavities, shown in Fig. 36 to 38,

indicated it is made essentially of iron oxide. At

ﬁrst, the hypothesis that this contaminant comes

from, for example, the atmospheric oxidation

after the fracture event of the component, was

discarded, since the inclusions measured by

microprobe were completely isolated inside the

metallic matrix, witho ut any possibility of

reaction with the environment.

It was concluded that the raw material used to

manufacture the fractured axle was probably

contaminated with iron oxide. The contaminant

was in the form of elongated inclusions, aligned

in the longitudinal part direction, making an

easy path for main crack propagation (long-

itudinal). The elongated format provided the

inclusions the capability to concentrate high

tensile stresses and then transform them into

potential crack nucleation sites.

Microanalysis also conﬁrmed the existence

of free cementite in the pearlitic grain contours

that form the 52100 steel. The presence of this

fragile phase may have contributed, to a certain

extent, to the intergranular secondary brittle

crack propagation in the catastrophic frac-

ture of the component during annealing heat

treatment.

Fig. 38

Alignment of the elongated inclusions (oriented in

the longitudinal direction of the part) act as an easy

propagation path in the 52100 steel axle. The main fracture

direction, that is, longitudinal, corresponds exactly to the elon-

gation and inclusion alignments. The arrows point to the exis-

tence of cracks among the microcavities that compose the

inclusions. SEM Original magniﬁcation: 100·;20kV

Fig. 39

52100 steel microstructure in the center of the

component thickness. Etched with nital. Solid arrows

point to free cementite in the globular form, and white arrows

point to Fe

C in the form of platelets in the pearlite contour.

Original magniﬁcations: (a) 3000·. (b) 10,000· . (c) 18,000·

Failures from the Casting Process / 175

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_151-176.pdf/Chap_05/ 18/8/2008 3:11PM Plate # 0 pg 175

ACKNOWLEDGMENTS

Thanks to the Department of Materials, Aero-

nautics and Automotive Engineering of the School of

Engineering of Sa

o Carlos, University of Sa

o Paulo,

on behalf of Professor Dr. Dirceu Spinelli, for the

collaboration on failure analysis case studies.

REFERENCES

1. Defects and Distortion in Heat-Treated

Components, Heat Treating, Vol 4, ASM

Handbook, ASM International, 1991,

p 1320–1325

2. R.E. Reed-Hill, Physical Metallurgy Prin-

ciples, 2nd ed., Van Nostrand, 1982

3. Casting Design, Casting, Vol 15, ASM

Handbook, ASM International, 1988,

p 1301–1322

4. Dimensional Tolerances and Allowances,

Casting, Vol 15, ASM Handbook, ASM

International, 1988, p 1336–1347

5. Quenching and Control of Distortion, ASM

International, 1988

6. Residual Stress, Heat Treating, Vol 4, ASM

Handbook, ASM International, 1991,

p 1325–1343

7. M.T. Milan, O. Maluf, D. Spinelli, and

W.W. Bose Filho, Metais—Uma Visa

Objetiva (Metals A Vision Object)

Suprema, 2004, p 148–149, 161–162

8. Gases in Metals, Casting, Vol 15, ASM

Handbook, ASM International, 1988,

p 175–189

9. Inclusi on-Forming Reactions, Casting, Vol

15, ASM Hand book, ASM Internati onal,

1988, p 190–211

10. R.D. Pelke and J. Elliott, Trans. TMS-

AIME, Vol 227, 1963, p 894

11. P.C. Glaws and R.J. Fruehan, Metall. Trans.

B, Vol 17, 1986, p 317

12. R.J. Fruehan, B. Lally, and P.C. Glaws,

Proceedings of the Fifth International Iron

and Steel Congress (Washington, D.C.),

Iron and Steel Society of AIME, 1986

13. A. Kagawa and T. Okamoto, Trans. Jpn.

Inst. Met., Vol 22 (No. 2), 1981, p 137

14. J. Campbell, Castings, Butterworth-

Heinenmann, 1993

15. J.G. Davies, Solidiﬁcation and Casting,

Applied Science, 1973

16. AFS Inclusion Atlas Homepa ge, http://

neon.mems.cmu.edu/afs/afs2/ (Accessed on

March 2005)

17. J.A. Horwath and G.M. Goodrich, Micro-

Inclusion Classiﬁcation in Steel Casting,

AFS Trans., 1995, p 495–510

18. J.M. Svoboda, R.W. Monroe, C.E. Bates,

and J. Grifﬁn, Appearance and Composition

of Macro-Inclusions in Steel Castings, AFS

Trans., 1987, p 187–202

19. C.R. Wanstall, J. Grifﬁn, and C.E. Bates,

“Clean Steel Cast Technology,” Research

Report 106, Steel Founders Society of

America

20. R. Sankaranarayanan and R. Guthrie, Slag

Entrainment through a “Funnel” Vortex

during Ladle Teeming Operations, Pro-

ceedings of the International Symposium on

Developments in Ladle Steelmaking and

Continuous Casting, Aug 1990 (Ontario),

CIM, 1990, p 66–87

21. R. Sankaranarayanan and R. Guthrie, A

Laboratory Study of Slag Entrainment

during the Emptying of Metallurgical Ves-

sels, Steelmaking Conference Proceedings

(Ontario), 1992, p 655–664

22. K. Noguchi et al., Zairyo to Purosesu (Curr.

Adv. Mater. Process.), Vol 4, 4th ed., 1991,

p 1194–1197

23. H. Nakamura, S. Kohira, J. Kubota, T.

Kondo, M. Suzuki, and Y. Shir atani, Tech-

nology for Production of High Quality Slab

at High Speed, Steelmaking Conference

Proceedings (Ontario), 1992

24. I. Manabu, S. Yutaka, O. Ryusuke, and M.

Zen-ichiro, Evaluation of the Critical Gas

Flow Rate Using Water Model for the

Entrapment of Slag into a Metal Bath Sub-

ject to Gas Injection, Tetsu-to-Hagane

(J. Iron Steel Inst. Jpn.), Vol 79 (No. 5), p 33

25. J.M. Harman and A.W. Cramb, A Study on

the Effect of Fluid Physical Properties on

Droplet Emulsiﬁcation, Steelmaking Con-

ference Proceedings, 1996, p 773–784

176 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_151-176.pdf/Chap_05/ 18/8/2008 3:11PM Plate # 0 pg 176

Sources of Failures in Carburized

and Carbonitrided Components

Małgorzata Przyłe˛cka and Wojciech Ge˛stwa, Poznan University

of Technology

Lauralice C.F. Canale, University of Sa˜o Paulo

Xin Yao, Portland State University

G.E. Totten, Associac¸a˜o Instituto Internacional de Cieˆncia and

Portland State University

MANY COMPONENTS, such as fasteners,

crankshafts, camshafts, bearings, and others, re-

quire a differentiated response of the surface and

core to external loading. This can be accom-

plished by surface (case) hardening methods

such as induction and ﬂame hardening or by

surface diffusion processes such as carburizing

and carbonitriding. Raja et al. have reported that

case carburizing is one of the most common heat

treatments for steel, accounting for 50% of all

surface treatments (Ref 1). Case carburizing

involves the creation of a gradient that exhibits

high hardness, brittleness, and strength in the

surface and greater toughness and ductility in

the softer core in order to provide optimal

(Ref 2):

 Wear resistance

 Resistance to scoring

 Bending and/or torsional fatigue strength

 Rolling-contact fatigue strength

These properties are optimized by maximiz-

ing surface compressive stresses, and carburiz-

ing is one of the most effective and commonly

used methods to impart compressive stresses to

the surface of a component (Ref 3). The focus of

this chapter is on carburized and carbonitrided

materials.

Gas carburizing, which is the most widely

used carburizing process, is a surface diffusion

process where the carbon concentration in a

surface layer (case) of a steel matrix that is

predominantly iron, chromium, and nickel is

increased by heating the component at approxi-

mately 850 to 950



C with endothermic gas

(Endogas), which is a blend of carbon mon-

oxide, hydrogen, and nitrogen (with smaller

amounts of carbon dioxide, water vapor, and

methane). Endogas is produced by reacting a

hydrocarbon gas, such as natural gas (methane),

propane, or butane, with air.

After the diffusion process is completed, the

component may be quenched from the carbur-

izing temperature or reheated to austenitize the

steel, and then quenched. Bainite formation in

the case is strongly inhibited by the presence of

molybdenum and chromium. Since the surface

contains higher carbon content than the core, it is

harder than the softer core. Core hardness is

most strongly affected by the presence of molyb-

denum and manganese. Chromium exhibits a

moderate effect, and nickel exhibits a weak

effect (Ref 3). Core hardness is strongly affected

by the quenchant selection and quenching tem-

perature.

In addition to strengthening the case, the

increased carbon content also provides desirable

increased compressive stresses that will inhibit

fatigue crack initiation. The lower carbon con-

tent in the core also will produce improved

fatigue strength.

Carbonitriding is similar to carburizing in that

it is a diffusion process that involves the simul-

taneous diffusion of carbon and nitrogen (from

ammonia) into the steel surface. To obtain

maximum strength, the carbonitriding process

produces a surface that is enriched in nitrogen

and carbon in the form of an epsilon

(e)-carbonitride layer and a diffusion zone

containing chromium-iron carbide, (Cr,Fe)

;

chromium carbide nitride, Cr

5N

0.3

; chro-

mium nitride, (Cr

N) or [Cr, Fe(2N

i ... x

)]; and

N phases (Ref 4, 5). Typical case thicknesses

range from 50 to 200 mm with a hardness

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_177-240.pdf/Chap_06/ 18/8/2008 3:16PM Plate # 0 pg 177

Failure Analysis of Heat Treated Steel Components

L.C.F. Canale, R.A. Mesquita, and G.E. Totten, editors, p 177-240

DOI: 10.1361/faht2008p177

www.asminternational.org

between 750 and 900 HV. Like carburizing, the

case depth of carbonitrided steel is dependent

on both the carbonitriding diffusion time and

temperature, as illustrated in Fig. 1 (Ref 2, 6).

Deeper case hardnesses may be obtained by ﬁrst

precarburizing prior to carbonitriding. Karamis¸

showed that carbonitrided AISI 5115 steel

exhibited greater surface hardness and wear re-

sistance than carburized AISI 5115 steel (Ref 4).

Carbonitriding processes are typically con-

ducted in either a gas (ammonia) or a salt bath

based on trade names such as Tufftride, Nitrotec,

and Nitrox. Alternatively, a plasma nitriding

process may be conducted. A brief summary

comparison of carburizing and carbonitriding

processes is provided in Table 1 (Ref 2).

Carter has reported that failures of carburized

gears are primarily due to service-related causes,

such as misalignment, poor lubrication, and

overloading, which constitute the greatest

source of all gear failures, as shown in Table 2

(Ref 7). Heat treatment was the second most

often cited cause for failure. However, it is often

difﬁcult to detect the root cause of a speciﬁc

failure under the conditions in which the failure

occurred, and many of the service-related fail-

ures could have been reduced with more atten-

tion to the other potential causes of failure

shown, since they are often interrelated.

Palaniradja et al. reported that 10 to 12% of

carburized parts are rejected due to various

process-related defects (Ref 8). To examine this

in more detail, they conducted a Taguchi ana-

lysis of gas carburization of AISI 8620 and 3310

steels, and their results showed that relative

contribution to surface hardness was holding

time (20%), carbon potential (20%), carburizing

temperature (0%), and quenching time (60%).

Similarly, they also studied the effects of process

variables on case depth and found: holding

time (60%), carbon potential (9%), carburizing

temperature (14%), and quenching time (10%)

(Ref 8). These results show that an adequate

understanding of failure analysis of carburizing

and, by implication, carbonitriding must be

accompanied by understanding the contribution

of process parameters on resulting potential

failures.

Some of the most common contributors to

failure of carburized gears include surface ﬁnish,

microstructure, excessive or inadequate case

depth, incorrect case and/or core hardness,

improper carbon concentration and hardness

gradients, undesirable surface carbon content,

excessive retained austenite, large amounts of

globular and network carbides, intergranular

oxidation, internal oxidation, residual stress,

extremely coarse case or core grain structure,

untransformed core with free ferrite, quenching

and grinding cracks, surface decarburization,

excessive heating during grinding, excessive

removal of the case during grinding, micro-

cracking, and so on (Ref 9, 10).

This chapter provides an overview of various

contributors to failure of carburized and carbo-

nitrided components, with the primary focus on

carburized components.

Fig. 1

Correlation of case depth of carbonitrided steels with

varying diffusion times and temperatures

Table 1 Comparison of carburizing and

carbonitriding processes

Process Comments

Carburizing Hard, highly wear-resistant surface (medium case

depths), excellent contact load potential, good

bending fatigue strength, good seizure

resistance, excellent quench cracking resistance,

low-to-medium-cost steels required, high capital

investment

Carbonitriding Hard, highly wear-resistant surface (shallow case

depths), fair contact load potential, good bending

fatigue strength, good seizure resistance, good

dimensional control, excellent quench cracking

resistance, low-cast steels usually satisfactory,

medium capital investment

Table 2 Survey summary of sources of

gear failures

Cause of gear failure %

Material quality and forming 0.8

Design 6.9

Service-related causes 74.7

Manufacturing 1.4

Heat treatment 16.2

178 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_177-240.pdf/Chap_06/ 18/8/2008 3:17PM Plate # 0 pg 178

Design

Component design may contribute directly or

indirectly to component failure. Deﬁciencies

such as insufﬁcient radii or sudden changes in

section size are signiﬁcant contributors to fail-

ure. In addition, the presence of stress raisers,

such as those shown in Fig. 2, are among the

most common design contributors to quench

cracking and fatigue failure.

A more comprehensive insight into design

is provided by Kuehmann et al., who developed

a systems analysis ﬂow chart to describe the

effects of case-core hardening in designing a

carburizing process/metallurgical structure/

resulting properties and performance for the

production of gears produced by three routes:

conventional forging, near-net shape casting,

and powder metal processing (Fig. 3) (Ref 11).

To properly design a component, it is neces-

sary to estimate surface loading, distortion after

heat treatment, case depth and carbon proﬁle,

case and core hardness, and core strength. As an

estimate, for hardnesses within the range of 30 to

45 RC, the required case depth can be calculated

from (Ref 2):

Case depth to 50 HRC=(1:2 · 10

7

W)=F

where W is the force in pounds pressing the

surfaces together, and F is the length of the line

contact (inches).

Carter has recommended the following gen-

eral design criteria (Ref 7):

 If a component is carburized from both sides,

the case depth should not be greater than

20% of the wall thickness.

 At the base of gear teeth, 30% of the core

material should remain uncarburized.

 Shallow case depths usually require higher

case hardness.

 Case depths should be ﬁve times the accep-

table wear limit.

Fig. 2

Effected of stress raisers on stress concentration and distribution of stress at several changes of form in components. (a) to (c)

Progressive increases in stress with decreasing ﬁllet radii. (d) to (f) Relative magnitude and distribution of stress resulting from

uniform loading. (g) Stress caused by the presence of an integral collar of considerable width. (h) Decrease in stress concentration that

accompanies a decrease in collar width. (i) Stress ﬂow at the junction of a bolt head and a shank. (j) Effect of a single sharp notch. (k) Effect

of a continuous thread. (l) Effect of a groove or gauge. Source: ASM Handbook, Volume 11, 2002, p 715

Sources of Failures in Carburized and Carbonitrided Components / 179

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_177-240.pdf/Chap_06/ 18/8/2008 3:17PM Plate # 0 pg 179

 Although machining is proportional to the

case depth, it should be minimal.

Kern and Suess have recommended the fol-

lowing general guidelines for heat treatment of

gas-carburized gears (Ref 2):

 For forgings, normalize or anneal (as re-

quired by the alloy being heat treated) from a

temperature at least 28



C (50



F) above the

carburizing temperature.

 Assure that the gears are machined prior to

heat treatment.

 Bring the gear to the carburizing temperature

with sufﬁcient circulation of a neutral

atmosphere, and then introduce the gas used

for carburizing.

 For deep cases (41.5 mm, or 0.060 in.),

adjust the carburizing atmosphere and time

to produce uniform carbon diffusion from

the surface to the core. A decrease of 0.15

to 0.20%/0.25 mm (0.010 in.) of depth is

nearly ideal. The gradient may be steeper for

shallower case depths.

 The recommended surface carbon is 0.90 to

1.10% for 4300, 4600, 8600, 8800, and 9400

carburizing steels. Although the same case

depth is generally acceptable for grades such

as 4800, they are preferably reheated for

hardening. The recommended surface car-

bon is 0.65 to 0.85% for high-nickel steels

such as the 4800 series, which is usually

direct quenched.

 To minimize cost and distortion, use direct

quenching whenever possible.

 To assure optimal dimensional control,

properly maintain quenching dies and

plugs.

 Quench as rapidly and uniformly as prac-

tical, and use spray impingement ﬁxtures on

large, solid pinions that are four pitch and

coarser.

 Use hot oil quenching on ﬁne pitch gears.

Fig. 3

Kuehmann et al. ﬂow chart to summarize design elements of a carburizing process/metallurgical structure/resulting properties

and performance comparison of gas carburizing gears produced by conventional forging, near-net shape casting, and powder

metal processing

180 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_177-240.pdf/Chap_06/ 18/8/2008 3:17PM Plate # 0 pg 180

 Case hardness of the ﬁnished gear should be

60 HRC or greater.

 If possible, test each gear for partial dec-

arburization and/or upper transformation

products.

 To minimize distortion and to permit quieter

operation, the surface carbon content should

be uniform throughout the production cycle.

Steel Selection and Hardenability

Steels typically used for case hardening con-

tain carbon contents of less than approximately

0.25%. The carbon content of the case is usually

controlled to between 0.8 and 1% C. The actual

surface carbon content is generally limited to

0.9%, because excessively high carbon content

may lead to the presence of unacceptably high

retained austenite and brittle martensite. Some

of the most commonly used AISI grades of steel

used for carburizing are shown in Table 3

(Ref 12).

Plain carbon steels may be carburized; how-

ever, relatively poor hardenability due to the

lack of alloying elements reduces the carburiz-

ing response of the case. Because of the stabi-

lizing effect of the nitrogen relative to austenite,

carbonitriding provides greater hardenability

than attainable with carburizing. Therefore,

plain carbon steels respond well to carboni-

triding.

Proper steel selection is a critically important

process to provide the desired case depth and

microstructure and the required core properties.

Typically, the case structure should be fully

martensitic, with the exception of allowing for

required application design limits on retained

austenite content. For example, the steel must

possess sufﬁcient hardenability to provide the

desired hardness and microstructure in both the

case and the core. After carburizing, the com-

ponent must possess sufﬁcient toughness with-

out exhibiting brittle failure.

Most steels that are carburized are deoxidized

by the addition of aluminum (commonly desig-

nated as killed steels). Deoxidation will provide

ﬁner grain sizes to temperatures of approxi-

mately 1040



C. Coarser grained steels may be

carburized if grain reﬁnement by double quench-

ing is possible. Double quenching typically in-

volves direct quenching followed by reheating

to a lower-temperature quenching a second time

(Ref 13).

Selection of proper hardenability of steels for

both carburizing and carbonitriding is critically

important, both of the core and the case, since

improper hardenability design can lead to un-

desirable nonmartensitic transformation pro-

ducts in the case, leading to a potential reduction

in static and dynamic fatigue strength of up to

30% and a reduction of impact fatigue of up to a

factor of 2.5 times (Ref 14). The hardenability

gradient of the case and the core is dependent

on a number of factors, including cooling rate

during quenching, variability of the chemical

composition (alloy content, carbon and nitro-

gen) of the case, and the carburizing or carbo-

nitriding method being used.

Core hardenability is being used increasingly

to specify alloy steels used for case hardening

where the hardenability of both case and core

must be considered. Details for the traditional

approach for the experimental determination of

hardenability of carburizing steels are provided

in Ref 15. Jominy curves for a number of car-

burizing steel alloys with varying hardenability

are shown in Fig. 4 (Ref 16).

Procedures have also been described for

determining ideal diameter (D

) values and

hardenability of carburizing steels from Jominy

data using regression equations for composition

and grain size (Ref 17, 18). The ideal diameter is

deﬁned as the diameter of a cylindrical steel bar

that will form 50% martensite at the center when

subjected to an “ideal” quench. Hardenability

differences may be substantially greater for

some case-hardening steel grades relative to

others due to the difference in carbon content in

the case and core. This is more critical for heavy-

sectioned components that are reheated and

quenched.

The hardness gradient through the case is due

to the relationship between the thermal gradient

Table 3 Common carburizing grades of steel

and their relative processing features

AISI

steel

grade Note

4620 Lower-cost, chrome/nickel/molybdenum steel where only

nominal hardenability and core response is required

8620 Most commonly speciﬁed grade. Excellent carburizing

response, with good hardenability for most section sizes

4320 Higher hardenability for improved core response in heavier

sections

4820 Increased nickel content for improved core toughness; slower

response results in longer process times

9310 Maximum nickel content for maximum core toughness;

slower response results in longer process times

Sources of Failures in Carburized and Carbonitrided Components / 181

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_177-240.pdf/Chap_06/ 18/8/2008 3:17PM Plate # 0 pg 181

and the carbon gradient during quenching.

Therefore, an increase in case hardenability

required to produce greater amounts of marten-

site for a given carbon content will result in an

increased case depth. In such cases, a reduced

(shallower) carbon proﬁle and shorter carburiz-

ing times will be necessary to obtain the desired

hardness proﬁle in the carburized component.

SAE J1975 standard “Case Hardenability of

Carburized Steels” summarizes characteristics

of carburized steels and factors involved

in controlling hardness, microstructure, and

residual stress. Methods of determining case

hardenability are also provided.

Parrish reported the following scheme that

was developed to classify the case hardenability

of steels (Ref 19):

 Level 1: Surface carbon contents 40.8% C

are martensitic.

 Level 2: All carbon contents from the surface

to 50% C are martensitic.

 Level 3: All carbon contents from the surface

to 0.27% C are martensitic.

 Level 4: A martensitic case occurs at all

carbon levels, including the core material

just beneath the case.

Figure 5 illustrates the core hardenabilities for

a number of carburizing steels (Ref 19). This

ﬁgure is used by estimating the equivalent dia-

meter for the critically stressed section of the

component of interest, and then the expected

level of case hardenability of that steel is deter-

mined. Figure 5 indicates that level 4 is attain-

able only for small section sizes of more alloyed

steels, and level 3, depending on the section size,

is more readily attainable for most of the steels

shown. Level 2 is more typical of the more

common case-hardened parts and should repre-

sent a minimum target to be attained.

Case hardenability may vary widely even for

steels with equivalent core hardenabilities. Kern

Fig. 5 Case hardenabilities of a number of carburizing steels with oil quenching

Fig. 4

Jominy hardenability data for a number of carburizing

steels

182 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_177-240.pdf/Chap_06/ 18/8/2008 3:17PM Plate # 0 pg 182

and Suess provided the following guidelines

(Ref 2):

 Steel grades in which the case hardenability

is due to carbide-forming metals such as

chromium (8600 series) are sensitive to

microcracking, especially when direct quen-

ched from the carburizing temperature. This

can be controlled by restricting the carbon

content in the case to 0.9%.

 Steel grades with relatively high nickel

content, for example, 4800 and 9300 series,

may form excessive (430%) amounts of

retained austenite when direct quenched

unless the carbon content of the case is

maintained at 50.75%.

 Carburizing round section sizes greater than

76 mm (3.0 in.) may lead to difﬁculty in

achieving the desired case and core micro-

structures when quenching in oil. In such

situations, consider induction hardening or

nitriding or using a highly alloyed steel

grade such as AISI 9310.

 Some standard grades of steel exhibit nar-

rower core hardenability bands than other

grades. For example, 8620H exhibits a

hardenability band spread of 14 HRC at J 4,

and 9310 exhibits only 8 HRC spread at the

same J-value. This provides a greater

amount of distortion control in addition to

some possible application-dependent prop-

erty advantages as well.

One problem that can arise during the steel-

making process or that may be observed as a mill-

to-mill variant is the presence of segregation

effects through the section of the steel billet

during a continuous casting process, which re-

sults in the presence of a white band (Ref 20).

White band is a type of negative segregation

often observed in electromagnetically stirred

continuous castings. The white banding pro-

duces a signiﬁcant hardness gradient across the

billet. After subsequent rolling and forging or

machining to produce a component, the resulting

grainﬂow can produce nonuniform hardenability

and/or soft spots that can signiﬁcantly affect

distortion.

In addition to proper hardenability selection,

to achieve maximum core toughness, proper

austenitization and quenching to martensite is

necessary. These topics are discussed subse-

quently.

Case Depth. The case of a carburized (or

carbonitrided) steel alloy is that portion

extending inward from the surface, where the

hardness is greater than that of the core. The total

case depth is the distance or thickness of the

carbon-enriched surface layer. The effective

case is the point where 0.4 to 0.5% C (percent is

called points in the industry) is present if the part

is hardened to 50 HRC (510 HIV). The depth of

the case is a function of carburizing time and

carbon (carbon potential) at the surface. Genel

and Demirkol have reported that the following

equation model can be used to predict effective

case depth (Ref 21):

Effective case depth (mm)=

0:41 ½Carburizing time (h)

1=2

The carbon potential of a furnace atmosphere

at a speciﬁed temperature is deﬁned as the carbon

content of pure iron that is in thermodynamic

equilibrium with the atmosphere. The carbon

potential of the furnace atmosphere must be

greater than the carbon potential of the surface of

the workpieces for carburizing to occur. The

carbon potential is a measure of the ability of

a gas to react with the steel surface. It is this

difference (carbon content in the gas and at

the steel surface) that provides the driving force

for carbon transfer to the parts being carburized.

The composition of a gas that will produce a

given surface carbon content is dependent on

equilibrium data for the gas. The amount of

carbon transferred will depend on factors that

include temperature, time, and steel composi-

tion.

The alloy composition of the steel will affect

the effective carbon potential at the surface. The

presence of elements such as chromium, man-

ganese, and molybdenum that form stable car-

bides of iron will increase the effective carbon

potential. Elements such as silicon and nickel

form less stable carbides and reduce the effec-

tive carbon potential. Alloying elements that

stabilize austenite or ferrite also reduce the

effective carbon potential. The effect of alloying

elements on carbon potential can be calculated

from (Ref 22):

Log (correction factors) =0: 005 (%Si)

 0:013 (%Mn)  0:040 (%Cr)+0:014 (%Ni)

 0:013 (%Mo)

It is important to note that, except for long

carburizing times (3 to 410 h), the surface

carbon content is typically not equal to the car-

bon potential, because the surface of the part

Sources of Failures in Carburized and Carbonitrided Components / 183

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_177-240.pdf/Chap_06/ 18/8/2008 3:17PM Plate # 0 pg 183