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

Подождите немного. Документ загружается.

cylinder head of a sport utility vehicle. It was

observed that the valve seats were expanding

during shrink-ﬁtting instead of contracting.

This resulted in interference between the

valve seat and the cylinder head, causing rejec-

tion of the cylinder heads. The parameters that

can cause dimensional instability during shrink-

ﬁtting are:

Operation 4M parameter Observations

Powder chemistry Not applicable ...

Blending Not applicable ...

Compaction Not applicable ...

Sintering Not applicable ...

Tempering High retained austenite Yes

A signiﬁcant amount of retained austenite

was observed even after tempering. This

retained austenite was transforming to marten-

site during shrink-ﬁtting, where the products

were cooled to 60



C, which resulted in a

volume expansion.

Corrective Measures. A complete trans-

formation of retained austenite can be effected

by:

 Multiple tempering

 Cooling to below the martensite ﬁnish tem-

perature (in this case, a deep cryogenic

treatment)

 Combination of tempering and deep cryo-

genic treatment

Trials revealed that a combination of subzero

treatment and tempering gave the best results

with respect to dimensional stability as well as

metallurgical properties.

Results. The valve seats subjected to a

combination of subzero treatment and tem-

pering did not expand during shrink-ﬁtting.

Further dimensional measurements taken

at 60



C were in line with the theoretically

calculated shrinkage in contrast to the earlier

expansion.

Case Study 5: Wear after Sintering

Bushes were produced with Fe-0.45%P–

2%Cu-2.5%C and supplied in the oil-impreg-

nated condition. These bushes were deliberately

undersintered so that some free graphite was

retained, which acted as a solid lubricant that

in turn minimized wear in application. The

parameters that can cause wear of a free-graphite

bush are:

Operation 4M parameter Observations

Powder chemistry Not applicable ...

Blending Not applicable ...

Compaction High density No

Sintering High sintering temperature Yes

High sintering time No

Cementite network Yes

Insufﬁcient free graphite Yes

Oil impregnation Low oil content No

Metallography of the worn sample revealed a

cementite network in the product (Fig. 14). This

is a clear indication that excess carbon had gone

into solution, which resulted in the solid lubri-

cant being depleted.

Corrective Measures. Sintering was carried

out at 1080



C for 20 min, which resulted

in excess combined carb on. The temperature

was reduced so that the combined carbon

was 50.8%. The sintering temperature for the

product was ﬁxed at between 1040 and 1060



Results. Bushes sintered at 1060



for 20 min resulted in a pearlitic-steaditic

microstructure with no cementite network

(Fig. 15).

Case Study 6: Fracture of Steam-Treated

Part

Valve plates for compressors were subjected

to grinding after steam treatment. Fracture of

a steam-treated valve plate was observed during

grinding after steam treatment. The parameters

Fig. 14

Pearlite-steadite microstructure with cementite net-

work

Failure Analysis of Powder Metal Steel Components / 407

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_395-415.pdf/Chap_12/ 18/8/2008 3:53PM Plate # 0 pg 407

that can cause the fractu re of a steam-treated

part are:

Operation 4M parameter Observations

Powder chemistry Not applicable ...

Blending Not applicable ...

Compaction Low density No

Sintering Low sintering temperature No

Low soaking time No

Oxidizing atmosphere No

Decarburizing atmosphere No

High dewaxing temperature No

Overloading of parts Yes

Steam treatment High temperature No

Insufﬁcient steam No

Low soaking time No

Overloading of parts No

Masking of parts No

Hardness measurements reveale d very low

hardness. Product with hardness conforming to

speciﬁcation and product with low hardness

were subjected to standard metallographic

analysis. The product with low hardness

revealed a predominantly ferritic structure with

free copper (Fig. 16), even though the blend

graphite addition was 0.8% (Fig. 17). Thus,

undersintering was suspected to be the possible

cause. However, the sintering control charts

revealed that the temperature as well as the

belt speed were meeting the speciﬁcations.

The loading pattern was then checked. It was

observed that the loading pattern followed

that which was recommended for a 46 cm

(18 in.) belt, whereas the belt width was 30 cm

(12 in.). This resulted in overloading of

parts, which consequently resulted in under-

sintering.

Corrective Measures. Suitable one-point

lessons were imparted to the sintering operators

to ensure that the right process plans were fol-

lowed.

Case Study 7: Oxidation after Sintering

Exhaust valve seats for internal combustion

engines are produced with a chromium-rich

hard-phase alloy. The hard-phase alloy is

essential for wear resistance at elevated tem-

peratures, to which the exhaust valve seat

is subjected. Oxidation of the hard phase was

observed after sintering. The parameters that can

cause oxidation after sintering are:

Operation 4M parameter Observations

Powder chemistry Chromium-ri ch alloy Yes

Blending Not applicable ...

Compaction Rusted compacts No

Sintering High rapid burnoff temperature Yes

High cooling zone dewpoint Yes

Fig. 16

Low-hardness sample with almost no combined

carbon. Only ferrite and free copper are observed.

Fig. 17

Typical micros tructure with ~0.7% combined

carbon

Fig. 15

Pearlite-steadite microstructure free from cementite

network

408 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_395-415.pdf/Chap_12/ 18/8/2008 3:53PM Plate # 0 pg 408

Chromium-rich particles are susceptible to

oxidation when the product is heated and cooled

in the range of 650 to 1050



C. Oxidation of the

hard-phase particles is favored by:

 High rapid burnoff temperature

 High cooling zone dewpoint

Figure 18 shows the microstructure with

oxidized hard-phase particles. The oxide ﬁlm

around the hard phase does not allow the hard

phase to interact with the matrix. As a result, the

matrix will not have chromium carbides, which

helps the valve seats to retain wear resistanc e at

high temperatures. The rapid burnoff temper-

ature was observed to be 650



C, and the cooling

zone dewpoint was between 20 to 30



These are potential causes for oxidation.

Corrective Measures. The furnace was

allowed to stabilize for a period of 4 to 16 h. The

gas ratio was maintained at 70% N

and 30% H

during this period, and the gas volume was

increased by ~25%. The rapid burnoff tem-

perature was restricted below 600



C. The valve

seats were sintered when the operating dewpoint

in the hot and cooling zones is 35



C or lower.

Results. The modiﬁcation in sintering para-

meters resulted in the production of oxidation-

free valve seats (Fig. 19).

Case Study 8: Cracks after Induction

Hardening

A PM transmission gear was subject ed to

induction hardening for an application that

demanded high wear resistance as well as high

contact and bending strength. The gear had an

Fe-Cu-C chemistry and was compacted to a

density of 7 g/cc. Subsequently, the gear was

sintered, and a sizing operation was then carried

out for dimensional correction. Induction hard-

ening of the gear was carried out after sizing, and

cracks were observed after induction hardening.

The parameters that can cause cracks after

induction hardening are:

Operation 4M parameter Observations

Powder chemistry High carbon Yes

Blending Not applicable ...

Compaction Low density No

Sintering Low sintering temperature No

Low sintering time No

Sizing Not applicable ...

Induction hardening High heating time No

Severe quench No

Figure 20 shows the crack. This crack origi-

nates in the middle of the root of the gear and

Fig. 19

No oxide layer observed in the hard-phase particles

after implementation of corrective measures

Fig. 20

Crack in the middle of the root of the primary driven

gear

Fig. 18

Exhaust valve seat with oxidized hard-phase

particles

Failure Analysis of Powder Metal Steel Components / 409

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_395-415.pdf/Chap_12/ 18/8/2008 3:53PM Plate # 0 pg 409

does not originate at any sharp corners. Hardness

proﬁles from the root indicated that the root has

been hardened to a very high hardness (Fig. 21),

compared to the earlier sample in which no crack

was observed. Microstructure near the root

revealed untempered martensite, which gradu-

ally became a mixture of martensite and bainite

and subsequently ferrite and pearlite. This is in

contrast to the earlier sample where the root was

not hardened and had a bainitic structure. Sec-

ondly, the combined carbon estimated from the

pearlite content in the sintered microstructure

was 0.7% compared to 0.6% in the product

that was successfully induction hardened. The

strain caused by the martensitic transformation

increases in proportion to the combined carbon

in the microstructure. Thus, the possible causes

for the crack were:

 Too high a soaking time, which resulted in a

severe quench

 High combined carbon in the sintered

microstructure

Corrective Measures. The formation of

martensite in the root was a result of quenching

from very high temperatures. The time was

reduced, and the coil design and quench were

modiﬁed. However, the martensitic micro-

structure coul d not be avoided in the root, and

cracking was still observed. It was concluded

that the product was unable to withstand the

stresses associated with the transformation. The

carbon content was reduced in the blend to 0.6%,

which resulted in a lesser volumetric strain due

to martensitic transformation in the product.

Induction hardening with this chemistry yielded

no cracks, and the hardness traverse matched

that obtained in the rapid prototype samples.

Results. Products with a lower combined

carbon content did not crack when induction

hardened.

Case Study 9: Cracks after Quenching

A ﬂyweight for a governor assembly in a

diesel engine was carbonitrided to meet the

product requirements for strength. This is a

complex part wherein multiple punches are used

to form the sections of different thicknesses.

The product was observed to crac k after carbo-

nitriding. The parameters that can cause cracks

after quenching are:

Operation 4M parameter Observations

Powder chemistry Not applicable ...

Blending Not applicable ...

Compaction High density difference No

Underﬁlling of powder Yes

Sintering Low sintering temperature No

Low sintering time No

Carbonitiriding High hardening temperature No

Severe quench No

The parts were subjected to magnetic particle

inspection prior to heat treatment. No cracks

were observed. Metallography was then carried

Fig. 21 Hardness traverse from root

410 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_395-415.pdf/Chap_12/ 18/8/2008 3:53PM Plate # 0 pg 410

out at the location where the crack occurred in

heat treatment. This is shown in Fig. 22. Insuf-

ﬁcient powder ﬁlling was observed (Fig. 23).

The underﬁll results in a localized region of low

strength, which gives way during subsequent

quenching and results in crack formation.

Corrective Measures. The localized under-

ﬁlling was due to the presence of a sharp corner

in the tool. A trial was conducted wherein the

sharp corner was smoothed by ﬁling the parts.

No crack was observed after carbonitriding. It

was resolved to increase the radius in the tool so

that a sharp corner and consequently an underﬁll

are avoided. Products thus produced were heat

treated, and no cracks were observed.

Results Products produced with an improved

radius in the tool resulted in elimination of

product rejections due to quench cracks.

Case Study 10: Dimensional Change

in Carbonitriding

An angle lever for a diesel inject ion pump was

produced with FD-02-00 chemistry and carbo-

nitrided to a case depth of 0.2 to 1.2 mm. When

the carbonitriding was established at a new heat

treatment source, it was observed that the bore

did not shrink sufﬁciently during carbonitriding.

The parameters that could cause a dimensional

change in carbonitriding are:

Operation 4M parameter Observations

Powder chemistry Not applicable ...

Blending Not applicable ...

Compaction Bore undersized due to tool wear No

Sintering High sintering temperature No

High sintering time No

Carbonitiriding Low soaking time Yes

Insufﬁcient carbon and nitrogen

enrichment

High quench oil temperature No

Insufﬁcient transformation Yes

Carbonitriding was carried out at 840



C for

45 min with a carbon potential of 0.8% in the

atmosphere. In the case of the vendor, better

furnace gas sealing ensured that this case depth

was achieved in 20 min. The hardness traverse

indicated that the longer soaking time had a

higher carbon martensite in the case as well as

a higher core hardness, which in turn resulted in

a case depth at the lower end of the speciﬁcation

for a lower soaking time. This is shown in

Fig. 24.

Corrective Measures. Higher shrinkage is

ensured by a higher amount of martensitic

transformation. This was ensured by increasing

the soaking time from 20 min to 45 min. The

increased soaking time resulted in better expo-

sure of the product to carbon and nitrogen

enrichment, which in turn resulted in higher

martensite content and the desired level of

dimensional shrinkage.

Results. The increased martensite content

resulted in increased volume expansion. Thus,

the shrinkage in the bore is higher with a

soaking time of 45 min. Products soaked at

840



C for 45 min were found to shrink sufﬁ-

ciently, and dimensions conformed to speciﬁ-

cations.

Case Study 11: Low Surface Hardness

after Carbonitriding

Low surface hardness was observed in

an iron-copper part after carbonitriding. The

Fig. 23

Metallography of the region where the crack occur-

red in the sintered sample, showing a pre-existing

defect caused by poor powder ﬁlling

Fig. 22 Quench crack in ﬂyweight

Failure Analysis of Powder Metal Steel Components / 411

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_395-415.pdf/Chap_12/ 18/8/2008 3:53PM Plate # 0 pg 411

parameters that can cause low surface hardness

after carbonitriding are:

Operation 4M parameter Observations

Powder chemistry Not applicable ...

Blending Not applicable ...

Compaction Low density No

Sintering Low sintering temperature No

Low sintering time No

Carbonitiriding Low soaking time No

Insufﬁcient carbon and

nitrogen enrichment

Excess carbon and nitrogen

enrichment

Yes

Delayed quench No

Pearlite/bainite in case No

High retained austenite Yes

After ensuring that the density conformed

to speciﬁcation, the low-hardness products

were subjected to metallographic analysis. The

case microstructure revealed a high amount

of retained austenite, which is the cause of low

hardness (Fig. 25). High retained austenite

can result if the carbon potential as well as

the nitrogen potential are high. The oxyprobe,

which senses the carbon potential, was exam-

ined and found to be covered with soot. This

resulted in improper sensing of carbon potential,

leading to an excess of active carbon in the

furnace atmosphere, which in turn resulted in

high retained austenite.

Corrective Measures. The temporary cor-

rective measure adopted to salvage the parts

was to carr y out a deep freeze, because repro-

cessing may result in scrapping of the product

due to dimensional distortion. On a permanent

basis, a periodic preventive maintainence

schedule for the oxyprobe was drawn up and

adhered to.

Results. Correcting the oxyprobe yielded

products with a martensitic/510% retained

austenite structure that met the requirements for

surface hardness.

Case Study 12: Variation in Bore Diameter

after Heat Treatment

A valve retainer for a shock absorber was

produced from an FC-02-08 blend that was

hardened for wear resistance. High ovality was

observed after heat treatment. This resulted in

~10% of the products being out of speciﬁcation.

Fig. 25 High retained austenite in the case

Fig. 24 Hardness traverse in bell crank lever

412 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_395-415.pdf/Chap_12/ 18/8/2008 3:53PM Plate # 0 pg 412

The parameters that can cause a wide variation

in bore diameter after heat treatment are:

Operation 4M parameter Observations

Powder chemistry Excess copper content No

Blending Elemental additions No

Compaction Low density No

Sintering Low sintering temperature No

Low sintering time No

Carbonitiriding Volume expansion due to

martensitic transformation

Yes

Thermal contraction during

quenching

Yes

The ovality was checked after sintering and

after heat treatment. The data revealed that

hardening contributes signiﬁcantly to distortion,

and the contribution of sintering is negligible.

Thermal contraction due to quenching and the

accompanying martensitic phase transformation

contributed signiﬁcantly to the distortion.

Corrective Measures. To minimize the

distortion, it was reso lved to eliminate the

hardening operation and achieve the required

hardness in sintering. Prealloyed powder con-

taining nickel and molybdenum, blended spe-

cially to produce products with negligible

distortion after sintering, was adopted to achieve

both the hardness and dimensions after sintering.

Results. Changing the process from con-

ventional hardening to sinter hardening reduced

the dimensional spread to 50% of the variation

observed in products with a separate heat treat-

ment operation.

Case Study 13: Low Breaking Load after

Carbonitriding

A component for a gearbox application pro-

duced with a chemistry of FD-02-00 and having

an overall density of 7.2 g/cc was subjected

to carbonitriding to meet the product require-

ments for wear and mechanical strength. During

assembly, the product failed by fracture. The

parameters that can cause low breaking load

after carbonitriding are:

Operation 4M parameter Observations

Powder chemistry Not applicable ...

Blending Not applicable ...

Compaction Low density No

Low sectional density Yes

Sintering Low sintering temperature No

Low sintering time No

Carbonitiriding High carbonitriding temperature Yes

High time of carbon potential

attainment

Yes

Variation in batch quantity Yes

Martensite in core Yes

No case depth inspection

in thin section

Yes

Low breaking load indicated poor toughness

of the product. Poor toughness is related to both

microstructure and density. Lower densities

and a through-hardened microstructure with

martensite in the core can cause premature

fracture. Metallography was carried out in dif-

ferent sections of the failed product. Metallo-

graphy of the sections indicated that the thinner

sections in the product were through hardened.

Figure 26 shows the through-hardened micro-

structure. Sectional density measurements indi-

cated that the thicker sections had a density of

47.2 g/cc, and the thinner sections had a density

of 7.1 g/cc.

However, this through hardening of thinner

sections was not observed in the initial samples

that were tested. To determine the cause of this

variation across different heat treated batches,

the carbon potential attainment was studied for

various batches.

During establishment of the carbonitriding

process, the process parameters were ﬁxed for

~500 parts. The same was subsequently ﬁxed

for the full batch quantity of 3500. In this par-

ticular case, the carbon potential pick up in the

furnace was completely different for the trial

and bulk. Figure 27 shows the difference in

carbon potential pickup. There is a signiﬁca nt

difference in the time for attainment of a carbon

potential of 0.8%. Longer times between 0.6 and

0.8% in the bulk lots result in additional carbon

pickup, which resulted in through hardening of

the thinner sectio ns.

Corrective Measures. The carbonitriding

temperature and time were reduced. The batch

quantity was ﬁxed. The density in the compo-

nent was increased to 7.2 g/cc min in all

Fig. 26 Martensite in the core of thin sections in the product

Failure Analysis of Powder Metal Steel Components / 413

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_395-415.pdf/Chap_12/ 18/8/2008 3:53PM Plate # 0 pg 413

sections, which ensured closure of intercon-

nected pores. Inspection standards for sectional

density and case depth measurements in the

thinner section were introduced so that any

nonconformance would be detected and not

reach the customer.

Results. The corrective measures resulted in

a pearlitic-bainitic microstructure in the core of

the thin section, and the toughness and, conse-

quently, the breaking load were achieved con-

sistently.

Case Study 14: Dimensional Change in

Steam Treatment

Pistons having a density of 7 g/cc were steam

treated for wear resistance and corrosion resist-

ance. During steam treatment, it was important

that the bore dimensions be maintained within

speciﬁcation. It was observed that after steam

treatment, the bore was undersized. The para-

meters that can cause dimensional change in

steam treatment are:

Operation 4M parameter Observations

Powder chemistry Not applicable ...

Blending Not applicable ...

Compaction Not applicable ...

Sintering Not applicable ...

Machining Bore undersized No

Steam treatment High temperature Yes

Thick oxide layer Yes

After ensuring that the machined dimensions

of the bore were in accordance with speciﬁcation

prior to steam treatment, metallography was

carried out on the samples with the undersized

bore. An oxide layer of up to 15 mm was

observed (Fig. 28). Typically, Fe

layers are

3to5mm, because the lattice of Fe

does not

allow further diffusion of oxygen molecules

once this layer is formed. Such a thick layer is

due to the formation of Fe O. The steam treat-

ment set temperature was 560



C for this pro-

duct. Because the Fe

reaction is exothermic,

the temperature of the products exceeded

570



C, resulting in the formation of FeO.

Corrective Measures. The steam treat-

ment temperature was subsequently reduced to

530



C, where an oxide layer of 4 to 6 mm was

obtained (Fig . 29). This resulted in the bore

dimensions meeting the speciﬁcation.

Fig. 27 Difference in carbon potential attainment for change in batch quantity

Fig. 28

Thick oxide layer indicative of FeO formation that

was obtained at a steam treatment temperature of

560



414 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_395-415.pdf/Chap_12/ 18/8/2008 3:53PM Plate # 0 pg 414

Case Study 15: Low Core Hardness after

Steam Treatment

Crankshaft sprockets with an Fe-Cu-C

chemistry and a density of 6.6 to 6.8 g/cc were

supplied in the steam-treated condition. The

customer demanded that the hardness in the

surface and core meet the speciﬁcation of

140 HV

min. The parameters that can cause

low core hardness after steam treatment are:

Operation 4M parameter Observations

Powder chemistry Low carbon Yes

Blending Not applicable ...

Compaction High density No

Sintering Low sintering temperature No

Low sintering time No

High ferrite content Yes

Sizing Closed surface pores No

Oil in pores Yes

Steam treatment Insufﬁcient time Yes

Low temperature Yes

High temperature (4570



C) No

Masking of products Yes

Nonuniform steam circulation No

It was observed that within a batch, some

products had a core hardness greater than 140

. Low and high core hardness samples were

subjected to metallography. It was observed that

higher-hardness samples had better steam

penetration compared to low-hardness samples.

Corrective Measures. To ensure higher

core hardness, the products required greater

exposure to steam. This was achieved by chan-

ging the steam treatment temperature and time

from 530



C/30 min to 550



C/2 h. Masking of

products can also potentially prevent uniform

exposure to steam. To eliminate masking, the

loading pattern was changed from a double layer

to a single layer. However, even after increasing

the steam exposure time and improving the

loading pattern, variation of hardness was

observed within the same batch. A trial was

conducted wherein oil-free sintered samples

were steam treated with sized and oiled parts. It

was observed that the sintered samples have

uniform steam penetration, and the thickness of

the steam oxide layer is uniform throughout the

part. This is in contrast to the sized products with

low core hardness, wherein the oxide layer is

minimal. The presence of oil or a related organic

residue was suspected of interfering with the

steam penetration. An oil that has a higher

volatile content and a lower organic residue

content was used for rust prevention, and the

same was used as a lubricant in sizing. The

change of oil resulted in a core hardness con-

forming to speciﬁcation. However, the hardness

obtained ranged from 145 to 190 HV. The pre-

sence of products with hardness marginally

above 140 HV

indicated that the hardness

must exceed 160 HV to avoid the chances

of core hardness going below speciﬁcation.

This additional improvement in hardness

was achieved by having a completely pearlitic

microstructure instead of a ferrite+pearlite

microstructure, which was effected by a blend

addition of 0.8% C.

Results. Implementation of the corrective

measures resulted in the core hardness being

achieved consisently in accordance with speci-

ﬁcation, as follows:

Properties Speciﬁcation Before After

Surface

hardness

140

10 min

152–208 185–235

Core hardness 140

10 min

110–173 170–207

Sintered

microstructure

... Ferrite+

pearlite

100% pearlite

Steam-treated

microstructure

... Incomplete

steam

penetration

with low-oxide

layer52 mm

in core

Complete steam

penetration with

oxide layer of

4–6 mm in core

REFERENCES

1. Asm Handbook, Volume 7, Powder Metal

Technologies and Applications, 1998

2. E. Klar and P. Samel, Powder Metallurgy

Stainless Steels: Processing, Micro-

Structures, and Properties, ASM Interna-

tional, 2007

Fig. 29

Oxide layer of 6 mm, indicative of Fe

obtained on

reduction of the steam treatment temperature to

530



Failure Analysis of Powder Metal Steel Components / 415

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_395-415.pdf/Chap_12/ 18/8/2008 3:53PM Plate # 0 pg 415

Induction Hardening

Janez Grum, University of Ljubljana

INDUCTION HEATING, in most applica-

tions, is used to selectively heat only a portion of

the workpiece that requires treatment. This

usually means that the process can be accom-

plished in a relatively short time and with high

efﬁciency, because energy is applied to the

workpiece only where it is needed.

Induction surface hardening is applicable to

axisymmetric or near-axisymmetric machine

parts in steel or cast iron that are produced in

substantial volumes. There are two basic tech-

niques for induction hardening machine parts:

single-shot and scanning. The former employs

selective heating and quenching to harden a

speciﬁc area or areas of the machine part in one

operation. The latter is usually applied to harden

progressively long, continuous sections, such as

shafts and spindles. In this instance, the scanning

inductor traverses the length of the section,

heating only a relatively small area at any given

time, and is followed closely by the quench

arrangement, which is often an integral part of

the inductor.

These advantages make it possible for in-

duction hardening to be fully automated and are

especially suitable for a large series of work-

pieces.

The induction-hardening procedure enables

an engineer, by simply adapting the shape of the

induction coil, to ensure the desired shape of the

hardened proﬁle of the surface layer. Likewise,

the engineer can surface harden only that part of

the surface (local hardening) on which a certain

increased level of hardness and wear resistance

are desired. One of the main advantages of in-

duction hardening is the ability to harden a sur-

face layer only in certain places at a deﬁned

penetration depth and shape.

For dynamically loaded machine parts, it is

very important to ensure the total compressive

stresses in the thin, most-loaded surface layer.

The total stresses are a sum of residual stresses in

a machine part and of load stresses produced by

the action of external forces and moments. To

ensure a long life of the machine part, knowl-

edge of the residual stresses in the machine part

and how to adjust the size and distribution of the

residual stresses by means of the selection of an

appropriate production technology are very

important.

In surface hardening, compressive residual

stresses always occur in the thin surface layer

due to martensite transformation. The size and

variation of the residual stresses depend pri-

marily on carbon content and less on the type

and content of alloying elements in heat treat-

ment and surface-hardening steels. The variation

of residual stresses in the surface layer can be

modiﬁed by varying the induction-heating con-

ditions and by a quenching method.

Induction surface hardening creates very

desirable residual stresses in the hardened sur-

face layer. Residual stresses are always of a com-

pressive nature and are usually present to the

depth of the induction-hardened layer. Residual

internal stresses, that is, the so-called residual

stresses, are the stresses present in a material

or a workpiece when there is no external force

and/or external moment acting on it. The resi-

dual stresses in metallic machine parts have

attracted the attention of technicians and engi-

neers only after manufacturing processes im-

proved to the level at which the accuracy of the

manufacture exceeded the size of deformation,

that is, distortion, of a workpiece/product.

Thus, it was almost 150 years ago that the

effect of internal stresses on plasticizing, that is,

destruction, distortion, and plastic deformation,

of workpieces was already known. It was then

that experts introduced measurement of indivi-

dual dimensions of products. For a given type of

machining process, they connected the inﬂuence

of the selected machining conditions with the

size of dimensional deviations. This was also the

beginning of an expert approach to the selection

of the most suitable machining and/or heat

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_417-501.pdf/Chap_13/ 18/8/2008 3:58PM Plate # 0 pg 417

Failure Analysis of Heat Treated Steel Components

L.C.F. Canale, R.A. Mesquita, and G.E. Totten, editors, p 417-501

DOI: 10.1361/faht2008p417