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

Подождите немного. Документ загружается.

been reached (d

= 0), and this is reﬂected on

only a partial depth of the overheated layer, d

Despite surface quenching, volume changes in

the surface layer are not affected, which means

that there is no distortion of the workpiece, so the

size of bending is f = 0. Bending distortion

may be noted quite rarely, and only if very in-

tensive cooling is applied with temperature dif-

ferences between the surface and subsurface

deﬁned by the depth at which the temperature

in the specimen is equal to the temperature of

the environment. However, such cases can be

neglected, since the goal is to achieve the aus-

tenitization and homogenization temperature,

which is much higher than the temperature of the

ferrite/pearlite-to-austenite transformation, T

In practice, the cooling rate is adapted to the

quenching medium, which ensures the forma-

tion of a hardened microstructure with minimum

internal stresses.

Figure 67(c) illustrates conditions after

induction surface hardening achieves a very

small depth of the hardened layer, d

, and a

correspondingly small depth of the heat-affected

zone. The achieved depth of penetration (con-

duction) of heat after induction hardening is

equal to d

. The volume changes in

the specimen are inﬂuenced only by the hard-

ened layer, that is, the volume of the hardened

layer versus the entire volume of the specimen.

If the volume changes are small, then the

bending of the specimen is negligibly small. For

a given hardened layer depth, it is possible at

each point during the cooling process to ensure a

higher yield point of the material (R

p, 0:2

) than

the internal stress in the axial direction of the

specimen. When the depth of the hardened layer

is equal to or smaller than one-twentieth of the

specimen height, d

h, there is no bend-

ing deformation of the specimen, only a slight

increase in residual stresses in the axial direc-

tion, despite a minimum volume change of the

hardened surface layer.

Figure 67(d) illustrates the conditions after

induction surface hardening of a specimen with a

slightly greater depth of the hardened layer, d

than on the specimen in Fig. 67(b), d

. The

achieved depth of penetration (heat conduction)

after induction heating is also slightly greater

and is d

or d

. If the achieved hardened-

layer thickness is d

h, then hardening is

accompanied by deformation of the specimen,

described as bending deformation in the middle

of the specimen ( f40). Surface hardening

is followed ﬁrst by annealing to remove the

stresses and then by mechanical straightening

and ﬁnally grinding. The procedures of sub-

sequent treatment of machine components are

chosen mainly with respect to the size of the

bending deformation.

Figure 67(e) shows the conditions after

induction surface hardening of the specimen

on both sides. Several different procedures can

be chosen to ensure two-sided hardening. The

ﬁrst procedure involves hardening of the top

surface to a certain depth, d

, and then of

the bottom surface to the same depth. In the

second procedure, both top and bottom surface

are hardened at the same time. The third proce-

dure involves hardening of the top surface

from the left to the right side and the bottom

surface from the right to the left. The most sui-

table procedure is the second one, where the

top and bottom surfaces are hardened simulta-

neously to the same depth. The achieved depth

of heating is equal to d

, whereas the depth of

the hardened layer is equal to d

. When equal

heat treatment conditions are ensured on both

sides of the specimen, the depth of the hardened

layer on the top surface, d

, is equal to the

depth of the hardened surface on the bottom

surface, d

This induction surface-hardening procedure

ensures that the bending deformation in the

middle of the specimen is equal to zero ( f = 0).

Figure 68 shows two crankshafts that differ in

the position of the crankpin with respect to the

main journal. The ﬁrst example presents a

crankshaft where the distance, a, between the

axes of the crankpin and the main journal is

greater than the sum of the radii of both the

crankpin and the journal:

In the second example, the crankshaft is

slightly different, having the distance between

the two axes equal to the sum of the radii of the

crankpin and journal:

This must be considered in the analysis of the

deformation of the crankshaft after hardening

cylindrical parts that alternate in following one

another (crankpin and main journal).

For the conditions after induction surface

hardening of the crankshaft described in the ﬁrst

468 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_417-501.pdf/Chap_13/ 18/8/2008 4:02PM Plate # 0 pg 468

example, Fig. 68(a) shows induction surface

hardening of the crankpin, which causes the

deformation of the main journal (Ref 15). Then

follows induction surface hardening of both

main journals on the left and right side. Due to

tensile residual stresses in the web between both

main journals, the web sections come closer to

one another, and the length of this crankshaft

section shortens by 2Dl. Considering the number

of crankpins or the number of main journals, the

entire length of the crankshaft decreases by Dl.

This value depends on the number of the main

journals, N, and the size of contraction per one

embedment (Dl). The entire contraction of the

crankshaft is thus DL

= NDl.

In the second example (Fig. 68b), the crank-

shaft is designed differently, with the axial dis-

tance between the crankpin and the main journal

being equal to the sum of both radii (Ref 15). As

in the previous case, induction surface hardening

of the crankpin induces tensile stresses in the

web, causing distortion of the main journal. Both

adjacent main journals come closer to one

another, but the distortion is smaller than in the

ﬁrst case. After induction surface hardening of

both adjacent main journals, any temporary

distortion that may have occurred vanishes, and

the crankshaft has equal length, just as prior to

heat treatment. Research has conﬁrmed that the

stress state in the workpiece is more favorable

and distortions are smaller in progressive hard-

ening than in single-shot hardening. The only

problem is whether the size and shape of the

workpiece allow the application of progressive

hardening.

Fujio et al. (Ref 52), in their third report on

induction hardening, focused on studies of dis-

tortion of gear-wheel teeth and residual stresses

in gear wheels. The authors measured the outer

diameter and the root diameter of the gear wheel

across two opposite teeth and the root parts of

the teeth with a micrometer before and after

quenching. The same dimensions were also

calculated theoretically, taking into account the

volume changes due to phase transformations,

and serve as a basis for determination for

deviations.

Figure 69 shows a change of the outer dia-

meter of the gear-wheel tooth for gear wheels 2,

3, and 4 (Ref 20, 52). The gear wheel has 26

teeth, which means that the deviation was mea-

sured between the ﬁrst and twenty-sixth tooth,

the third and sixteenth, and so on, so that thirteen

measurements were made in total. The results of

the measurements are shown as points in the

diagram. The dotted line represents the theore-

tically calculated increase in diameter due to

heat treatment.

Figure 70 shows a change of tooth height for

gear wheels 2, 3, and 4, as in the previous

Fig. 68 Crankshaft distortion after surface induction hardening of individual journal locations. Source: Ref 15

Induction Hardening / 469

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_417-501.pdf/Chap_13/ 18/8/2008 4:02PM Plate # 0 pg 469

example (Ref 20, 52). The dotted lines in the

individual diagrams represent the theoretical

deviations of height of the individual teeth by

0.18 mm and the deviations found by measure-

ment, which are mainly greater than the theo-

retical values. The greatest deviation can be

found with gear wheel 4. It exceeds 0.2 mm.

Figure 71 shows changes of the root diameter.

Calculations show that after heat treatment,

the root diameter decreases by approximately

0.02 mm (Ref 20, 52). The measured values

of the same diameter, however, remained

unchanged for gear wheels 2 and 3 and de-

creased by 0.1 mm maximum for gear wheel 4.

A comparison of the outer tooth diameter and

the tooth root diameter, also taking into account

the tooth height, shows that distortion of indi-

vidual gear wheels is considerably more com-

plicated and cannot be described by the selected

measurement methods; therefore, changes of

tooth proﬁles along with tooth height were

measured. Figure 72 shows tooth proﬁle error

curves after induction surface heating and after

quenching (Ref 20, 52). Prior to quenching, the

left tooth surface was marked with the letter “a”

and the right tooth surface with the letter “b.”

Measurements of the height proﬁle are shown

for two teeth, those marked 1 and 14, of gear

wheels 2 and 3. Considering that the tooth height

is 10 mm and the deviations are plotted in milli-

meters, their absolute value can be evaluated. It

ranges between 9 and 100 mm at each tooth

surface concerned.

One method of induction surface hardening

appropriate for large gear modules is known as

gap-by-gap hardening. It belongs to the pro-

gressive hardening methods. In this case, the

coil is placed so that it ensures a uniform gap

between the coil and the ﬂanks of two adjacent

teeth. The tooth gap-hardening method is very

demanding and requires much experience and

knowledge to achieve the desirable properties of

the gear. This method is also known as contour

Fig. 70

Change of whole depth of gear wheel after quench-

ing. Source: Ref 20, 52

Fig. 71

Change of root gear-wheel meter after quenching.

Source: Ref 20, 52

Fig. 69

Change of tip gear-wheel diameter after quenching.

Source: Ref 20, 52

470 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_417-501.pdf/Chap_13/ 18/8/2008 4:02PM Plate # 0 pg 470

hardening. It is an ideal method for heat treat-

ment of gears because it increases the hardness

on the tooth surface only slightly and decreases

the load-bearing capacity in the tooth root. Gears

heat treated in this way exhibit very good

behavior in operation, because compressive

residual stresses are present in the root of the

tooth. Gears with induction-hardened ﬂanks,

given that the dimensioning is carefully carried

out, can achieve the highest fatigue strength. To

verify the results of induction surface hardening,

it is necessary to discuss certain measures for

controlling the quality of the hardened layer. For

this purpose, hardness and microhardness mea-

surements supported by microstructural analysis

are commonly used. A disadvantage of this

procedure is that, due to the method of heating

and quenching (nonuniformly overheated left

and right tooth ﬂanks), slightly higher dimen-

sional deviations may be obtained than in

simultaneous hardening of both ﬂanks of the

same tooth (Ref 27).

Figure 73 shows the deviation in the dimen-

sions of the tooth after induction surface hard-

ening by heating with a coil that encircles the

gear tooth (Ref 15, 20, 55).

Measurements of gear teeth and gear gaps

after induction surface hardening show an

increased volume in the tooth root and thus a

smaller gap and increased volume at the tip of

the tooth (increased gear diameter). These

volume changes result in a slightly smaller width

of the tooth above the pitch circle and slightly

increased tooth width below the pitch circle

of the gear. These dimensional deviations are

relatively small and negligible in the case of

gears with small diameter or small module, but

may become more important in gears with larger

modules and greater tooth width (Ref 55, 56).

Surface hardening of gears is one of the most

frequent applications of induction surface hard-

ening. In the ﬁrst part of this chapter, different

methods of induction heating of gears was dis-

cussed, depending on gear size but also on the

properties expected of gears after heat treatment.

Fig. 72 Tooth proﬁle error curves (a) after induction surface heating and (b) after quenching. Source: Ref 20, 52

Fig. 73

Distortion of individual tooth shape after induction

hardening caused by volume changes. Source:

Ref 15, 20, 55

Induction Hardening / 471

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_417-501.pdf/Chap_13/ 18/8/2008 4:03PM Plate # 0 pg 471

Gears are axisymmetric machine elements;

therefore, volume changes due to the trans-

formation into a martensitic microstructure in-

duced by heat treatment are to be expected.

Teeth themselves are symmetrical, which means

that careful application of uniform heating of

particular gear teeth and uniform quenching,

regardless of the method applied, should not

cause distortion after heat treatment. In induc-

tion surface hardening, compressive residual

stresses are created in the surface, which, in

addition to hardness and wear resistance of

the surface, create a considerably high fatigue

strength and resistance to bending loads.

Figure 74 shows shape distortion after single-

frequency and dual-frequency induction surface

hardening (Ref 20, 57). Figure 74(a) shows a

simulation of distortion by showing the initial

shape of the tooth half-cut prior to hardening

(dotted line) and the modiﬁed shape after dual-

frequency induction surface hardening (con-

tinuous line). Figure 74(b) shows a similar

simulation of distortion but for conventional

single-frequency induction surface hardening.

In Fig. 74, the calculated actual size of dis-

tortion is 50 mm, with an equivalent geometrical

scale of 2.0 mm. A comparison of both cases of

hardening from the viewpoint of tooth distortion

clearly shows that distortion in single-frequency

surface induction hardening is twice that

obtained after dual-frequency hardening.

This difference in tooth distortion can be

attributed to the fact that in single-frequency

surface induction hardening, the martensitic

transformation occurs in the whole tooth. Con-

sequently, the volume change of the tooth

is greater and so is the distortion. In dual-

frequency surface induction hardening, only

the tooth contour becomes hardened. Con-

sequently, the volume change of the tooth is

smaller and so is the distortion, that is, it equals

only half of that obtained after single-frequency

hardening.

Residual Stresses after Induction Surface

Hardening and Finish Grinding

The last phase in the manufacturing of

crankshafts is ﬁne grinding, where achieving the

desirable condition of the surface and the surface

layer requires:

 Suitable dimensions of the particular bearing

locations with respect to the allowable

deviations

 Suitable surface roughness

 Grinding stresses are compressive or the

lowest tensile to maintain the favorable

stress proﬁle obtained by induction surface

hardening of the surface layer

 Smallest changes possible in the micro-

structure and also smallest changes in the

hardness and microhardness proﬁles in the

heat-affected zone after grinding

How is it possible to assure a desirable surface

and surface layer quality after induction surface

hardening and ﬁne grinding? Finding an answer

to this question requires a very good knowledge

of the process of grinding on the microlevel as

well as all mechanical and heat effects acting on

the layer of the workpiece, including the type

and condition of the grinding wheel. An all-

inclusive consideration of the numerous inﬂu-

ences of the kind and condition of the tool on the

changes on the surface and in the surface layer of

the workpiece in the given machining conditions

can be based on the descriptions of surface

integrity (Ref 58–63).

For the grinding process, the following con-

ditions have been selected:

 Different kinds of grinding

 Different grinding conditions (gentle,

normal, abusive)

Because of thermomechanical loads in the

thin surface layer during the grinding process,

Fig. 74

Distortion after induction surface hardening with

(a) dual frequency and (b) single frequency. Source:

Ref 20, 57

472 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_417-501.pdf/Chap_13/ 18/8/2008 4:03PM Plate # 0 pg 472

very complex physical and chemical processes

take place on the microlevel. For an accurate

description of the conditions in the contact zone

for machining a given material, it is necessary to

consider the grinding method as well as the kind

of material from which the grinding wheel is

made, its structure, and the grinding conditions.

The heat conditions in grinding are a result of the

contact conditions between the individual grind-

ing grains and the workpiece material as well as

deformation work in chip formation in the shear

zone. The amount of generated heat is strongly

dependent on the chosen machining conditions

and is abducted mainly through the chip, while a

smaller part of the generated heat is transferred

through heat transfer into the thin surface layer

of the workpiece. Heating of the chip does not

cause any particular difﬁculties, but heating of

the thin surface layer of the workpiece creates

the conditions for different mechanical and

thermokinetic processes, which cause micro-

chemical changes. Heating up the thin surface

layer of the workpiece can leave certain unde-

sirable effects that change the properties of the

part surface layer and thus harm its operational

abilities. The generated friction heat is trans-

ferred through the heat-transfer phenomenon

from the contact between the grinding grain and

the workpiece into the grinding grain.

The increased amount of heat on the grinding

grain or grinding wheel intensiﬁes the wear

processes and damage of particular grinding

grains, which follow very different and complex

mechanisms and, in the ﬁnal phase, affect the

serviceability and operational life of the tool.

The grinding tool consists of grinding grains,

each representing a process. Grinding grains are

interconnected with an appropriate binder,

which is deﬁned by different degrees of porosity

of the grinding wheel structure. Therefore, it is

necessary to also know the force needed to break

the grinding grain from a given structure, ex-

pressed by the grinding wheel hardness. Simul-

taneous changes in volume proportions of the

grinding grains and binder can create different

structures on the grinding wheel that do not be-

have in the same way. This means that by

changing the kind of material for the grinding

wheel and the binder, it is possible to achieve

equal effects by changing the structure of the

grinding wheel. It follows that by a suitable

combination of inﬂuential parameters, it is pos-

sible to achieve a longer life and wear resistance

of the wheel in equal kinematic conditions of the

wheel and the workpiece. The wear of the

grinding grains is a result of mechanical and

thermal effects, which are reﬂected in reduced

cuttability. Figure 75 shows the basic forms of

wear and damage on the grinding wheel grains,

expressed by characteristic changes on the

grinding grains (Ref 15, 64). Due to mechanical

loads on particular grinding grains, short-lived

but intensive heat effects are created in the con-

tact or on the friction surfaces with the work-

piece. In these cases, blunting of the grinding

grains (Fig. 75a), breaking off of the grind-

ing grains (Fig. 75b), or splitting of the grinding

grains (Fig. 75c) may happen. Requirements are

often set that a worn-out grain should fall from

the grinding wheel at a certain moment. A worn-

out grain typically has an increased contact

surface, which causes the forces on the grinding

grain during the cutting process to be higher than

the binding forces between the grains, and there-

fore, the grain falls out (Fig. 75d). It is expected

that the manufacturing engineer will carefully

choose all the parameters of the process so that

the worn-down grains fall out. This ensures

more efﬁcient machining and reduced thermo-

mechanical effects in the thin surface layer of the

workpiece material. To create a suitable quality

of the workpiece surface and surface layer, it

is necessary to ensure that grinding grains will

fall out, and that the activation of new, sharp

grinding grains is made possible.

It is very important to choose a cutting con-

dition that would make the grains fall out only

when their cuttability is reduced. Figure 75(e)

illustrates characteristic blunting of the grinding

grain due to chemical reactions at high tem-

peratures. Chemical reactions at high tempera-

tures are frequently followed by ﬁlling the pores

with overheated, highly plastic chips of the

workpiece material (Fig. 75f). When the pores

Fig. 75

Basic forms of wear and damage on the grinding

wheel grain. Source: Ref 15, 64

Induction Hardening / 473

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_417-501.pdf/Chap_13/ 18/8/2008 4:03PM Plate # 0 pg 473

on the wheel surface become ﬁlled, the tem-

perature in the contact zone rises and changes

the progress of the temperature cycle into the

depth of the workpiece. The maximum tem-

perature of the temperature cycle on the surface

rises, as does the temperature in the particular

depths of the workpiece material. This results in

greater depth of the heat-affected zone in the

material, which may have fatal consequences in

terms of surface layer properties of the work-

piece material.

Generally, blunting of the grinding grains

makes the contact surfaces between the grains

and the workpiece larger, which creates the

conditions for increased mechanical effects in

the contact zone and higher heat input, accom-

panied by stronger heat effects, in the surface

layer of the workpiece. A plastically deformed

layer is created because of interaction between

the grinding tool and the workpiece on the place

where the chip separates from the base.

A result is hardening of the thin surface layer

of the workpiece material and occurrence of

internal stresses that may lead to failure of the

thin surface layer and/or deformation of the

workpiece, with the presence of residual stresses

at the end of machining.

Macro- and microanalysis with optical and/or

electronic microscopes shows microcracks and/

or other damage on the surface caused by an

inadequate grinding method or procedure.

Microscopic assessment of the surface state and

damage on the surface quite often points to

inadequate grinding conditions. The most fre-

quent surface damage includes hollows, mars,

torn-off areas, built-up edges of the workpiece or

the tool, and so on. It is often necessary to con-

sider the generated heat effects that cause micro-

structural and/or chemical changes accompanied

by dimensional changes. The damage on the

workpiece surface should be taken very ser-

iously, since this may give rise to very detri-

mental friction conditions during operation with

another element in the mating pair.

In the analysis of microstructural changes in a

thin surface layer of the material after ﬁnish

grinding, it is possible to evaluate the size of the

heat-affected layer. Figure 76 shows the time

variation of the maximum temperature on the

surface at the particular depths with respect to

the workpiece speed, V

(work speed, V

)(Ref

15, 65, 66). By knowing the melting temperature

and the austenitization temperature of the dis-

cussed steel, the depths of the remelted layer and

the heat-affected layer can be deﬁned. Under

different grinding conditions, different tem-

perature cycles were obtained on the surface and

in the depth of the heat-affected layer, which has

effected microstructural changes and changes

in the microhardness and residual stresses

(Ref 64–66).

Figure 77 presents the temperature cycles on

the specimen surface as a function of depth in

the hardened steel at given grinding conditions

(Ref 15, 65). The temperature cycles can be

treated separately as a heating phase and then a

cooling phase.

The maximum temperatures achieved at the

surface and in the surface layer, respectively, are

also very important (Ref 64). A distinction can

be made between three characteristic cases of

temperature cycles:

 The maximum temperatures at the surface

and in the surface layer, respectively, are

higher than the melting temperature of the

specimen material, depending on the tem-

perature cycles (Fig. 78a) (Ref 65). Such

conditions may occur due to very sharp

grinding conditions or the selection of an

inappropriate grinding wheel with regard to

the specimen material. The depth of the re-

melted layer is only a few micrometers and

makes a very ﬁne ledeburite microstructure

containing ﬁne cementite spread in residual

austenite. The newly formed microstructure

has a slightly lower hardness than marten-

site. The residual stresses in the thin surface

layer will be tensile, due to plastic defor-

mation of the surface layer in grinding

caused by tensile forces in the contact zone

of the specimen material. To this should be

Fig. 76

Maximum temperature drop as a function of depth in

the hardened steel during grinding with various work

speeds, V

. Source: Ref 15, 65, 66

474 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_417-501.pdf/Chap_13/ 18/8/2008 4:03PM Plate # 0 pg 474

added the tensile stresses induced by the

occurrence of residual austenite.

 The maximum temperature in the contact

zone is lower than the temperature required

for the beginning of melting of the given

material and higher than the austenitizing

temperature. The lower temperature, that is,

the austenitizing temperature, will shift

because of a high rate of heating of the

specimen toward higher temperatures, as

known from transformation diagrams

(Fig. 78b) (Ref 65). Provided the previous

microstructure of the surface layer was

martensite-cementite-carbide, a ﬁner mar-

tensitic microstructure with a higher carbon

content, obtained at the expense of cemen-

tite-carbide phases and with the possibility

of a lower content of residual austenite, may

be expected in the thin surface layer after

grinding. The modiﬁed content of the

cementite-carbide phase depends on the

heating conditions, whereas the content of

the residual austenite depends on the cooling

conditions.

 The maximum temperature in the contact

zone is lower than the temperature required

for the beginning of austenitization and

higher than the lower temperature that is

limited by the temperature of steel temper-

ing. It is equal to approximately 200



(Fig. 78c) (Ref 65). The grinding conditions

are very mild, so that with the selection of

the correct type of grinding wheel, no

important changes are expected in the sur-

face layer. Only martensitic tempering may

occur in the surface layer if it was not per-

formed during the induction surface hard-

ening of the specimen.

The engineer should be aware that surface

integrity depends on the tribological conditions

in the operation of a component/assembly.

Therefore, adequate knowledge for the assess-

ment of tribological conditions of a component

in operation is very important for the prescrip-

tion of machining that would result in the desired

surface condition and subsurface layer. In gen-

eral, there are two tribological systems. The ﬁrst

is the one present during machining, and the

second is acting in operation. In both tribologi-

cal systems, however, it is the properties of the

workpiece material and its condition prior to and

after machining that play a key role. Under dif-

ferent machining conditions of grinding, differ-

ent temperature cycles were obtained on the

surface and in the depth of the heat-affected

zone, which has effected microstructural chan-

ges and changes in the microhardness and resi-

dual stresses. Thus, on the surface, a maximum

temperature higher than the temperature of

melting of the workpiece material was obtained.

The depth of the remelted layer is only a few

micrometers and makes a very ﬁne ledeburite

microstructure containing ﬁne cementite spread

in residual austenite. The newly formed

Fig. 78

Maximum temperature drop as a function of depth in the induction surface-hardened steel at various speed, V

. Source:

Ref 65

Fig. 77

Grinding temperature cycles in different depths in

the hardened steel at given grinding conditions.

Source: Ref 15, 65

Induction Hardening / 475

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_417-501.pdf/Chap_13/ 18/8/2008 4:03PM Plate # 0 pg 475

microstructure has a slightly lower hardness than

martensite. The residual stresses in the thin sur-

face layer will be tensile, due to plastic defor-

mation of the surface layer in grinding caused

by tensile forces in the contact zone of the

workpiece material as well as the tensile stresses

induced by the occurrence of residual austenite.

The relative grinding stress is obtained by

measuring the residual stress after induction

surface hardening, followed by measuring the

same spot after induction hardening and grind-

ing, then calculating their difference. Figure 79

shows the measured absolute residual-stress

proﬁle after induction surface hardening and the

measured residual-stress proﬁle after induction

surface hardening and grinding (Ref 15).

A relative grinding stress represents the dif-

ference between measured residual stress on

specimens that were induction surface hardened

and induction surface hardened and grinded.

A relative grinding tensile stress equals

+425 N/mm

, and then the sign changes in the

depth at approximately 175 mm, as shown in

Fig. 80.

Figure 80 shows the absolute residual-stress

proﬁle after induction surface hardening and

grinding on bearing location “A” as well as the

average residual stress after induction surface

hardening. From this, the relative grinding stress

can be calculated (Ref 15).

Figure 81 shows a zone of relative grinding

stresses occurring under gentle grinding con-

ditions (Ref 54).

With gentle grinding conditions, compressive

relative grinding stresses will prevail at the

surface. It is only in the subsurface that the

tensile inﬂuence can be felt.

The relative grinding stresses are low and

practically negligible, although they exert a

positive inﬂuence on the original residual

stresses after induction hardening. The zone was

deﬁned by the upper limiting values character-

istic of greater depths under normal grinding

conditions and by the lower limiting values

characteristic of gentle grinding conditions.

Generally, care should be taken when choosing

grinders and grinding conditions. The chosen

grinder should ensure regeneration of the abra-

sive grains, so that the worn-out grains may fall

off and the new ones grind the surface. The

selected grinding conditions should ensure that

a suitable force be applied to the worn-out

abrasive grains so that they may fall off and new,

sharp abrasive grains become active. Figure 82

Fig. 80

Subsurface proﬁle of relative grinding stress on

bearing location “A”. Source: Ref 15

Fig. 79

Subsurface residual-stress proﬁle after induction

surface hardening and grinding (absolute stress) on

bearing location “A”. Source: Ref 15

Fig. 81

Relative grinding stress region through thin surface

layer after gentle grinding conditions. Source: Ref 54

476 / Failure Analysis of Heat Treated Steel Components

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_417-501.pdf/Chap_13/ 18/8/2008 4:03PM Plate # 0 pg 476

shows a zone of relative grinding stresses

occurring under abusive or normal grinding con-

ditions (Ref 54). The zone of grinding stresses

was determined from the proﬁle of residual

stresses after induction hardening and also after

induction hardening and grinding (Ref 64).

By deducting the two residual stresses, the so-

called relative grinding stresses were deter-

mined. The abusive and normal grinding con-

ditions produce remelting of the thin surface

layer, that is, its rehardening. With the abusive

and normal grinding conditions, the tensile

relative grinding stresses are always predomi-

nant at the surface. At greater depths, there is

still the inﬂuence of remelting under the abusive

grinding conditions, whereas rehardening oc-

curs under normal grinding conditions.

For the sake of clearness, the zone was deﬁned

by the upper limit of residual-stress values,

which is characteristic of the abusive grinding

conditions, and the lower limit, which is char-

acteristic of the normal grinding conditions

(Ref 64).

The results conﬁrm as predominant the resi-

dual stresses, stresses induced by the plastic

deformation of the material, and as a lesser

inﬂuence, tensile stresses caused by the forma-

tion of residual austenite. On the basis of the

measurements of residual stresses after induc-

tion surface hardening or induction surface

hardening and grinding, it can be concluded:

 For residual stresses after induction surface

hardening and grinding, the conditions of

abusive grinding are a more favorable

choice. They lower to a lesser extent the

desirable compressive residual stresses after

induction surface hardening.

 Grinding conditions can be chosen so that

the melting temperature of the workpiece

material (gentle grinding conditions) is not

exceeded. Then, the favorable compressive

stresses after induction surface hardening

are lowered due to plastic deformation of the

workpiece material during the process, and

thus, relatively low tensile residual stresses

are obtained. However, this will signi-

ﬁcantly lower the productivity.

 Special attention should be paid to the

selection of the type of grinding wheel in

terms of grinding wheel material, binding

agent, hardness, and pore density, since a

correct selection can contribute to higher

cutting efﬁciency concerning the plastic de-

formation of the workpiece material. In this

way, the grinding tensile stresses are kept as

low as possible and make the compressive

residual stresses induced by induction sur-

face hardening of the prevailing variety.

Induction surface hardening creates a very

desirable residual stress state. Residual stresses

are always of a compressive nature and are

usually present to the depth of the induction

surface-hardened layer.

However, a major difﬁculty in induction sur-

face hardening is ensuring a very slight/slow

variation in microhardness and the existence of

compressive residual stresses in transition areas

to the microhardness of the core material. By

gently grinding and varying the hardness and

existence of compressive stresses in the transi-

tion area, it is possible to diminish the notch

effect induced by stress concentration. Addi-

tional grinding of an induction surface-hardened

surface deteriorates the stress state in the surface

layer, since grinding has always induced tensile

stresses. By the correct selection of machining

conditions and grinding wheel properties, the

engineer will contribute to less tensile residual

stresses and will avoid deteriorating the favor-

able residual-stress state after induction surface

hardening (Ref 44, 59).

Hardness Proﬁles in the Induction

Surface-Hardened Layer

The induction surface-hardened layer was

analyzed by measuring the hardness and

microhardness and their relation to the micro-

structure (Ref 44). To establish some relation-

ships between microstructural changes versus

Fig. 82

Relative grinding stress region through thin surface

layer after abusive or normal grinding conditions.

Source: Ref 54

Induction Hardening / 477

Name ///sr-nova/Dclabs_wip/Failure_Analysis/5113_417-501.pdf/Chap_13/ 18/8/2008 4:03PM Plate # 0 pg 477