64 The Open Materials Science Journal, 2010, 4, 64-73

Open Access
Modeling of Phase Transformation in Induction Hardening
K. Palaniradja*, N. Alagumurthi and V. Soundararajan

Department of Mechanical Engineering, Pondicherry Engineering College, Pondicherry – 605 014, India

Abstract: Mechanical properties of machined components are controlled by inducing phase transformations in the outer layer
of the materials. Induction hardening is one such manufacturing process where the surface hardness is enhanced while the
core is retained with the original structure and characteristics. In this study, a mathematical model had been developed to
predict the hardness and the volume fraction of martensite present in the hardened surface. Experiments applying induction
hardening were conducted on the following specimen materials AISI 1040, AISI 4140, AISI 4340, AISI 1055, AISI 6150 and
AISI 9255. The microstructures obtained from the experiment showed a moderate phase transformation of austenite to
martensite. The hardness and the volume fraction of martensite estimated from the experiments were found to match the
results of mathematical modeling as well as the theoretical model using regression analysis.
Keywords: Modeling, phase transformation, induction hardening, volume fraction, validation

1. INTRODUCTION component with any composition within the allowed range

for the induction hardening steel type. Hence, in this paper a
Induction hardening is an important manufacturing
mathematical modeling has been developed to predict the
process to control the mechanical properties of metal parts
volume fraction of martensite and hardness for the materials
where surface hardness is enhanced while the core retains the
considered in the current research [5].
original structure and toughness characteristics [1]. This is
possible by inducing suitable phase transformation in the 2. MODELING OF INDUCTION HARDENING
outer layer of the material. Two types of modification occur
Modeling of induction hardening is a complex issue
when phase transformation takes place while quenching. The
which involves numerically solving the Maxwell’s equations
first type called metallurgical interaction is kinetic modifi-
for electromagnetic field around workpiece, coupled with
cation which sometimes leads to a different morphology in
heat transfer principles for modeling the product thermal
the phase produced. The second type (commonly called
process (Nemkov, V.S., and Goldstein, R., 1999). The
transformation plasticity) is a mechanical modification
related to the progress of transformation, and takes place system is inevitably highly nonlinear because of temperature
dependent material properties of workpiece. In the past,
when plastic deformation occurs under [2] stresses lower
more attention of numerical simulation was paid to modeling
than the yield stress of the material (Denis, S., et al., 1985).
only certain aspects of the induction hardening problem;
The importance of thorough knowledge of the little effort has been paid to model the integrated process
transformation characteristics of steel in the solution of heat from heating to final hardening. Numerical simulation of the
treatment and steel selection problems and in the explanation magnetic field was ever conducted. They provided valuable
of stress related phenomena has long been realized. It is analysis for the magnetic vector potential which was a very
obvious that for the quick decisions, which are the order of important parameter for eddy current and Joule heat
the day, metallurgists and materials engineers cannot determination. Simulation of induction heating has been
undertake expensive and time-consuming metallographic carried out and provided models for solving coupled
experiments to determine volume fraction of martensite and electromagnetic/thermal analysis. Recently, quenching and
maximum hardness value obtainable, considering heating phase transformation process has been simulated. And the
and cooling curves relevant to their actual technology and Koistinen-Marburger law and Avrami equation was applied
the work material. Instead, they must turn to the literature, for determination of volume fraction of micro-constituents
but unfortunately the data found, i.e. the chemical formed in the quench cooling process. However, few studies
composition, grain size, heating and cooling curves indicated have been found to predict the hardness distribution using
on the existing diagrams, differ from those required [3,4]. It numerical simulation.
is felt that the most direct assistance to the materials
Jiankun Yuan et al., (2003) [6] developed an integrated
engineers employed in heat treatment industry is to provide
FEA based modeling system with the capabilities to simulate
them with solution in the form of mathematical models to
the electromagnetic field induced heating process, austenite
predict the hardness and volume fraction of martensite at any
state holding processes, and severe quenching processes.
point in the Induction hardened zone of a given steel
This modeling system helps in determining volume fraction
of micro constituent such as martensite formed in the quench
*Address correspondence to this author at the Department of Mechanical cooling process, and final hardness distribution pattern in
Engineering, Pondicherry Engineering College, Pondicherry-605014, India; work piece. However, there is a restriction in using this
Tel: 91 413 2655281-287, Ext. 252,259; Fax: 91 413 2655101; E-mail: model because this model involves analysis of interaction of
[email protected]

1874-088X/10 2010 Bentham Open

Modeling of Phase Transformation in Induction Hardening The Open Materials Science Journal, 2010, Volume 4 65

continuous cooling curve with Temperature –Time – 3.1. Model Regression Analysis
Transformation diagram. It must be remembered that cooling
AISI 1040
curves are difficult to obtain in Induction hardening
applications. This is because the exact conditions are Coeff =
difficult to duplicate in the laboratory (Scott Mackenzie.,
2002). 1.0000 5.5000 1.3400 15.0000 80

This paper deals with the model, which may be 1.0000 5.5000 1.3400 17.5000 78
considered as an extension of Yuan model, developed to 1.0000 5.5000 1.3400 20.0000 79
predict the volume fraction of martensite theoretically. A
1.0000 5.5000 1.7200 15.0000 80
Regression model has also been developed to estimate the
Hardness in HRA for steel materials, AISI 1040, AISI 4140, 1.0000 5.5000 1.7200 17.5000 78
AISI 4340, AISI 1055, AISI 6150 and AISI 9255. These
1.0000 5.5000 1.7200 20.0000 82
models overcome the drawbacks in the modeling system
explained above. Experimental validation of the developed 1.0000 5.5000 2.1400 15.0000 83
model has also been presented here. 1.0000 5.5000 2.1400 17.5000 74
3. MATHEMATICAL MODEL TO PREDICT THE 1.0000 5.5000 2.1400 20.0000 81
HARDNESS
1.0000 7.0500 1.3400 15.0000 74
In order to develop the mathematical model to predict the
hardness and volume fraction of martensite for Induction 1.0000 7.0500 1.3400 17.5000 78
hardening of steel materials (G1 – Material and G2 – 1.0000 7.0500 1.3400 20.0000 77
Material), Induction hardening experiments have been
1.0000 7.0500 1.7200 15.0000 78
conducted.
1.0000 7.0500 1.7200 17.5000 74
Table 1 shows the details about the operating conditions
and the materials. 1.0000 7.0500 1.7200 20.0000 76
3
Table 2 shows the experimental results in the 3 Design 1.0000 7.0500 2.1400 15.0000 78
Matrix for the G1 and G2 Induction hardening rack materials 1.0000 7.0500 2.1400 17.5000 74
respectively.
1.0000 7.0500 2.1400 20.0000 78
Regression analysis has been carried out using MATLAB
and the Regression equations (Equation to predict the 1.0000 8.5000 1.3400 15.0000 69
hardness of the materials) have been found and the same are 1.0000 8.5000 1.3400 17.5000 65
given below:

Table 1. Experimental Conditions and Work Materials

Levels Actual Code

S. No. Variables Unit Notation
Low Medium High Low Medium High

1 Power potential kW/inch2 P 5.5 7.05 8.5 L1 L2 L3

2 Scan speed m/min S 1.34 1.72 2.14 L1 L2 L3
3 Quench flow rate Litres/min Q 15 17.5 20 L1 L2 L3
3
Design of Experiment: Factorial Design of Experiment with 3 Design matrix.
Regression analysis.
Material: Group 1 (Carbon content ranging from 0.35 -0.45%, Maximum HRC possible 60).
Group 2 (Carbon content ranging from 0.45 – 0.6%, Maximum HRC possible 65).

Material Steel Designation Chemical Composition in Percentage

Group 1 AISI 1040 C-0.35%, Si-0.10%, Mn-0.60% S&P each 0.06%

AISI 4140 C-0.35%, Si-0.10%, Mn-0.50%, Cr-0.90, Mo-0.20% S&P each 0.05%
AISI 4340 C-0.35%, Si-0.10%, Mn-0.45%, Cr-0.90%, Mo-0.20%, Ni-1.30% S&P each 0.05%

Group 2 AISI 1055 C-0.45%, Si-0.10%, Mn-0.50%, S&P each 0.06% (max)
AISI 6150 C-0.50%, Si–0.50%, Mn-0.50%, Cr-0.80%, V-0.15% S&P each -0.05 (each)
AISI 9255 C-0.55%, Si–1.50%, Mn-0.70%, S&P each -0.05 (max)
66 The Open Materials Science Journal, 2010, Volume 4 Palaniradja et al.

Table 2. Experimental Results for G1 and G2 Materials for Induction Hardening

Hardness in HRA for G1 Materials Hardness in HRA for G2 Materials

S. No. P S Q
AISI AISI AISI AISI AISI AISI
1040 4140 4340 1055 6150 9255

1 5.5 1.34 15 80 83 82 84 83 84
2 5.5 1.34 17.5 78 81 83 83 82 80
3 5.5 1.34 20 79 80 78 84 80 84
4 5.5 1.72 15 80 78 80 80 80 83
5 5.5 1.72 17.5 78 76 78 80 84 82
6 5.5 1.72 20 82 78 79 84 79 81
7 5.5 2.14 15 83 82 80 81 80 83
8 5.5 2.14 17.5 74 80 77 83 84 82
9 5.5 2.14 20 81 82 80 82 84 84
10 7.05 1.34 15 74 74 74 78 78 78
11 7.05 1.34 17.5 78 72 72 80 70 75
12 7.05 1.34 20 77 76 73 78 73 71
13 7.05 1.72 15 78 78 70 70 70 70
14 7.05 1.72 17.5 74 75 75 77 76 77
15 7.05 1.72 20 76 78 75 80 77 80
16 7.05 2.14 15 78 75 74 75 73 80
17 7.05 2.14 17.5 74 75 72 73 75 78
18 7.05 2.14 20 78 74 75 73 74 75
19 8.5 1.34 15 69 69 75 78 78 73
20 8.5 1.34 17.5 65 67 69 70 70 73
21 8.5 1.34 20 67 65 67 68 67 70
22 8.5 1.72 15 64 68 68 70 68 68
23 8.5 1.72 17.5 62 68 63 74 64 69
24 8.5 1.72 20 63 70 65 70 67 70
25 8.5 2.14 15 68 66 68 67 70 67
26 8.5 2.14 17.5 67 68 66 69 69 69
27 8.5 2.14 20 65 68 64 72 70 74

1.0000 8.5000 1.3400 20.0000 67 Therefore, for AISI 1040,

YH = 107.2820-4.5995P+0.1873S-0.0889Q (1)
1.0000 8.5000 1.7200 15.0000 64
1.0000 8.5000 1.7200 17.5000 62 For AISI 4140,
YH = 102.33421-4.0993P+0.4140S+0.0001Q (2)
1.0000 8.5000 1.7200 20.0000 63
For AISI 4340,
1.0000 8.5000 2.1400 15.0000 68 YH = 111.5611-4.1474P-2.3059S-0.2889Q (3)
1.0000 8.5000 2.1400 17.5000 67 For AISI 1055,
1.0000 8.5000 2.1400 20.0000 65 YH = 106.7885-3.8179P-3.8671S+0.1778Q (4)

The coefficients for the formation of hardness equation For AISI 6150,
are, YH = 108.2939-4.1947P-0.2082S-0.2000Q (5)

107.2820 For AISI 9255,

YH = 102.9862-4.0.759P+0.6014S+0.0667Q (6)
-4.5995
0.1873
-0.0889
Modeling of Phase Transformation in Induction Hardening The Open Materials Science Journal, 2010, Volume 4 67

4. MATHEMATICAL MODEL TO PREDICT THE variables P, Q and S (P = 5.5 kW/inch2, S = 1.72 m/minutes
VOLUME FRACTION OF MARTENSITE and Q = 15 litres/minutes) obtained in the optimization study
and it is found that,
To predict the volume fraction of marteniste for
Induction hardening of materials (G1 and G2 materials) Constant c = 75.28 = 13.68P
mathematical model has been developed as a part of the b = 90.85 = 6.05Q
current research work. For this modeling work, a computer
model given by Yuan et al., (2003), has been taken into a = 77.85 = 45.26S
account and extended to make it suitable for the present Therefore, the mathematical model to find the volume
research work. Yuan et al., (2003), developed the following fraction of martensite with the process variables P, Q and S
mathematical model (equation 7) to predict HRC and is derived as,
Volume fraction of martensite using FEM for the Induction
hardened components. fm = [(HRC) max – 13.68P +6.05Q] / 45.26S (14)
2
HRC = afm + bfm +c (7) (or)
where a, b and c are constants and dependent only on carbon fm = {[(HRC) max – 13.68P + 6.05Q] 0.02206S-1} (15)
content in the materials. Further, it is given that for the For the Group 1 materials (carbon content 0.35 – 0.45%)
materials, the hardness lies in the range 56 – 60 HRC (Thelning, K.E.,
AISI 1070 (0.65% Carbon) a = 80.91 b = -97 c = 81.61 1984) [7]. By extrapolation and interpolation it is found that
the constants a, b and c for the Group 1 materials as a =
AISI 1053 (0.55% Carbon) a =77.84 b = -90.85 c = 75.28 74.77, b = -83 and c = 68.95.
Equation 7 can be rewritten in the form of In order to verify the obtained constants the following
afm2 + bfm = HRC – c (8) calculation is carried out using the equation 11, by assuming
HRC as 60 when fm = 1.
fm (afm + b) = HRC – c (9)
fm = (HRC – c – b) / a
In general for the Induction hardened medium carbon
alloy steels the hardness value lies in the range 56 – 65 HRC fm = (60 – 68.95 +83) / 74.77
and volume fraction of martensite between 0.9 and 1.0 fm = 0.9903
(Thelning, K.E., 1984). Hence it is assumed that fm = 1,
therefore from the equation 9, it is obtained that, It shows that the obtained values of a, b and c are
absolutely correct. By following the same procedure as that
afm +b = HRC- c (10) adopted in the development of volume fraction of martensite
The equation 10 is used in the model development. of Group 2 materials, the mathematical model to predict the
Equation 10 becomes, ‘fm’ for the Group 1 materials has been obtained and
presented below.
fm = (HRC – c – b) / a (11)
fm = {[(HRC) max – 11.96P+5.27Q] / 43.5S} (16)
On substituting a, b, and c values of 0.55% carbon steel
material (AISI 1053) in the equation 11, we get (or)
fm = [(HRC) - 75.28 + 90.25] / 77.85 (12) fm = {[(HRC) max – 11.96P + 5.27Q] 0.0230S-1} (17)
For a maximum hardness of 65 HRC the volume fraction Thus the mathematical models( equations 1,2,3,4,5, and
of martensite should be equal to one. The following 6- pertaining to HRA; 14(or) 15 and 16 (or) 17 – pertaining
calculations will justify the above said statement. to fm) developed are used to predict the volume fraction of
martensite and hardness respectively without conducting the
fm = [65 – 75.28 + 90.25] / 77.85 = 1.027 (approximately = 1) trials on Induction hardening furnace [8,9].
Therefore, the equation 12 can be rewritten as
5. VALIDATION OF THE MODEL
fm = [(HRC) max - 75.28 + 90.25] / 77.85 (HRC) max = 65
In this study Induction hardening is carried out on AISI
(13)
1040, AISI 4140, AISI 4340, AISI 1050, AISI 6150 and
In the present analysis the process variables Power AISI 9255. These types of steel materials chosen for the
potential (P), Scan speed (S) and Quench flow rate (Q) of studies because such steels are typical candidate materials
Induction hardening process are taken for the study of phase for a variety of automobile components especially Rack that
transformation, model development, simulation and require local hardening [10]. The theoretical simulations
validation. Further, it is construed that there is a definite (results obtained from the model developed) are compared
relation between the above said process variables with the with results produced experimentally. Koistinen equation 18
hardness and volume fraction of martensite of the Induction is used to find the volume fraction of marteniste
hardened materials. theoretically,
In order to develop a mathematical model to predict the fm = 1- exp {-0.011 [Ms – Tq]} (18)
volume fraction of marteniste for the materials (Group 2 –
where, Ms – starting temperature of marteniste formation
carbon content 0.5 – 0.6%) using the above mentioned
(Rajan, T.V., et al., 1998) [11]
process variables an analysis has been carried out by
replacing the constants a, b and c with the optimal process
68 The Open Materials Science Journal, 2010, Volume 4 Palaniradja et al.

Ms (°C) = 561 -474(%C) – 33(%Mn) -17(%Ni) -17(%Cr)-21 The volume fraction of martensite (theoretical) is found
(%Mo) (19) as per the details given below using the model developed for
AISI 1040 steel material with the values of process variables
Tq – Quenchant temperature at which fraction of martensite
used in the trials.
is formed
fm = {[(HRC) max – 11.96P+5.27Q] / 42.2S}
For the material AISI 1040, the Ms temperature is found
as per the following procedure fm={[60-12*5.5+5.3*15]/42.2*1.72}
AISI 1040 - Carbon = 0.35%, Si = 0.1% and Mn = 0.6%, fm={[60-66+79.05]/72.6}
Ms (°C) = 561 -474(%C) – 33(%Mn) -17(%Ni) -17(%Cr)- fm=1.006
21(%Mo)
The hardness value (theoretical) is found by using the
= 561 – 474 (0.35) – 33 (0.6) Regression model equation.
= 375.3°C Regression equation for the AISI 1040
Similarly, for the other materials also the Ms YH = 107.2820-4.5995P+0.1873S-0.0889Q
Temperature is calculated and given in Table 3.
where, P=5.5, S=1.72 and Q=15
Table 3. Ms Temperature for G1 and G2 Materials
On substitution theoretical, YH =107.2820 -
4.5995*5.5+0.1873*1.72-0.0889*15
AISI Grade Ms (°C)
= 107.2820-25.3+0.322-1.3
1040 375.3 = 81 HRA
4140 359.1 For different conditions, theoretical “HRA” and “fm”
4340 338.65 values are predicted using the mathematical models and
results are compared with the experimental values in the
1055 331.2
following Tables 4-9 for the G1 and G2 materials.
6150 293.9
6. RESULTS AND DISCUSSION
9255 277.2
Micro structural analysis [12] of Induction hardened
Rack components can provide important information
Random trials are carried out by varying the process regarding material properties, reliability and intensity of
variables P, Q and S by subjecting the G1 and G2 materials surface hardening process. The photo micrographs [13] of
in the Induction hardening furnace and the temperature Tq the Induction hardened specimens are studied. The surface
(Quenchant temperature at which fraction of martensite is hardness for the different heat treatment conditions for the
formed) is measured using thermocouple. Hardness of the materials AISI 1040 AISI 4140, AISI 4340, AISI 1055, AISI
each specimen is measured using Rockwell hardness tester in 6150 and AISI 9255 are measured and reported in the Table
HRA. 2.
The volume fraction of marteniste (experimental) is From the microstructures of medium carbon steels (Figs.
found as per the calculation given below for the material 1-6) it is evident that there is a considerable conversion of
AISI 1040, (Ms = 375.3 from Table 3 and Tq = 65°C austenite to martensite at the outer surface which improves
through measurement when P = 5.5kW/inch2,, Q = 15 litres the hardness in Induction hardened components. It is seen
/min and S = 1.72 m /minutes) from the Tables 4-9 under optimal conditions that the
volume fraction of martensite is almost 95% for both the G1
fm = 1- exp {-0.011 [Ms – Tq] and G2 materials and also the required hardness (79 – 82
fm = 1 – exp {-0.011 [375.3 – 65]} HRA) is obtained.
fm = 0.9670 Because of the temperature dependent non-linear
properties in Induction hardening process, analytical solution

Table 4. Validation of Simulated Results for the AISI 1040 Steel Material

Process Variables Theoretical Values Experimental Values

Tq % Error
P S Q HRA fm HRA fm

4.5 1.72 15 - 85.6 1.177 Over hardening More than one

5.5 1.72 15 65°C 81 1.006 80 0.96 1.25 4.123
5.8 1.74 15 110°C 79.6 0.95 80 0.95 0.50 0.00
6.0 1.76 15 140°C 78.7 0.9088 79 0.92 0.379 1.217
7 1.72 15 74.105 0.760 Under hardening Martensite formation is less
Modeling of Phase Transformation in Induction Hardening The Open Materials Science Journal, 2010, Volume 4 69

Table 5. Validation of Simulated Results for the AISI 4140 Steel Material

Process Variables Theoretical Values Experimental Values

Tq % Error
P S Q HRA fm HRA fm

4.5 1.72 15 - 84.60 11779 Over hardening More than one -

5.5 1.72 15 63°C 80.5 1.006 81.0 0.962 0.617 4.573
5.8 1.74 15 107°C 79.2 0.951 80.0 0.938 1.000 1.385
6.0 1.76 15 140°C 78.4 0.908 79.5 0.910 1.383 0.219
7 1.72 15 74.30 0.7646 Under hardening Martensite formation is less -

Table 6. Validation of Simulated Results for the AISI 4340 Steel Material

Process Variables Theoretical Values Experimental Values

Tq % Error
P S Q HRA fm HRA fm

4.5 1.72 15 - 84.0 1.177 Over hardening More than one -

5.5 1.72 15 60°C 80.2 1.01 82 0.967 2.195 4.446
5.8 1.74 15 105°C 79.0 0.95 81 0.930 2.469 2.150
6.0 1.76 15 137°C 78.0 0.90 79 0.90 1.265 0.000
7 1.72 15 74.0 0.760 Under hardening Martensite formation is less -

Table 7. Validation of Simulated Results for the AISI 1055 Steel Material

Process Variables Theoretical Values Experimental Values

Tq % Error
P S Q HRA fm HRA fm

4.5 1.72 15 - 85.0 1.21 Over hardening More than one -

5.5 1.72 15 65°C 82.0 1.03 83.0 0.95 1.204 8.420
5.8 1.74 15 110°C 81.0 0.96 81.0 0.92 0.000 4.347
6.0 1.76 15 140°C 80.0 0.92 79.5 0.89 0.628 3.370
7 1.72 15 76.0 0.76 Under hardening Martensite formation is less -

Table 8. Validation of Simulated Results for the AISI 6150 Steel Material

Process Variables Theoretical Values Experimental Values

Tq % Error
P S Q HRA fm HRA fm

4.5 1.72 15 - 86.0 1.209 Over hardening More than one -

5.5 1.72 15 47°C 82.0 1.03 81.0 0.96 1.234 7.291
5.8 1.74 15 65°C 81.0 0.96 82.0 0.92 1.219 4.347
6.0 1.76 15 77°C 80.0 0.92 80.0 0.90 0.000 2.222
7 1.72 15 75.5 0.768 Under hardening Martensite formation is less -

Table 9. Validation of Simulated Results for the AISI 9255 Steel Material

Process Variables Theoretical Values Experimental Values

Tq % Error
P S Q HRA fm HRA fm

4.5 1.72 15 - 86.0 1.20 Over hardening More than one -

5.5 1.72 15 36°C 82.2 1.03 84 0.94 2.142 9.574
5.8 1.74 15 52°C 81.0 0.96 83 0.92 2.409 4.347
6.0 1.76 15 70°C 80.0 0.92 81 0.92 1.234 0.000
7 1.72 15 76.08 0.76 Under hardening Martensite formation is less -
70 The Open Materials Science Journal, 2010, Volume 4 Palaniradja et al.

AISI1040-Un alloyed carbon steel

(Carbon Content :0.35%)

Surface hardened zone showing

martensite

Case to Core Transition zone

100X

Fig. (1). Microstructure of Induction hardened AISI 1040.

AISI 4140-chromium alloy steel

(Carbon Content : 0.35%)
Surface hardened zone showing
martensite

Case to Core Transition zone

100X

Fig. (2). Microstructure of Induction hardened AISI 4140.

AISI 4340-chromium Mo. Ni alloy steel

(Carbon Content : 0.35%)
Surface hardened zone showing
martensite

100X
Fig. (3). Microstructure of Induction hardened AISI 4340.
Modeling of Phase Transformation in Induction Hardening The Open Materials Science Journal, 2010, Volume 4 71

AISI 1055-Unalloyed Constructional steel

(Carbon Content : 0.45%)
Surface hardened zone showing
martensite

100X

Fig. (4). Microstructure of Induction hardened AISI 1055.

Surface hardened zone showing AISI 6150-silicon alloy spring steel

martensite
(Carbon content: 0.45%)

Case to Core
Transition zone

100X

Fig. (5). Microstructure of Induction hardened AISI 6150.

Surface hardened zone showing AISI9255-Silicon constructionl steel

martensite
(Carbon Content : 0.50%)

Case to Core
Transition zone

100X

Fig. (6). Microstructure of Induction hardened AISI 9255.

72 The Open Materials Science Journal, 2010, Volume 4 Palaniradja et al.

Table 10. Induction Hardening Mathematical Model

Hardness Equations
S. No. Materials Volume Fraction of Martensite (fm)
(Hardness in HRA)

01 AISI 1040 YH = 107.2820-4.5995P+0.1873S-0.0889Q

02 AISI 4140 YH = 102.33421- 4.0993P+0.4140S+0.0001Q fm = {[(HRC)max – 11.96P + 5.27Q]0.0236S-1}

03 AISI 4340 YH = 111.5611-4.1474P-2.3059S-0.2889Q (HRC)max = 60

04 AISI 1055 YH = 106.7885-3.8179P-3.8671S+0.1778Q
05 AISI 6150 YH = 108.2939-4.1947P-0.2082S-0.2000Q fm = {[(HRC) max – 13.68P + 6.05Q] 0.022S-1}
06 AISI 9255 YH = 102.9862-4.0.759P+0.6014S+0.0667Q (HRC)max = 65
2
P-Power Potential in kW/inch , S-Scan speed in m/min, Q-Quench flow rate in liters/min.

Table 11. Induction Hardening Mathematical Model

S. No. Materials Hardness Equations (Hardness in HRA) Volume Fraction of Martensite (fm)

01 AISI 1040 YH = 108-4.695P+0.4S-0.888Q

02 AISI 4140 YH = 118.68-6.63P+0.32866S-0.0667Q fm = {[(HRC)max – 11.96P + 5.27Q]0.0236S-1}

03 AISI 4340 YH =111.66-4.14P-2.358S-0.2884Q (HRC)max = 60

04 AISI 1055 YH=184.43-3.8175P-46.177S+0.177Q

05 AISI 6150 YH =110.7688-4.167P-0.2776S-0.244Q fm = {[(HRC) max – 13.68P + 6.05Q] 0.022S-1}
(HRC)max = 65
06 AISI 9255 YH =102.64-4.221P+0.41666S+0.088Q
where,
P-Power Potential in kW/inch2.
S-Scan speed in m/min.
Q-Quench flow rate in liters/min.

is very difficult to obtain [14]. When surfaces with complex 7. CONCLUDING REMARKS
geometric shape are heat treated the problems become more
cumbersome. It must also be remembered that cooling • Microstructural analysis indicates that there is a
curves are difficult to obtain in Induction hardening moderate conversion of austenite to martensite (Phase
applications [15]. Hence, mathematical models will allow Transformations) at the outer surface which improves
engineers/metallurgists to experiment with the critical the Hardness in Induction hardened components.
process variables such as Power potential, Scan speed • A Mathematical Model (details are given in the Table
Quench flow rate and Frequency before expensive 11) to predict volume fraction of martensite and
investment or unforeseen failures in mass produced items. hardness in the induction hardening process has been
A Mathematical Model (details are given in the Table 10) developed. It is shown that the numerical predictions
to predict volume fraction of martensite and hardness in the of the hardened layer are in good agreement with the
Induction hardening process has been developed for the experimental results.
Group 1 and Group 2 materials. ACKNOWLEDGEMENTS
Comparisons between the theoretically simulated and The authors are thankful to Rane (Madras) Pvt. Ltd.,
experimental results are shown in the Tables 4-9. It can be Thirubhuvanai, Pondicherry and IGCAR Kalpakkam for
observed that there is a good agreement between the two providing the experimental, testing and measurement
results. The difference between the results (error) may be facilities.
attributed to the lack of precision in measurement of Tq. It
can also be seen that there is a reduction in the starting REFERENCES
temperature of Martensite formation (Table 3) when the [1] Avner SH. Introduction to physical metallurgy. 2nd ed. New Delhi:
carbon percentage increases from 0.35 – 0.55. However, the Tata McGraw Hill Publishing Company Ltd. 1982; pp. 315-45.
addition of alloying elements improves the properties like [2] Denis S, Gautier E, Simon A, Beck G. Stress-phase-
wear resistance, corrosive resistance, ductility etc. but transformations-basic principles, modeling, and calculation of
internal stresses. Mater Sci Technol 1985; 1: 805-14.
decreases the hardness. [3] Colegate GT. The surface hardening of steel, metal treatment and
drop forging. Trans Metall Soc AIME 1951; 249-55.
[4] Colegate GT. The surface hardening of steel – Part – II –
mechanism of the carburising process and the effect of alloying
Modeling of Phase Transformation in Induction Hardening The Open Materials Science Journal, 2010, Volume 4 73

elements and heat-treatment. meal treatment and drop forging. [10] Melander M. Computer predictions of progressive induction
Trans Metall Soc AIME 1951; 63-70. hardening of cylindrical components. Mater Sci Technol 1985: 1;
[5] Zgaraja J. Computer simulation of induction hardening of moving 877-82.
flat charge. IEEE Trans Magnet 2003; 69: 1523-6. [11] Rajan TV, Sharma CP, Sharma A. Heat treatment principles and
[6] Yuan J, Kang J, Rong Y, Richard D, Siison Jr. Development of a techniques. 8th ed. India: Prentice-Hall Pvt. Ltd 1985.
computer model for induction hardening of steel proceedings of [12] Wang KF, Chandrasekar S, Yang HTY. An efficient 2d finite
IMECE’03. 2003 ASME International Mechanical Engineering element procedure for the quenching analysis with phase change.
Congress and Exposition 2003; pp.1-7. Trans ASME 1993; 115: 124-38.
[7] Thelning K. Heat and its heat treatment. 2nd ed. London: [13] Lakhtin YM. Engineering physical metallurgy and heat treatment.
Butterworths 1984. Moscow: Mir Publishers 1979.
[8] Liu CC, Xu XJ, Liu Z. A FEM modeling of quenching and [14] Halliday JS. Application of reflection electron microscopy to the
tempering and its application in industrial engineering. Int J Finite study of wear, proceedings of first international conference on
Elem Anal Des 2003; 39: 1053-70. lubrication and wear. Institution of Mechanical Engineers, London,
[9] Maity J, Mandal A, Behera AK, Ghosh AK. A mathematical model Paper 40, 1975; pp. 647-65.
to determine hardening behaviour of steels. IE(I)J MM 2004; 85: [15] Raghavan V. Physical metallurgy. Principles and practice. New
27-32. Delhi: Prentice-Hall of India 1984.

Received: October 6, 2009 Revised: October 8, 2009 Accepted: October 27, 2009

© Palaniradja et al.; Licensee Bentham Open.

This is an open access article licensed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/
3.0/) which permits unrestricted, non-commercial use, distribution and reproduction in any medium, provided the work is properly cited.