See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/305489957

Modeling of Carburization Parameters Process for Low Carbon Steel.

Article · January 2016

CITATIONS READS

6 753

4 authors, including:

Abbas Khammas Laith K Abbas

University of Technology, Iraq University of Technology, Iraq
20 PUBLICATIONS 36 CITATIONS 41 PUBLICATIONS 144 CITATIONS

SEE PROFILE SEE PROFILE

Jamal Jalal
University of Technology
12 PUBLICATIONS 23 CITATIONS

SEE PROFILE

All content following this page was uploaded by Laith K Abbas on 21 July 2016.

The user has requested enhancement of the downloaded file.

Eng. &Tech.Journal, Vol.34,Part (A), No.6,2016

Modeling of Carburization Parameters Process for Low Carbon Steel

Dr. Abbas Khammas Hussein

Materials Engineering Department, University of Technology/Baghdad
Dr. Laith Kais Abbas
Materials Engineering Department, University of Technology/Baghdad
Jamal Jalal Dawood
Materials Engineering Department, University of Technology/Baghdad
Nadeen Jaafar Ismae
Materials Engineering Department, University of Technology/Baghdad

Received on:26/11/2015 & Accepted on:9/3/2016

ABSTRACT:
This paper represents the carburization parameters for steel (1020) using Desirability Function
Analysis-DFA. The experiments were conducted using Taguchi (L9) orthogonal array.
Carburization parameters such as carburization temperature, carburization time and tempering
temperature were optimized by multi - response considerations depending on micro hardness and
were rate measurements. The optimal carburizing parameters had been determined by composite
desirability value obtained from desirability function analysis while significant contribution of
parameter was determined by analysis of variance (ANOVA). The analyses results showed that
optimal combination for higher hardness and lower wear rate were at (A2=920 oC, B2= 3 hours and
C3=120 oC). Confirmation test was also conducted to validate the test results. Mathematical models
for composite desirability, micro hardness and rate wear were determined. Experimental results
showed that the carburization performance can be improved effectively through desirability
approach.
Keywords: Taguchi, ANOVA, carburization temperature, carburization time, tempering
temperature.

INTRODUCTION:

O
ne of the most important groups of engineering materials is steels which represent the
widest applications in materials engineering. There are many types of steels, each of them
was designed to a particular application in engineering fields [1]. Recently, the researchers
were very interested in surface engineering of the steels to improve the life and performance of
products which may be used in automobiles and aerospace [2].
Several techniques like thermochemical treatments have been well studied and were widely used in
the industry to improve products life. In this method, the metals or non-metals will be modified and
a chemical reaction is followed into the surfacethe composition of product surface, the structure of
product surface and the properties of product surface will be changed at the end of the
thermochemical treatments. Carburizing, nitriding, aluminizing, chromizing and carbonitriding are
the most popular methods for industrial applications [3]. Carburizing is a thermochemical process
which is used to increase the resistance ofproducts surface wear and hardness [4].Carburizing is the
operation of adding carbon to the surface of low carbon steel at temperatures of austenitic region,
this operation usually is done at (850-950oC),On such range of temperatures, there will be a high
solubility of carbon which had stable crystal structure. So, a tough low carbon steel core can be

1069
Eng. &
&Tech.Journal,, Vol.34,Part (A
A), No.6,2016 Modelling of Carburizaation Parameeters Process for
f Low
Carrbon Steel

produuced with goood hardness and wear reesistance whhen a high caarbon surface layer quennched to
form martensite haardening [5]..
Usuallly carbon diffusion rate depends onn: steel chem mical compossition, carburrization tempperature
and cchemical com mposition of the carburizing mixture [6]. The succcessful of caarburizing opperation
can bbe done wheen a control on three faactors is achhieved, thesee factors are the temperaature of
carbuurizing, the timme of carburrizing and chemical compposition [7].
In this research, TTaguchi approoach was useed which is a powerful sttatistical tooll had multi sttatistical
techniques and abbilities of jooining and innteract with multiple proocesses variables identiffications
basedd on a few exxperiments. This T approachh is widely uused becausee of its efficieency in optim mize the
operaational variabbles for experimental dessign [8].The ttwo major toools used in Taguchi metthod are
orthoggonal array (OA) and thhe signal too noise ratio (S/N) ratioo. Taguchi’s method, in general
providde a significcant reductioon in the sizze of experiiments whicch can proviide speedingg up the
experrimental proccesses [9].
This research
r focuuses on the ccarburizing pprocess param meters optimmization of thhe (low carboon steel)
specimmens by (Taaguchi’s-(L9))) orthogonaal array conccept and theereby determ mining the ooptimum
values which will maximize the hardnesss valueand m minimize thee wear rate.IIn general thhere are
manyy approachess to solve thhe optimizattion problem m like genettic algorithm m, response surface,
factorrial design annd Taguchi approach
a whiich is considdered the mosst stable, as iit reduce the number
of expperiments. IImprovemennt in computeer technologyy made it eaasier to use T Taguchi apprroach in
appliccation. The m most commoon analysis technique byy computer and all tables and plotss in this
researrch was processed using ((MINITAB16).
Experimental woork:
Enginneering produucts like: cam ms, gears, shhafts, pinionss, keys, handd tools, agriccultural equippment...
etc, arre usually prroduced from m a low carboon steel alloyys. In this ressearch experiiments were done on
low ccarbon steel specimens
s faabricated acccording to (AASTM- A2766). The chem mical compossition of
the steeel is given iin table (1).

Taable (1): Speectrochemicaal analysis oof steel (10200) ASTM (A

A276)
Element Fe C Mnn P S Cr M
Mo Si
Wt% Rem. 0.17 0.33 0.01 0.01 00.006 0.0002 0.14
Thee prepared ssteel specimeens were em mbedded in the carburizing box, whhich was fillled with
carbuurizing agent (Graphite) aand then it was
w tightly sealed
s with clay cover tto prevent unnwanted
furnacce gas from eentering the carburizing
c bbox during heeating, this can be shownn in figure (1)).

(a) (
(b)
Figu
ure (1): a: C
Carburizing b
box clay covver, b: Carburized speciimens.

1070
Eng. &Tech.Journal, Vol.34,Part (A), No.6,2016 Modelling of Carburization Parameters Process for Low
Carbon Steel

The furnace temperature was adjusted to the required temperature (870, 910 and 950 oC), the loaded
steel carburizing box was charged into the furnace. When each of the furnace temperature reaches
the required carburizing temperature, it was then soaked at the temperature for the required time (2,
3 and 4 hours).The specimen was held at the specified time, the steel carburizing box was removed
from the furnace and the specimen was quenched in water.
The carburized test specimens were tempered at a temperature of (80, 100 and 120 oC), held for an
hour and then quenched in water. Hardness of carburized steel was determined using (Knoop
microhardness test). The same trial was repeated five times to reduce experimental errors, the
average value was quoted as the final hardness.Taguchi technique was used to plan the design of
experiments. The Taguchi method is a powerful tool for improving productivity during research and
development, so high quality products in a low cost can be achieved while production in short time
[10]thus the marriage ofexperiment design with optimization can control parameters to obtain best
results which is achieved in the Taguchi method. Orthogonal array (AO) provides a set of well
balanced (minimum experiments).
Taguchi’s signal-to-noise ratio (S/N) is a log-function of desired output, serve as objective function
for optimization, help in data analysis and prediction of optimum results.
Experiments of micro hardness were conducted based on the Taguchi method of three levels, each
level with three factors. The values taken by a factor are termed to be levels. The factors to be
studied and their levels chosen are detailed in table (2):

Table (2): Factors and their levels of carburization

Factor Factor Code Levels
1 2 3
Carburization temperature (oC) A 870 920 950
Carburization time (hrs) B 2 3 4
Tempering temperature (oC) C 80 100 120

The (S/N) ratio for micro hardness is calculated using the higher-the betterrule as depicted in the
following equation [11]:
S/N = 10 ∑ (1)
th
Where:(n) is the number of the experiments in the orthogonal array and (yi) is the (i ) value
measured. The corresponding (OA) according to the number of factors and their levels that will be
used in the experiment. Table (3) shows (L9) orthogonal array from table (2).

Table (3): The (L9) orthogonal array

Experiment No. Process Parameter
Carburization Carburization Time (hrs) Tempering
Temperature (oC) Temperature (oC)
1 870 2 80
2 870 3 100
3 870 4 120
4 910 2 80
5 910 3 100
6 910 4 120
7 950 2 80
8 950 3 100
9 950 4 120
1071
 Eng. &Tech.Journal, Vol.34,Part (A), No.6,2016 Modelling of Carburization Parameters Process for Low
Carbon Steel

In this research, a pin on disc wear tests were performed on carburized steel which has been
selected as work piece material. All experiments were performed by using a pin on disc wear
apparatus. This machine facilitates study of wear characteristics in sliding contact under desired
conditions. Sliding occurs between the stationary pin on a rotating disc. Normal load, sliding
distance and sliding speed can be varied to suit the test conditions. The pin specimen was tested in
pin on disc apparatus. To perform the test specimen was clamped in jaw. Pin weight losses were
measured using an electronic balance having an accuracy of (±0.001 mg). Dry sliding wear tests
were carried out using pin-on disk type wear tester at constant parameters, where the tests were
conducted at (10 min, 20 N, 250 rpm -sliding speed and 6 cm - sliding distance).
An (L9) orthogonal array was chosen to determine the responses of hardness and wear rate of the
specimens. The experiment consists of (9) tests, each raw in the (L9) orthogonal array and the
columns were assigned with parameters. The orthogonal array table in the Taguchi design method
was applied to testing data as shown in table (3).
Results and Discussion:
1- Taguchi results:
The Taguchi method depending on the objective, proposed three different mean square deviations
for the signal- noise (S/N) ratios, which are [10]:
 Nominal – the better.
 Larger – the better.
 Smaller – the better.
The three different signal-noises (S/N) ratios, corresponding to (n) experiments are presented
below:
 For nominal – the better:
S/N = ‐ 10 log ∑ = ‐ 10 log (2)
 For larger – the better:
S/N = ‐ 10 log ∑ (3)
 For – smaller the better:
S/N = ‐ 10 log ( ) (4)
Where: (S) denotes the standard deviation, (yi) the data obtained from experiments, (n) represents
the number of experiments.
In this research, an orthogonal array (L9) based on Taguchi method was applied to design of
experiments. The experimental results and response (S/N) ratio are summarized in table (4):

Table (4): Orthogonal experimental results for micro hardness and wear rate
Exp. No. A B C Knoop (S/N)Ratio Wearrate (S/N) ratio
hardness (g/cm)
1 1 1 1 610 55.7066 0.0000047 106.595
2 1 2 2 601 55.5775 0.0000043 107.230
3 1 3 3 577 55.2235 0.0000004 127.250
4 2 1 1 580 55.2686 0.0000005 125.368
5 2 2 2 550 54.8073 0.0000005 126.214
6 2 3 3 511 54.1684 0.0000067 103.427
7 3 1 1 518 54.2866 0.0000006 124.945
8 3 2 2 605 55.6351 0.0000054 105.304
9 3 3 3 603 55.6063 0.0000070 102.926

1072
Eng. &Tech.Journal, Vol.34,Part (A), No.6,2016 Modelling of Carburization Parameters Process for Low
Carbon Steel

Results of individual and composite desirability:

The objective of this study is to identify an optimal setting that minimizes the wear rate under wear
conditions and maximize the micro hardness for the carburized steel. To resolve this type of multi-
output parameter design problems, an objective function of (x), is defined as follows:

DF = ∏ ∑
(5)
F(x) = - DF (6)
Where: (diwi) is the composite desirability defined for the (ith) targeted output and the (wi) is the
weighting of the (diwi). For various goals of each targeted output, the desirability (diwi) is defined in
different forms [8].

For the one-sided transformation:

0
d i= (7)
1

The rate of increase for the desirability (di) in the (ith) response depends on the variable (r) in the
equation. For the three possible choices of (r) the following statements to hold, when the
desirability is constant, the relationship between (yi) and (di) is thought to be linear, and the value of
(r) is taken as one. When ( ), above the minimum acceptable values ( ) are of decreasing
marginal worth, the relationship between (di) and ( ) is thought to be concave and the value of (r)
is taken as less than one. When ( ) above the minimum acceptable values ( ) are of
increasing marginal worth, the relationship between ( ) and (di) is thought to be convex and the
value of (r) is taken as greater than one [10].

For the two-sided transformation:

di = , if (8)

di = , if (9)
0 Otherwise

As with (r), the value of (s) and (t) can be selected to reflect constant, increasing or decreasing
incremental worth of ( ) as it approach (Ti).The individual desirability (weighted desirability-(di))
is calculated for all responses (micro hardness and wear rate) depending upon the type of quality
characteristics, the selection of quality characteristic of micro hardness and wear rate are larger-the-
better and smaller-the-better respectively [10]. The computed individual desirability for each
quantity characteristics using equation (8) and (9) are shown in table (5):

1073
Eng. &Tech.Journal, Vol.34,Part (A), No.6,2016 Modelling of Carburization Parameters Process for Low
Carbon Steel

Table (5): Individual of composite desirability

Normalized Normalized Weighted Weighted Composite Rank
Values Values Desirability Desirability Desirability
(micro (wear Rate) (micro (wear Rate)
hardness) hardness)
1 0 1 0 0 9
0.916064670 0.083935079 0.957112673 0.289715514 0.277290390 4
0.685939842 0.314059757 0.828214853 0.560410347 0.464140173 2
0.715223565 0.284776053 0.845708913 0.533644126 0.451307593 3
0.415319535 0.584679888 0.644452896 0.764643635 0.492776805 1
0 1 0 1 0 8
0.076829285 0.923169919 0.27718096 0.960817318 0.266320267 5
0.953523283 0.046476490 0.976485168 0.215584067 0.210514644 7
0.934825037 0.065174724 0.966863505 0.255293408 0.246833879 6

The composite desirability values are calculated using the equation [10]. Equal weightage is given
to all responses (w1=w2= ½ and w=0.5). Finally these values are considered for optimizing the multi
response parameter design problem. The results are given in the table (6):

Table (6): Optimizing the multi response parameter design problem.

Level Carburization Carburization Tempering
Temperature (oC) Time (hrs) Temperature (oC)
1 0.24714 0.23921 0.07017
2 0.31469 0.32686 0.32514
3 0.24122 0.23699 0.40775
Delta 0.07347 0.08987 0.33757
Rank 3 2 1

From the value of composite desirability in table (6), the parameter effect and the optimal level are
estimated. The results are tabulated in table (6) and parameter effects are plotted in figure (2).

Figure (2): Mean plot of composite desirability.

1074
 Eng. &Tech.Journal, Vol.34,Part (A), No.6,2016 Modelling of Carburization Parameters Process for Low
Carbon Steel

Considering the maximization of composite desirability value in (table (6) and figure (2)), the
optimal parameter condition is obtained as:
 Carburization temperature level (2).
 Carburization time level (2).
 Tempering temperature level (3).

Results of (ANOVA) for composite desirability:

The purpose of using (analysis of variance-ANOVA)is to investigate which carburization parameter
significantly affect the performance characteristic [9]. This is accomplished by separating the total
variability of the composite desirability, which is measured by the sum of the squared deviations
from the total mean of the composite desirability, into contributions by each carburization
parameter and the error. Thus:

SST = SSF + SSe (10)

Where:

SSF = ∑   (11)
Where:

SST:Total sum of squared deviations about the mean.

 : Mean response for the (jth) experiment.
 : Ground mean of the response.
: Number of experiments in the orthogonal array.
SSF:Sum of squared deviations due to each factor.
SSe: Sum of squared deviations due to error.
In addition, (F) test was used to determine which carburization parameters have a significant effect
on the performance characteristic. ANOVA results for composite desirability are shown in table (7):

Table (7): ANOVA table for composite desirability

Source Degree of Sum of Mean sum F-ratio Contribution
Variation freedom squares of squares present
(DF) (SSF) (SSe)
Carburization 2 0.0564 0.0282 0.593 7%
temperature (oC)
Carburization 2 0.0498 0.205 4.305 48%
time (hrs)
Tempering 2 0.2842 0.143 2.985 34%
Temperature (oC)
Error 2 0.09523 0.048 - 11%
Total 8 0.48563 - - 100%

From the ANOVA results, the carburization time and tempering time were found to have significant
influence on the wear rate and micro hardness of carburized steel, compared to the influence of
carburization temperature.

1075
Eng. &Tech.Journal, Vol.34,Part (A), No.6,2016 Modelling of Carburization Parameters Process for Low
Carbon Steel

Mathematical model:
The relationship between the factors (carburization temperature, carburization time and tempering
time) and the performance measures (composite desirability, micro hardness and wear rate) were
modeled by multiple linear regression. The following equations are the final models in terms of
encoded parameters:

Composite Desirability = - 0.0617486 - 0.0029603 * Carburizing Temperature (oC)-

0.00110897 * Carburizing Time (hrs) +0.168787 * Tempering Temperature (oC).
(12)
Micro hardness (VHN) = 626.111 - 10.3333 * Carburizing Temperature (oC) -2.83333 *
Carburizing Time (hrs) - 13.5 * Tempering Temperature (oC). (13)

Wear Rate (g/cm) = 4.42758*10-6 + 6.11983*10-7 * Carburizing Temperature

(oC)+1.42152*10-6 * Carburizing Time (hrs) - 2.56018*10-6 *Tempering Temperature (oC).
(14)
From the equation (1), surface plot for composite desirability at different tempering temperature
and carburization temperature are plotted as shown in figure (3). This surface plot can help in the
prediction of the composite desirability at any zone of the experimental domain. It is clear from this
plot that the composite desirability increases with increase in tempering temperature and
carburization temperature.

Figure (3): Relationship of composite desirability and carburization temperature, tempering

temperature.

From equation (1), surface plot for composite desirability at different carburization time and
carburization temperature are plotted as shown in figure (4). It is observed from this plot that
composite desirability will improve at the middle values of carburization time and carburization
temperature.

1076
Eng. &Tech.Journal, Vol.34,Part (A), No.6,2016 Modelling of Carburization Parameters Process for Low
Carbon Steel

Figure (4): Relationship of composite desirability and carburization temperature,

carburization time.

Finally, figure (5) is the representation of surface plot which shows the effect of both the
carburization time and tempering temperature on the composite desirability.

Figure (5): Relationship of composite desirability and tempering temperature, carburization

time.

The surface plot for the figure (5) shows that the minimum value of composite desirability is
obtained at the lower levels of both carburization time and tempering temperature.

1077
Eng. &Tech.Journal, Vol.34,Part (A), No.6,2016 Modelling of Carburization Parameters Process for Low
Carbon Steel

Confirmation test:
Confirmation test is a final step recommended by Taguchi to verify experiment conclusion [8]. The
improvements of the performance characteristics, using the optimal level of carburization
parameters are verified. Estimated composite desirability [10] is calculated as follows:
  ∑   (15)

Where:
 : is the total composite desirability.
 : mean composite desirability at optimum level.
( ): the number of process parameter having significant contribution in multiple performance
characteristics.
As noted from table (4), the micro hardness is increased from (610 to 663) which were measured
by(Knoop) method and the wear rate decreased from (0.0000047 to 0.0000029). The estimated
composite desirability is increased from (0.2772 to 0.5138). It is clearly shown that the multiple
objectives of the carburization process are together improved remarkably.

Table (5): Optimum process parameters results

Initial process Prediction Experiments
parameters
Level A1 B1 C1 A2 B2 C3 A2 B2 C3
Micro hardness 610 - 663
Wear rate 0.0000047 - 0.0000029
Composite desirability 0.2772 0.5138 -
Improvement of composite - 0.2366 -
desirability

CONCLUSIONS:
For optimization of process which consists of less number of experiments mostly Taguchi
method is used. Essential requirements of carburization process are higher hardness and lower wear
rate. In this work, it was focused on using Taguchi method coupled with desirability function
analysis for solving multi-criteria optimization problem in the field of carburization process. The
following conclusions were obtained:
1- Experimental results showed that micro hardness and wear rate of carburized steel (1020) were
greatly optimized by using desirability function analysis.
2- The optimal set of parameters obtained were carburization temperature (A2=920 oC),
carburization time (B2= 3 hours) and tempering temperature (C3=120 oC).
3- The contribution of carburization time is maximum followed by tempering temperature and
carburization temperature.

1078
 Eng. &Tech.Journal, Vol.34,Part (A), No.6,2016 Modelling of Carburization Parameters Process for Low
Carbon Steel

REFERENCES:
[1]- FataiOlufemi, Simeon Ademola, IsiakaOluwole and Joseph Olatunde, "Pack Carburization of
Mild Steel, using Pulverized Bone as Carburizer: Optimizing Process Parameters", Leonardo
Electronic Journal of Practices and Technologies, Vol.1, Issue.16, PP:1-12,January-June, 2010.
[2]- Deshmukh .S.S,Thakare, S.R., “Implementationof Taguchi technique on Heat treatment process
of pinion gear, "International Journal of Engineering Research and Applications (IJERA) ISSN:
2248-9622, Vol.2, Issue.6, PP:592-598, November-December, 2012.
[3]- Linus O. Asuquo, Ihom P. Aondona, "Mathematical modeling of the hardness of case hardened
steel with respect to carburizing time" , International Journal of Metal and Steel Research
Technology, Vol.1, No.2, PP: 19 – 25,August 2013.
[4]- Belete K. A., Adhena A. W., Asmamaw T. A., Fekadu G. H., "An experimental study on the
mechanical characteristics oflow alloy carbon steels for better performance of traditional farm
implements in Ethiopia", International Journal of Materials Science and Applications, Vol.3, No.6,
PP:420-430, 2014.
[5]- Sunil Patidar, Abhishek Jain, Dushyant Singh, "Effect of tempering temperature and applied
load on various wear environment of carburized mild steel ", IOSR Journal of Mechanical and Civil
Engineering (IOSR-JMCE), ISSN: 2278-1684 Vol. 3, Issue.3, PP:62-69,Sep-Oct. 2012.
[6]-BenariouaYounes, Maza Samira, "Towards carburizing treatment applied on low alloy steel
16CN6", 2 èmeConférenceInternationale des Energies Renouvelables CIER-2014 Proceedings of
Engineering and Technology - PET Copyright – IPCO, 2015.
[7]- R.QIN AND G. HE, “Carburization Phenomenon of ENiCrFe7 NickelBased Alloy Covered
Electrode", School of Materials Science and Engineering, Shanghai Jiao Tong University,
Shanghai, China, WELDING JOURNAL,PP:83-90, March 2015.
[8]-Douglas C.Montgomery, "Design and Analysis of Experiments", John Wily & Sons, Inc., 2009.
[9]-Douglas C.Montgomery, Geroge C.Runger, “Applied Statistics and Probability for Engineers”,
5th edition, John Wiley, 2012.
[10]- Jiju Antony, "Design of Experiments for Engineers and Scientists", Butterworth Heinemann,
2009.
[11]-Mendenhall, Beaver, “Introduction to Probability and Statistics", 4th Edition, Cengage
Learning, 2013.
[12]- Bluman, "Elementary Statistics", 8th Edition, Mc-Grew Hill, 2012.

1079

View publication stats