Int J Adv Manuf Technol

DOI 10.1007/s00170-013-5008-8

ORIGINAL ARTICLE

Influence of parameters and optimization of EDM

performance measures on MDN 300 steel using Taguchi
method
A. M. Nikalje & A. Kumar & K. V. Sai Srinadh

Received: 26 January 2011 / Accepted: 8 April 2013

# Springer-Verlag London 2013

Abstract Maraging steel (MDN 300) exhibits high levels Keywords Electrical discharge machining (EDM) .
of strength and hardness. Optimization of performance mea- Maraging steel . Discharge current . Pulse on time .
sures is essential for effective machining. In this paper, Pulse off time . Orthogonal array
Taguchi method, used to determine the influence of process
parameters and optimization of electrical discharge machin-
ing (EDM) performance measures on MDN 300 steel, has 1 Introduction
been discussed. The process performance criteria such as
material removal rate (MRR), tool wear rate (TWR), relative Electrical discharge machining (EDM) is an important
wear ratio (RWR), and surface roughness (SR) were evalu- manufacturing process for machining hard metals and alloys
ated. Discharge current, pulse on time, and pulse off time [1]. This process is widely used for producing dies, molds,
have been considered the main factors affecting EDM per- and finishing parts for aerospace, automotive, and surgical
formance. The results of the present work reveal that the components [2]. The process is capable of getting required
optimal level of the factors for SR and TWR are same but dimensional accuracy and surface finish by controlling the
differs from the optimum levels of the factors for MRR and process parameters [3]. EDM performance is generally eval-
RWR. Further, discharge current, pulse on time, and pulse uated on the basis of material removal rate (MRR), tool wear
off time have been found to play a significant role in EDM rate (TWR), relative wear ratio (RWR), and surface rough-
operations. Detailed analysis of structural features of ness (SR) [2]. The important EDM machining parameters
machined surface was done by using scanning electron affecting to the performance measures of the process are
microscope (SEM) to understand the influence of param- discharge current, pulse on time, pulse off time, arc gap, and
eters. SEM of electrical discharge machining surface in- duty cycle [4].
dicates that at higher discharge current and longer pulse A considerable amount of work has been reported by the
on duration give rougher surface with more craters, glob- researchers on the measurement of EDM performance on the
ules of debris, pockmarks or chimneys, and microcracks basis of MRR, TWR, RWR, and SR for various types of steels.
than that of lower discharge current and lower pulse on Rao et al. [5] studied the influence of process parameters on
duration. EDM of MDN 250 steel. They have considered discharge
current, pulse on time, and duty factor as performance mea-
sures whereas process parameters are MRR and SR. However,
in their study, parametric optimization was not done. TWR
and RWR ratios were not considered. Furthermore, they
A. M. Nikalje (*) : A. Kumar : K. V. S. Srinadh extended their studies and developed a hybrid model for SR
Department of Mechanical Engineering,
National Institute of Technology (NIT),
is to predict the behavior of the MDN 250 steel [6].
Warangal, 506-004, Andhra Pradesh, India MDN 300 steel possesses an extreme resistance to
e-mail: amnikalje@yahoo.com crack propagation, even in the most extreme environ-
ments. It is used in applications where high fracture
A. M. Nikalje
toughness is required or where dimensional changes
Department of Mechanical Engineering,
Government College of Engineering, have to remain at a minimal level. It is specially used
Jalgaon, 425001, Maharashtra, India in effective in the design of power shafts and low-
 Int J Adv Manuf Technol

Table 1 Chemical composition of MDN 300 steel Table 3 Working range of the process parameters and their levels

Alloying Ni Co Mo Ti Fe C Al Symbol EDM parameter Unit Level 1 Level 2 Level 3

element (%) (%) (%) (%) (%) (%) (%)
A Discharge current A 10 15 20
Amount 17–19 12 3–5 0.2–1.6 65–68 0.01 0.10 B Pulse on time μs 25 45 65
C Pulse off time μs 24 36 48

temperature cooling systems, rocket motor casings, and

the landing gear for certain planes [7]. The properties The present work describes the optimization of the EDM
and applications of both MDN 250 steel and MDN 300 performance measures using Taguchi method. In this meth-
steel are different. MDN 300 steel has less thermal od, it is required to consider all aspects of the design that
conductivity than MDN 250 steel; hence, MRR is dif- affect the deviation of functional characteristics of the prod-
ferent. In addition to this, the chemical composition uct from target values. It is also essential to consider methods
shows cobalt is not present in case of MDN 250 steel to reduce undesirable and uncontrollable factors that can
which affects the hardness at elevated temperatures. The cause functional deviations [9]. It is possible to evaluate the
melting point of MDN 300 steel is 1,427–1,454 °C effects of individual parameter independent of other param-
whereas the melting point of MDN 250 is 1,435– eters and interactions on the indentified quality characteris-
1,505 °C. However, it has not been found that there is tics by using this method [10–12]. Taguchi method is one of
no available result of the EDM process of this material. the popular methods used for optimization as it requires
Therefore, it is imperative to develop a suitable technology minimum experimental cost and decreases the effect of the
guideline for optimum EDM of MDN 300 steel. source of variation effectively [13]. TWR is an important
In EDM, for optimum machining performance mea- performance measure as it affects to the geometrical and
sures, it is an important task to select proper combina- dimensional accuracy of the machined surface. It is an inde-
tion of machining parameters [8]. Generally, the pendent performance measure and depends on process pa-
machining parameters are selected on the basis of oper- rameters. Also, it gives an amount of tool material eroded
ator’s experience or data provided by the EDM manu- from the tool in a given time. RWR is an important perfor-
factures. When such information is used during EDM, mance measure as it gives the amount of tool material eroded
the machining performance is not consistent. Data pro- as compared to MRR. So, RWR depends on MRR and TWR.
vided by the manufacturers regarding the parameter Both the factors will minimize the objective function; hence,
settings is useful only for most commonly used steels. we considered both MRR and RWR in the present study.
Such data is not available for special materials like
Maraging steels, ceramics, and composites. For these
materials, experimental optimization of performance 2 Scheme of investigation
measures is essential. Optimization of EDM process
parameters becomes difficult due to more number of In order to maximize the desirable performance measures
machining variables. Slight changes in a single parame- and minimize undesirable performance measures, the inves-
ter significantly affect the process. Thus, it is essential tigation was done in the following sequence:
to understand the influence of various factors on EDM
process. Analytical and statistical methods are used to Table 4 Experimental layout using an L9 (34) OA
select best combination of process parameters for an
Sr. no. EDM parameter
optimum machining performance.
A B C
Discharge current Pulse on time Pulse off time
Table 2 Properties of MDN 300 steel
1 1 1 1
Property Quantity
2 1 2 2
Density 8.1 g/cm3 3 1 3 3
Specific heat, mean for 0–100° C 813 J/kg° K 4 2 1 2
Melting point 1,427–1,454° C 5 2 2 3
Thermal conductivity 25.8 W/m K 6 2 3 1
Yield tensile strength 758 MPa 7 3 1 3
Electrical resistivity 0.174×10−4 Ω cm 8 3 2 1
Hardness 34 BHN 9 3 3 2
Int J Adv Manuf Technol

Average
& Selection of workpiece material and electrode material.

5.62
6.61
7.71
6.48

7.32
5.92
6.96
8.15
&

7
Identify the important EDM process parameters.
& Determine the working range of the identified process
parameters.

Trial 3

5.59
6.63
7.73
6.51
7.12
7.36
5.89
7.03
8.18
& Select the orthogonal array (OA; design of matrix).
& Conduct the experiments as per the selected OA.
& Record the performance measures (i.e., MRR, TWR,

Trial 2

5.58
6.51
7.84
7.35

7.05
6.09
7.39
8.37
6.4
RWR, and SR).
& Find the optimum condition for performance measures

SR (μm)
and indentify the significant factors.

Trial 1

5.69
6.69
7.58
5.59

7.54
5.78
6.47
&

7.5

7.9
Conduct the confirmation test.

2.1 Selection of the workpiece material and electrode

Average

27.33
20.15
13.37
27.76
21.88
23.83

25.33
25.38
material

30.9
The workpiece material employed in this study was MDN

Trial 3
300 steel. Copper was selected as an electrode material as it

27.41
19.78
13.39
27.74

23.79
30.94
25.36
25.34
21.9
is commonly used because of its high thermal conductivity
and electrical conductivity. The chemical composition and
properties of MDN 300 steel are shown in Tables 1 and 2,

Trial 2

29.35

14.93
29.58
22.58
24.74
32.78
25.74
24.55
21.5
respectively.

2.2 Identify the important EDM process parameters

Trial 1

25.24
19.19

25.96
21.18
22.96
28.98
24.89
26.25
RWR

11.8
On the basis of the literature [14] and previous work done
[15, 16], it was concluded that the most important EDM
Average

4.29
5.22
4.03
7.97
8.73
7.14
9.48
10.46
process parameters which has greater influence on the

13
MRR, TWR, RWR, and SR are discharge current, pulse
on time, and pulse off time.
Trial 3

10.47
12.99
4.36
5.24
4.05
7.98
8.74
7.12
9.51
2.3 Determination of the working range of the process
parameters
Trial 2

4.65
5.65
4.57
8.61
9.06
TWR (mm3/min)

10.5
10.7
12.3
7.4

A large number of trials were conducted by varying one of

the process parameters and keeping the other parameters
Trial 1

constant. The working range of discharge current, pulse on

10.25
13.68
3.86
4.79
3.47
7.32
8.39

8.47
6.9

time, and pulse off time was explored by inspecting the

cavity produced in the workpiece by the electrode. The
working range of the process parameters selected under
Average

15.76
25.92
30.02
28.65
39.87
29.99
30.57
41.31
51.38

the present study is indicated in Table 3.

2.4 Selection of OA
Table 5 Performance measures of EDM

Trial 3

15.77
25.86

28.63
39.88
29.96
30.57
41.35
52.02
30

In this study, the number of process parameters considered

were three and the level of each parameter was 3. The
Trial 2

26.86
30.57
29.12

29.95
31.92

52.29
16.11

40.11
MRR (mm3/min)

degrees of freedom of all three parameters were 2 (i.e.,

41.4

number of levels 1) and the total degrees of freedom of all

the factors is 6 (i.e., 3×2=6). The selected orthogonal arrays
Trial 1

15.41
24.98
29.49

39.62
30.06
29.24
41.19
49.85

(OA) degrees of freedom (DOF) (i.e., number of experi-

28.2

ments, 1=9–1=8) must be greater than the total DOF of all

the factors (6). Hence, L9 (34) OA is considered for the
Sr. no

present study. Based on the preliminary experimentation,

1
2
3
4
5
6
7
8
9

there is no interaction between the selected process

 Int J Adv Manuf Technol

Fig. 1 Schematic
diagram of die sinking
EDM machine

parameters. Hence interaction is not considered for the weight loss (WWL) under a period of machining time in
present study. Three trails of each experiment were minutes, i.e.,
conducted to average of these values so that minimize the

 WWL ðgÞ  1; 000
pure experimental error. The selected OA is presented in MRR mm3 =min ¼
Table 4. ρðg=cm3 Þ  machining timeðminÞ
ρ=density of workpiece material
2.5 Conduct the experiments as per the selected OA The TWR is the tool weight loss (TWL) under a period of
machining time in minutes, i.e.,
The workpiece material of 60 mm in diameter and 8 mm
thick and the electrode of 12 mm diameter were used. The
 TWL ðgÞ  1; 000
TWR mm3 =min ¼
experiments were conducted as per the layout shown in ρðg=cm3 Þ  machining timeðminÞ
Table 5. A Formatics EDM 50 die sinking machine with
ρ=Density of tool material
Electronica PRS-20 controller was employed for conducting
The RWR is defined as the ratio of the TWR to the MRR
the EDM experiments. Each experiment was conducted for
from the workpiece and is usually expressed as a percent.
a duration of 3 min. Prior to machining, the work pieces and
electrode were cleaned and polished. The workpiece was
TWR
firmly clamped in the vice and immersed in the electrol RWR ð%Þ ¼  100
EDM oil. The positive polarity was used during the exper- MRR
iments. The schematic diagram of die sinking EDM ma- The SR is referred to the roughness or smoothness of a given
chine is shown in Fig. 1. surface. In this study, it was measured in terms of roughness
average (Ra), which is an arithmetic average of peaks and
2.6 Record the performance measures valleys of a workpiece surface measured from the centerline
(i.e., MRR, TWR, RWR, and SR) of evaluation length. It was measured by Zeiss (Make:Surcom
130A) surface roughness tester. The machining performance
The machining performance measures were evaluated by measures, i.e., MRR, TWR, RWR, and SR were evaluated for
MRR, TWR, RWR, and SR. The MRR is the workpiece all the conditions and presented in Table 5.

Table 6 Average experimental results of MRR, TWR, RWR, and SR and their corresponding S/N ratios

Sr. no. MRR (mm3/min) S/N ratio (dB) TWR (mm3/min) S/N ratio (dB) RWR S/N ratio (dB) SR (μm) S/N ratio (dB)

1 15.76 23.95 4.29 −12.64 27.33 −28.73 5.62 −14.99

2 25.92 28.27 5.22 −14.35 20.15 −26.08 6.6 −16.39
3 30.02 29.54 4.03 −12.1 13.37 −22.52 7.71 −17.74
4 28.65 29.14 7.97 −18.2 27.76 −28.87 6.48 −16.23
5 39.87 32.01 8.73 −18.82 21.88 −26.8 7 −16.9
6 29.99 29.53 7.14 −17.07 23.83 −27.54 7.32 −17.29
7 30.57 29.7 9.48 −19.53 30.9 −29.8 5.92 −15.44
8 41.31 32.32 10.46 −20.39 25.33 −28.07 6.96 −16.85
9 51.38 34.21 13 −22.28 25.38 28.09 8.15 −18.22
Int J Adv Manuf Technol

Table 7 S/N response table for MRR Table 8 S/N response table for TWR

Symbol EDM parameter Mean S/N ratio (dB) Symbol EDM parameter Mean S/N ratio (dB)

Level 1 Level 2 Level 3 Max-min Level 1 Level 2 Level 3 Max-min

A Discharge current 23.9 32.84 41.09 17.19 A Discharge current −13.4 −17.97 −20.97 7.7
B Pulse on time 24.99 35.7 37.13 12.14 B Pulse on time −16.74 −17.85 −18.22 1.12
C Pulse off time 29.02 35.32 33.49 6.3 C Pulse off time −16.7 −18.22 −16.82 1.52

The total mean S/N ratio=29.85 The total mean S/N ratio=−17.26

2.7 Find the optimum condition for performance measures On the other hand, lower-the-better quality characteristics
and identify the significant factors for TWR, RWR, and SR should be taken for obtaining
optimal EDM performance. The MSD for lower the better
In Taguchi method, the effects of machining parameters on quality characteristic can be expressed as:
performance measures are evaluated under optimal condi-
1 Xm
tion. It is used to determine appropriate combination of MSD ¼ TWR2i ð3Þ
machining parameters to maximize MRR and minimize m i¼1
TWR, RWR, and SR. The experimental results of MRR,
Where TWRi2 is the value of TWR for the ith test
TWR, RWR, and SR were further transformed into a signal-
to-noise ratio (S/N) ratio. The characteristic that higher 1 Xm

value represents better machining performance, such as MSD ¼ RWR2i ð4Þ

m i¼1
MRR, is called “higher the better”. The characteristics that
lower the value represents better machining performance, Where RWRi2 is the value of RWR for the ith test
such as TWR, RWR, and SR, is called “lower the better.”
Therefore, “higher the better” for the MRR, “lower the 1 Xm
MSD ¼ SR2i ð5Þ
better” for TWR, RWR, and SR were selected for obtaining m i¼1
machining performance. Taguchi method uses the S/N ratio
to measure the quality characteristic deviating from the Where SRi2 is the value of SR for the ith test.
desired value. The S/N ratio η is defined as The average experimental results of MRR, TWR, RWR,
and SR and their corresponding S/N ratios using Eqs. (1) to
η ¼ 10 logðMSDÞ ð1Þ (5) are presented in Table 6.
After calculation of S/N ratio, the effect of each machin-
where MSD is the mean square deviation for the output
ing parameter at different levels can be separated. The mean
characteristic.
S/N ratio for each machining parameter at each level was
To obtain optimal EDM performance, higher-the-better
calculated by averaging the S/N ratios for the experiments at
quality characteristic for material removal rate from work-
the same level for that particular parameter. Table 7 shows
piece must be taken. The MSD for higher-the-better quality
S/N response table for MRR and Fig. 2 shows the S/N
characteristic can be expressed as:
response graph for MRR.
1 Xm
1 For performance measures like TWR, RWR, and SR,
MSD ¼ ð2Þ greater S/N ratios were considered as they result in smaller
m i¼1 MRR2i
variance of the output about the targeted value. The exper-
Where m is the number of tests and MRRi is the value of imental results for TWR and the S/N response graph for
MRR and ith test. TWR are shown in Table 8 and Fig. 3, respectively. Table 9

-11 A1 A2 A3 B1 B2 B3 C1 C2 C3
34
Mean of S/N ratios

-13
Mean of S/N ratios

32
-15
30 -17
28 -19

26 -21
-23
24
A1 A2 A3 B1 B2 B3 C1 C2 C3 -25

Fig. 2 S/N graph for MRR Fig. 3 S/N graph for TWR
 Int J Adv Manuf Technol

Table 9 S/N response table for RWR Table 10 S/N response table for SR

Symbol EDM parameter Mean S/N ratio (dB) Symbol EDM parameter Mean S/N ratio (dB)

Level 1 Level 2 Level 3 Max–min Level 1 Level 2 Level 3 Max–min

A Discharge current −25.78 −27.74 −28.65 2.87 A Discharge current −16.38 −16.81 −16.84 0.47
B Pulse on time −29.13 −26.99 −26.05 3.08 B Pulse on time −15.56 −16.72 −17.75 2.19
C Pulse off time −28.12 −27.68 −26.37 1.74 C Pulse off time −16.38 −16.95 −16.7 0.57

The total mean S/N ratio=−27.22 The total mean S/N ratio=−16.71

presents S/N response table for RWR and Fig. 4 shows the 2.8 Conduct the confirmation test
S/N response graph for RWR. Table 10 indicates S/N re-
sponse table for SR and Fig. 5 shows the S/N response Optimum levels of design parameters were used for predic-
graph for SR. tion and confirmation of the performance measures im-
The optimization of process parameters using Taguchi provement. The estimated S/N ratio, b η using the optimal
method [17] permits evaluation of the effects of indi- level of the design parameters can be calculated as:
vidual parameters independent of the other parameters. X
o
The analysis of variance (ANOVA) is used to determine b
η ¼ ηm þ ðb
ηi  ηm Þ ð7Þ
which design parameters significantly affect the perfor- i¼1
mance measures [16]. In ANOVA, first total sum of
Where ηm is the total mean S/N ratio, b ηi is the mean S/N
squared deviations SST from total mean S/N ratio ηm
ratio at the optimal level, and o is the number of main design
can be calculated as
parameters that affect the quality characteristic.
X
n For validations of the optimum results, experiments were
SST ¼ ð ηi  ηm Þ 2 ð6Þ conducted as per the optimum conditions and machining
i¼1
performance measures were evaluated and the results are
Where n is the number of experiments in the orthogonal presented in Table 16. It is observed that, experimental
array and ηi is mean S/N ratio for ith experiment. values are closer to the optimum values.
ANOVA was applied to find out the significance of
main factors and the F test was used to determine the
process parameter significantly effect on the responses 3 Discussion
(MRR, TWR, RWR, and SR). Usually, the change of
the EDM process parameter has significant effect on the 3.1 Effect of parameters on EDM performance measures
response when F ratio is large. Table 11 shows the
results of ANOVA for MRR. Table 12 presents the Based on the S/N ratio and ANOVA analysis of the result,
results of ANOVA for TWR. Table 13 shows the results various conclusions are drawn. As shown in Table 7 and
of ANOVA for RWR. Table 14 shows the results of Fig. 2, factors at level A3 (discharge current, 20 A), B3 (pulse
ANOVA for SR. on time, 65 μs), and C2 (pulse off time, 36 μs) gives
ANOVA also provides an indication of which process maximum MRR. Factor C is having least significant effect
parameter combination is predicted and the optimum on improving MRR. The contribution order of machining
results. These optimum results are presented in parameters for MRR is discharge current, then pulse on
Table 15. time, and then pulse off time as shown in Table 11. The

-25
-25.5 A1 A2 A3 B1 B2 B3 C1 C2 C3 -15
A1 A2 A3 B1 B2 B3 C1 C2 C3
-26
-15.5
Mean of S/N ratios

-26.5
Mean S/N ratios

-27 -16
-27.5
-16.5
-28
-28.5 -17
-29
-17.5
-29.5
-30 -18

Fig. 4 S/N graph for RWR Fig. 5 S/N graph for SR

Int J Adv Manuf Technol

Table 11 Results of the ANOVA for MRR

Symbol EDM parameter DOF Sum of squares Mean of squares F ratio Contribution (%) P value

A Discharge current 2 35.53 17.77 3.3 52.4 0.011

B Pulse on time 2 23 11.5 1.54 33.92 0.029
C Pulse off time 2 7.1 3.5 0.35 10.47 0.072
Error 2 2.17 3.2
Total 8 67.8 100

heat energy supplied to remove the workpiece material is is of positive polarity. Thus, less amount of material is
controlled by the discharge current. Hence, the contribution removed from the electrode during pulse on time. Material
and significance of discharge current is largest. The pulse on is not removed from the electrode during the pulse off time.
time controls the duration of time for which the current is Hence, pulse on time and pulse off time have very less
allowed to flow per cycle. The material removed from the contribution and significance to the electrode wear.
workpiece is directly proportional to the amount of energy It is observed from Table 9 and Fig. 4 that the factors at
supplied during this period. Thus, it is the second factor as level A1 (discharge current, 10 A), B3 (pulse on time, 65 μs),
far as contribution and significance is concern. During the and C3 (pulse off time, 48 μs) can be used to minimize
pulse off time, no material is removed from the workpiece as RWR. Factors A and B are having significant effect for
there is no discharge current supplied. This results in lowest minimization of RWR, whereas factor C has least effect. It
significant effect and lowest contribution for MRR. is observed from Table 13 that the contribution of pulse on
From Table 8 and Fig. 3, it is recommended to use the time is more than the contribution of discharge current and
factors at level A1 (discharge current, 10 A), B1 (pulse on pulse off time for RWR. The pulse on time decides the time
time, 25 μs), and C1 (pulse off time, 24 μs) to minimize duration of heat energy supplied to both workpiece and
TWR. Table 12 shows that the contribution of factors B and electrode. The electrode is exposed to less heat than the
C is very less to minimize TWR. The amount of heat energy workpiece for the same time period when positive polarity
depending on the discharge current is also utilized to remove is used. This causes less amount of material removal from
material from the tool. This results in largest significance an electrode as compared to the workpiece. The discharge
and contribution of discharge current towards TWR. The current controls amount of heat energy required removing
electrode is exposed to less heat than the workpiece when it material from workpiece and electrode, hence its

Table 12 Results of the

ANOVA for TWR Symbol EDM parameter DOF Sum of squares Mean of squares F ratio Contribution (%) P value

A Discharge current 2 91.28 45.64 33.32 91.74 0.001

B Pulse on time 2 1.9 1 0.006 1.91 0.094
C Pulse off time 2 4.3 3.5 0.35 10.47 0.087
Error 2 2.02 3.2
Total 8 99.5 100

Table 13 Results of the ANOVA for RWR

Symbol EDM parameter DOF Sum of squares Mean of squares F ratio Contribution (%) P value

A Discharge current 2 12.93 6.46 1.64 35.41 0.026

B Pulse on time 2 14.98 7.49 2.09 41.03 0.020
C Pulse off time 2 4.93 2.47 0.47 13.5 0.065
Error 2 3.53 10.05
Total 8 36.51 100
 Int J Adv Manuf Technol

Table 14 Results of the ANOVA for SR

Symbol EDM parameter DOF Sum of squares Mean of squares F value Contribution (%) P value

A Discharge current 2 0.4 0.2 0.15 4.65 0.084

B Pulse on time 2 7.23 3.61 15.74 84.07 0.005
C Pulse off time 2 0.49 0.24 0.18 5.7 0.084
Error 2 0.48 5.58
Total 8 8.6 100

contribution and significance is more than pulse off time and fluid. This continuous rapid heating and followed by
less than pulse on time. The pulse off time shows least cooling resulted in a typical EDMed surface. When material
contribution and effect on RWR as no heat energy is sup- is removed from the workpiece and electrode, craters are
plied by the discharge current during this period. formed on their surfaces. During the flushing action, very
From Table 10 and Fig. 5, factor B (pulse on time) is the less amount of removed material is carried away. The
most significant factor among all factors where as factors A remaining material resolidifies to form an uneven surface.
(discharge current) and C (pulse off time) have insignificant Figure 6a–b indicates scanning electron microscope (SEM)
effect on SR. The Table 14 indicates that the contribution of for the optimum condition for MRR and SR respectively. It is
factor B is very high for SR as compared to factors A and C. observed that the size of craters, globules of debris, pock-
The optimum level of factors for SR is A1 (discharge current, marks or chimneys, and microcracks are larger in Fig. 6a due
10 A), B1 (pulse on time, 25 μs), and C1 (pulse off time, to higher discharge current (20 A) and longer pulse duration
24 μs). The time duration of heat energy available for material (65 μs). Thus, the surface is rougher as large amount of pulse
removal depends on pulse on time. This energy is shared by a energy is supplied for longer duration. Figure 6b shows
larger number of sparks results in reduction the size of the smoother surface due to lower discharge current (10 A) and
crater. This improves surface finish. Hence, contribution and lower pulse on time (25 μs). At the low discharge current and
significance of pulse on time is largest for SR. smaller pulse duration, low amount of pulse energy is supplied
It is interesting to note that optimal settings of parameters for smaller duration. No craters, globules of debris, pock-
for SR and TWR are same but differs from optimal settings marks or chimneys, and microcracks are observed. From
of parameters for MRR and RWR, and poses difficulty to Fig. 6b, the molten material is removed either by sheet forma-
achieve the target of all objectives. tion or ligament formation. Proper removal material leads to
smoother surface.
3.2 Morphology of EDMed surface

The EDMed surface micrographs are shown in Fig. 6. It 4 Conclusions

shows the overlapping craters, globules of debris, and pock-
marks or chimneys, formed by entrapped gases escaping In this study, the influence of the process parameters and
from the redeposited material. It is also observed that optimization of MDN 300 steel in the die sinking EDM was
microcracks are present. They are formed due to develop- studied by using Taguchi method. From the results, it was
ment of high thermal stresses produced and plastic defor- found that discharge current, pulse on time, and pulse off
mation of the material. In the EDM process, a large amount time have been found to play significant role in EDM
heat is generated during each spark. This heat is utilized to operations. Also, it was found that the optimal levels of
vaporize and melt the workpiece material as well as elec- the factors for SR and TWR are same but differs from the
trode material followed by rapid cooling by the dielectric optimum levels of the factors for MRR and RWR. From

Table 16 Validation of the optimum results

Table 15 Optimum values of the machining performance measures
Performance Optimum Optimum Experimental
Performance measure Optimum condition Optimum value measure condition value value

MRR (mm3/min) A3B3C2 48.18 MRR (mm3/min) A3B3C2 48.84 51.22

TWR (mm3/min) A1B1C1 3.44 TWR (mm3/min) A1B1C1 3.54 4.06
RWR A1B3C3 15.2 RWR A1B3C3 14.58 13.37
SR (Ra) (μm) A1B1C1 5.60 SR (Ra) (μm) A1B1C1 5.56 5.62
Int J Adv Manuf Technol

Fig. 6 SEM micrographs of Globules of debris

MDN 300 steel; a optimum
condition for MRR, b optimum Pockmark
condition for SR

Micro-cracks
Crater

a b

ANOVA, discharge current is more significant than pulse on Maraging steels—an experimental investigation. Proceedings of the
time for MRR and TWR; whereas pulse on time is more World Congress on Engineering. Vol. II, London, UK
6. Krishna Mohana Rao G, Rangajanardhaa G, Hanumantha Rao D,
significant than discharge current for RWR and SR. On the Sreenivasa Rao M (2009) Development of hybrid model and
other hand, pulse off time is less significant for all perfor- optimization of surface roughness in electric discharge machining
mance characteristics considered. using artificial neural networks and genetic algorithm. J Mater
Surface morphological study indicates that at higher dis- Process Technol 209:1512–1530
7. All Metals and Forge Group, USA. www.steelforge.com/
charge current and longer pulse on duration gives rougher
metaltidbits/maraging.html
surface characteristics with more craters, globules of debris, 8. Lin JL, Wang KS, Yan BH, Tarng YS (2000) Optimization of
and microcracks than that of lower discharge current and electrical discharge machining process based on the Taguchi meth-
lower pulse on duration. od with fuzzy logics. J Mater Process Technol 102:48–55
9. Taguchi G, Elsayed EA, Hsiang TC (1989) Quality engineering in
production systems. McGraw-Hill, New York
Acknowledgments The authors would like to thank the authorities 10. Ross PJ (1998) Taguchi techniques for quality engineering.
of National Institute of Technology (NIT), Warangal, India and Indian McGraw-Hill, New York, pp 24–98
Institute of Chemical Technology (IICT), Hyderabad, India for providing 11. Montgomary DC (1997) Design and analysis of experiments.
the facilities to carry out this work. Wiley, New York, pp 385–475
12. Phadke MS (1989) Quality engineering using robust design. Pren-
tice Hall, New Jersey
13. Mahapatra SS, Patnaik A (2007) Optimization of wire electrical
References discharge machining (WEDM) process parameters using Taguchi
method. Int J Adv Manuf Technol 34(9–10):911–925
14. Gao Q, Zhang Q-h, Shu-peng S, Zhan J-h (2008) Parameter
1. Abu Zaid OA (1997) On the effect of electro-discharge machining optimization model in electrical discharge machining process. J
parameters on the fatigue life of AISI D6 tool steel. J Mater Zhejiang Univ Sci A 9(1):104–108
Process Technol 68(1):27–32 15. A. M. Nikalje, A. Kumar, K. V. Sai Srinadh (2009) Modeling
2. Ho KH, Newman ST (2003) State of the art electrical discharge electrical discharge machining of Maraging steel using artificial
machining (EDM). Int J Mach Tool Manufac 43:1287–1300 neural network. Proceedings of Sixth International Conference on
3. Ekmekci B, Elkoca O, Erden A (2005) A comparative study on the Precision Meso. Micro Nano Engineering: G-44–48
surface integrity of plastic mould steel due to electric discharge 16. Nikalje AM, Kumar A, Sai Srinadh KV (2010) Optimization of
machining. Metall Mater Trans 36B:117–124 EDM parameters of Maraging steel (MDN300) using Taguchi
4. Puertas I, Luis CJ (2003) A study on machining parameters opti- method. Proceedings of the third international and twenty fourth
mization of electrical discharge machining. J Mater Process All India Manufacturing Technology. Des Res Conf 1:265–270
Technol 143–144:521–526 17. Wang WH, Tarng YS (1998) Design optimization of cutting pa-
5. Krishna Mohana Rao G, Satyanarayana S, Praveen M (2008) Influ- rameters for turning operations based on Taguchi method. J Mater
ence of machining parameters on electric discharge machining of Process Technol 84:122–129