137

CHAPTER 7

OPTIMIZATION OF MULTIPLE PERFORMANCE

CHARACTERISTICS WITH GREY RELATIONAL ANALYSIS FOR
SS 316

7.1 INTRODUCTION

The optimization of process parameters considering multiple

performance characteristics of the EDM process for SS 316 using the GRA is
presented. The GRA is then applied to examine how the factors influence the
quality targets of MRR, TWR, ROC and T. An optimal parametric combination
is then obtained. Through analyzing the Grey relational grade, the most
influential factors for individual quality targets of EDM operations can be
identified. Additionally, the ANOVA is also utilized to examine the most
significant factors for the EDM process as the MRR, TWR, ROC and T are
simultaneously considered. Also the analysis of micro-holes based on SEM
micrographs has also been presented.

7.2 EXPERIMENTAL RESULTS

The machined micro-hole of diameter 500 m using sinking EDM is

shown in Figure 7.1(a&b). The experimental runs of the L18 OA and the actual
values of parameters are presented in Table 7.1. Experimental Results of Entry
and Exit Radius of drilled micro holes are measured by optical microscope
 138

(OLYMPUS), which is presented in Table 7.2. Each experiment is repeated

thrice.

Figure 7.1 (a) SS 316 Workpiece with EDM drilled micro-holes

.Figure 7.1 (b) Macrograph of EDM drilled micro-hole of diameter 500 m

 139

Table 7.1 Experimental layout using an L18 orthogonal array

Levels of Parameters Actual values of Parameters
Expt.
No. A0 B0 D0
A0 B0 C0 D0 C0 s)
(A) (V) s)
1 1 1 1 1 10 30 100 20
2 1 2 2 2 10 40 150 30
3 1 3 3 3 10 50 200 40
4 2 1 1 2 12 30 100 30
5 2 2 2 3 12 40 150 40
6 2 3 3 1 12 50 200 20
7 3 1 2 1 14 30 150 20
8 3 2 3 2 14 40 200 30
9 3 3 1 3 14 50 100 40
10 1 1 3 3 10 30 200 40
11 1 2 1 1 10 40 100 20
12 1 3 2 2 10 50 150 30
13 2 1 2 3 12 30 150 40
14 2 2 3 1 12 40 200 20
15 2 3 1 2 12 50 100 30
16 3 1 3 2 14 30 200 30
17 3 2 1 3 14 40 100 40
18 3 3 2 1 14 50 150 20

The experimental results of machining time and tool wear are

presented in Table 7.3. To obtain a more accurate result, each combination of
experiments is repeated three times and the average is taken up. The
experimental performance responses of MRR and TWR are shown in Table 7.4.
Experimental performance Responses of ROC and T are shown in Table 7.5.
Experimental results of MRR, TWR, ROC and T are shown in Table 7.6.
 140

Table 7.2 Experimental Results of Hole entry and exit Radius

Expt. Hole entry radius ( m) Hole exit radius

No. Trial 1 Trial 2 Trial 3 Trial 1 Trial 2 Trial 3

1 285.4 294.5 290.2 242.5 251.4 248.1

2 296.9 289.8 292.4 249.9 239.5 241.7

3 299.3 282.6 298.4 263.4 252.4 261.5

4 312.1 354.4 325.6 236.5 284.4 265.6

5 325.7 359.2 360.4 248.4 247.0 252.0

6 328.1 366.4 352.4 230.5 247.0 243.0

7 316.2 325.7 320.4 275.4 245.5 260.0

8 366.4 325.7 360.0 292.6 281.4 240.3

9 320.9 332.9 328.6 248.4 247.0 238.2

10 311.3 306.5 309.7 273.9 275.4 271.4

11 325.7 313.7 320.7 252.9 266.4 259.2

12 318.5 335.3 332.3 281.4 276.9 269.8

13 324.8 317.3 319.1 252.9 252.9 251.6

14 318.4 330.8 326.5 272.4 264.9 275.3

15 312.8 304.4 315.3 303.8 260.4 240.3

16 308.9 321.8 311.7 285.9 276.9 269.8

17 318.8 327.8 325.3 273.9 272.4 255.6

18 296.3 299.3 295.1 267.9 266.4 262.6

 141

Table 7.3 Experimental Results of machining time and tool wear

Expt. Machining time (sec) Tool wear (mm)

No. Trial 1 Trial 2 Trial 3 Trial 1 Trial 2 Trial 3

1 44.48 34.50 48.52 6 6 6

2 34.01 40.10 38.82 2 4 4

3 42.10 46.03 46.62 6 5 6

4 31.63 26.13 28.75 9 4 5

5 20.16 21.87 22.42 6 8 8

6 20.79 26.15 23.65 5 4 4

7 24.15 23.61 28.82 7 7 6

8 18.16 20.30 20.60 6 10 9

9 30.53 27.96 29.65 8 7 8

10 39.82 33.48 38.89 6 7 6

11 36.05 34.34 35.58 5 5 6

12 42.26 41.36 42.65 6 3 5

13 43.48 53.16 46.85 5 7 6

14 48.54 39.81 44.86 7 6 7

15 35.29 49.16 46.62 5 4 5

16 26.17 23.12 24.48 7 8 8

17 27.13 27.14 28.86 6 6 7

18 18.64 16.05 16.98 5 10 8

 142

Table 7.4 Experimental performance Responses of MRR and TWR

Expt. MRR (g/min) TWR (g/min)

No. Trial 1 Trial 2 Trial 3 Trial 1 Trial 2 Trial 3

1 0.01421 0.01959 0.01354 0.01383 0.01783 0.01268

2 0.01994 0.01586 0.01668 0.00603 0.01023 0.01057

3 0.01704 0.01409 0.01524 0.01461 0.01114 0.01320

4 0.02167 0.03548 0.02760 0.02918 0.01570 0.01783

5 0.03722 0.03846 0.03825 0.03052 0.03751 0.03659

6 0.03431 0.03297 0.03430 0.02466 0.01568 0.01734

7 0.03285 0.03148 0.02655 0.02972 0.03040 0.02135

8 0.05434 0.04116 0.04011 0.03388 0.05051 0.04480

9 0.02415 0.02742 0.02473 0.02687 0.02567 0.02767

10 0.01949 0.02291 0.01968 0.01545 0.02144 0.01582

11 0.02112 0.02222 0.02147 0.01422 0.01493 0.01729

12 0.01929 0.02057 0.01930 0.01456 0.00744 0.01202

13 0.01746 0.01390 0.01580 0.01179 0.01350 0.01313

14 0.01630 0.02025 0.01831 0.01479 0.01545 0.01600

15 0.02438 0.01471 0.01507 0.01453 0.00834 0.01100

16 0.03061 0.03515 0.03131 0.02743 0.03548 0.03351

17 0.02936 0.03012 0.02659 0.02268 0.02267 0.02487

18 0.03868 0.04517 0.04150 0.02750 0.06389 0.04831

 143

Table 7.5 Experimental performance Responses of ROC and TAPER

Expt. ROC (µm) T

No. Trial 1 Trial 2 Trial 3 Trial 1 Trial 2 Trial 3

1 70.80 89.00 80.40 0.00715 0.00718 0.00702

2 93.80 79.60 84.80 0.00783 0.00838 0.00845

3 98.60 65.20 96.80 0.00598 0.00503 0.00615

4 124.20 208.80 151.20 0.01260 0.01167 0.01000

5 151.40 218.40 220.80 0.01288 0.01870 0.01807

6 156.20 232.80 204.80 0.01627 0.01990 0.01823

7 132.40 151.40 140.80 0.00680 0.01337 0.01007

8 232.80 151.40 220.00 0.01230 0.00738 0.01995

9 141.80 165.80 157.20 0.01208 0.01432 0.01507

10 122.60 113.00 119.40 0.00623 0.00518 0.00638

11 151.40 127.40 141.40 0.01213 0.00788 0.01025

12 137.00 170.60 164.60 0.00618 0.00973 0.01042

13 149.60 134.60 138.20 0.01198 0.01073 0.01125

14 136.80 161.60 153.00 0.00767 0.01098 0.00853

15 125.60 108.80 130.60 0.00150 0.00733 0.01250

16 117.80 143.60 123.40 0.00383 0.00748 0.00698

17 137.60 155.60 150.60 0.00748 0.00923 0.01162

18 92.60 98.60 90.20 0.00473 0.00548 0.00542

 144

Table 7.6 Experimental results of MRR, TWR, ROC and T

Average values of all responses

Expt.
No. MRR ROC
TWR (g/min) T
(g/min) (µm)

1 0.0158 0.0147 80.07 0.0071

2 0.0175 0.0089 86.07 0.0082

3 0.0155 0.0129 86.87 0.0057

4 0.0283 0.0209 161.40 0.0114

5 0.0379 0.0348 196.87 0.0166

6 0.0339 0.0192 197.93 0.0181

7 0.0303 0.0271 141.53 0.0101

8 0.0452 0.0430 201.40 0.0132

9 0.0254 0.0267 154.93 0.0138

10 0.0207 0.0175 118.33 0.0059

11 0.0216 0.0154 140.07 0.0101

12 0.0197 0.0113 157.40 0.0088

13 0.0157 0.0128 140.80 0.0113

14 0.0183 0.0154 150.47 0.0091

15 0.0181 0.0112 121.67 0.0071

16 0.0324 0.0321 128.27 0.0061

17 0.0287 0.0234 147.93 0.0094

18 0.0418 0.0465 93.80 0.0052

 145

7.3 BEST EXPERIMENTAL RUN

The OA with the GRA is used to determine the optimal machining

parameters with consideration of the multiple performance characteristics.
Basically, the MRR belongs to the “larger-the-better” methodology that is
Equation (6.1) which is employed for data preprocessing whereas smaller values
of the TWR, ROC and T are desirable and hence they belong to “smaller-the-
better” methodology that is Equation (6.2) which is employed for data
preprocessing. The values of the MRR, the TWR, the ROC and T are set to be
the reference sequence xi* (k ) , k = 1–4. Moreover, the results of nine experiments

are the comparability sequences xi (k ) , i=1, 2. . . 18, k = 1–4.

Table 7.7 lists all of the sequences after implementing the data
preprocessing using Equations 6.1 & 6.2. The reference and the comparability
sequences are denoted as x *i (k ) and x i (k ) respectively. Also, the deviation
sequences 0i (k ) , max (k ) and min ( k ) for i = 1–18, k = 1–4 can be calculated.

The distinguishing coefficient can be substituted for the Grey relational

coefficient in Equation (6.4). If all the process parameters have equal weightage,
is set to be 0.5. Table 7.8 lists the Grey relational coefficients and the grade for
all nine comparability sequences.
 146

Table 7.7 The data preprocessing of the each individual quality

characteristic

ROC
Expt. No. MRR (g/min) TWR (g/min) T
(µm)

Reference
1.0000 1.0000 1.0000 1.0000
sequence

1 0.0109 0.8448 1.0000 0.8525

2 0.0685 1.0000 0.9505 0.7670

3 0.0000 0.8926 0.9440 0.9604

4 0.4301 0.6821 0.3297 0.5193

5 0.7570 0.3108 0.0374 0.1225

6 0.6187 0.7266 0.0286 0.0000

7 0.4987 0.5159 0.4934 0.6234

8 1.0000 0.0931 0.0000 0.3809

9 0.3354 0.5271 0.3830 0.3336

10 0.1760 0.7707 0.6846 0.9441

11 0.2067 0.8262 0.5055 0.6225

12 0.1434 0.9363 0.3626 0.7240

13 0.0089 0.8972 0.4995 0.5271

14 0.0953 0.8280 0.4198 0.7021

15 0.0874 0.9376 0.6571 0.8530

16 0.5681 0.3835 0.6027 0.9312

17 0.4448 0.6156 0.4407 0.6724

18 0.8850 0.0000 0.8868 1.0000

 147

Table 7.8 The grey relational coefficient and grey relational grade of the
each individual quality characteristic

ROC Grey
Expt.
MRR (g/min) TWR (g/min) T Relational
No. (µm) Grade

1 0.3358 0.7631 1.0000 0.7722 0.7178

2 0.3493 1.0000 0.9100 0.6821 0.7353

3 0.3333 0.8232 0.8992 0.9267 0.7456

4 0.4673 0.6113 0.4272 0.5099 0.5039

5 0.6730 0.4205 0.3418 0.3630 0.4496

6 0.5673 0.6465 0.3398 0.3333 0.4717

7 0.4993 0.5081 0.4967 0.5704 0.5186

8 1.0000 0.3554 0.3333 0.4468 0.5339

9 0.4293 0.5139 0.4476 0.4287 0.4549

10 0.3776 0.6856 0.6132 0.8995 0.6440

11 0.3866 0.7420 0.5028 0.5698 0.5503

12 0.3686 0.8870 0.4396 0.6443 0.5849

13 0.3353 0.8295 0.4997 0.5139 0.5446

14 0.3559 0.7440 0.4629 0.6266 0.5474

15 0.3540 0.8890 0.5932 0.7728 0.6522

16 0.5365 0.4478 0.5573 0.8791 0.6052

17 0.4739 0.5653 0.4720 0.6042 0.5288

18 0.8130 0.3333 0.8154 1.0000 0.7404

 148

This investigation employs the response table of the Taguchi method

to calculate the average Grey relational grades for each factor level, as illustrated
in Table 7.9. Since the Grey relational grades represent the level of correlation
between the reference and the comparability sequences, the larger Grey relational
grade means the comparability sequence exhibiting a stronger correlation with
the reference sequence. Based on this study, a combination of the levels can be
selected so that it can provide the largest average response. In Table 7.9, the
combination of A01, B03, C02, and D02 shows the largest value of the Grey
relational grade for the factors A0, B0, C0, and D0, respectively. Therefore, it is
observed that A01B03C02D02 with current of 10 A, voltage of 50 V, pulse-on time
of 150 µs, and pulse-off time of 30 µs is the optimal parametric combination of
the EDM operations.

Table 7.9 Response table for the grey relational grade

Machining Gray Relational Grade Main
Symbol
Parameter level 1 level 2 level 3 Effect
A0 Current (A) 0.6630 0.5282 0.5636 0.1347
B0 Voltage (V) 0.5890 0.5576 0.6083 0.0507
C0 Pulse-on time ( s) 0.5680 0.5956 0.5913 0.0276
D0 Pulse-off time ( s) 0.5910 0.6026 0.5612 0.0413

7.4 MOST INFLUENTIAL FACTOR

In this study, the GRA is applied to examine how the EDM parameters
influence the quality targets of workpiece. The values of the factor level in
eighteen experimental runs are set to the comparability sequences for four
controllable factors. Table 7.10 lists all of the sequences.
 149

Table 7.10 The reference sequences and comparability sequences for MRR,
TWR, ROC, and T results and experimental factor levels

Expt. Comparability sequences Reference sequences

No. A0 B0 C0 D0 MRR TWR ROC T

1 10 30 100 20 0.0158 0.0148 80.07 0.0071

2 10 40 150 30 0.0175 0.0089 86.07 0.0082

3 10 50 200 40 0.0155 0.0130 86.87 0.0057

4 12 30 100 30 0.0283 0.0209 161.40 0.0114

5 12 40 150 40 0.0380 0.0349 196.87 0.0166

6 12 50 200 20 0.0339 0.0192 197.93 0.0181

7 14 30 150 20 0.0303 0.0272 141.53 0.0101

8 14 40 200 30 0.0452 0.0431 201.40 0.0132

9 14 50 100 40 0.0254 0.0267 154.93 0.0138

10 10 30 200 40 0.0207 0.0176 118.33 0.0059

11 10 40 100 20 0.0216 0.0155 140.07 0.0101

12 10 50 150 30 0.0197 0.0113 157.40 0.0088

13 12 30 150 40 0.0157 0.0128 140.80 0.0113

14 12 40 200 20 0.0183 0.0154 150.47 0.0091

15 12 50 100 30 0.0181 0.0113 121.67 0.0071

16 14 30 200 30 0.0324 0.0321 128.27 0.0061

17 14 40 100 40 0.0287 0.0234 147.93 0.0094

18 14 50 150 20 0.0418 0.0466 93.80 0.0052

 150

Table 7.11 The sequences after data preprocessing for the reference
sequences and comparability sequences

Expt. Comparability sequences Reference sequences

No.
A0 B0 C0 D0 MRR TWR ROC T

1 1 1.00 1 1 1.0000 1.0000 1.0000 1.0000

2 1 1.33 1.5 1.5 1.1086 0.6049 1.0749 1.1553

3 1 1.67 2 2 0.9795 0.8783 1.0849 0.8041

4 1.2 1.00 1 1.5 1.7904 1.4141 2.0158 1.6050

5 1.2 1.33 1.5 2 2.4066 2.3591 2.4588 2.3255

6 1.2 1.67 2 1 2.1459 1.3009 2.4721 2.5480

7 1.4 1.00 1.5 1 1.9196 1.8372 1.7677 1.4161

8 1.4 1.33 2 1.5 2.8647 2.9133 2.5154 1.8564

9 1.4 1.67 1 2 1.6117 1.8087 1.9351 1.9422

10 1 1.00 2 2 1.3113 1.1886 1.4779 0.8337

11 1 1.33 1 1 1.3693 1.0473 1.7494 1.4176

12 1 1.67 1.5 1.5 1.2499 0.7671 1.9659 1.2334

13 1.2 1.00 1.5 2 0.9963 0.8665 1.7585 1.5909

14 1.2 1.33 2 1 1.1591 1.0428 1.8793 1.2732

15 1.2 1.67 1 1.5 1.1443 0.7638 1.5196 0.9992

16 1.4 1.00 2 1.5 2.0505 2.1743 1.6020 0.8571

17 1.4 1.33 1 2 1.8181 1.5834 1.8476 1.3271

18 1.4 1.67 1.5 1 2.6479 3.1504 1.1715 0.7322

 151

Table 7.11 presents the normalized results after data preprocessing.

Subsequently, the deviation sequences are calculated using the method
mentioned above. The deviation sequences and the distinguishing coefficient
then are substituted in Equation (6.4) to obtain the Grey relational coefficients.
Additionally, the Grey relational coefficients are averaged using an equal
weightage to obtain the Grey relational grade. Table 7.12 presents the Grey
relational coefficients and the grade of the MRR of the reference sequence and
comparability sequences.

Table 7.13 gives the Grey relational coefficients and the grade of the
TWR for the reference sequence and the comparability sequences. Table 7.14
shows the Grey relational coefficients and the grade of the ROC for the reference
sequence and the comparability sequences. Similarly, Table 7.15 shows the Grey
relational coefficients and the grade of the T for the reference sequence and the
comparability sequences.

The Grey relational grades in Tables 7.12–7.15 can be further

arranged in Table 7.16. By comparing Row 1, Row 2, Row 3 and Row 4, some
conclusion can be drawn. In the first row, it is found that (MRR, A0) > (MRR,
B0) > (MRR, D0) > (MRR, C0) which means that the order of importance of
the controllable factors for MRR in sequence is the factor A0, B0, D0 and C0. The
second row (TWR, A0) > (TWR, D0) > (TWR, B0) > (TWR, C0) means the
order of importance of the controllable factors for TWR in sequence is the factor
A0, D0, B0, and C0. The third row (ROC, D0) > (ROC, A0) > (ROC, B0) >
(ROC, C0) means the order of importance of the controllable factors for ROC in
sequence is the factor D0, A0, B0, and C0. Similarly, the fourth row (T, A0) >
(T, D0) > (T, B0) > (T, C0) means the order of importance of the controllable
factors for T in sequence is the factor A0, D, 0 B0, and C0.
 152

Table 7.12 The calculated grey relational coefficient and grey relational
grade for experimental factors to experimental result of the MRR

A0 B0 C0 D0

Grey relational coefficient

1.0000 1.0000 1.0000 1.0000

0.8709 0.7731 0.5945 0.6779

0.9728 0.5270 0.3600 0.4467

0.5537 0.4921 0.4207 0.7394

0.3777 0.4164 0.3877 0.6696

0.4364 0.6151 0.7973 0.4183

0.5850 0.4543 0.5777 0.4726

0.3333 0.3333 0.3989 0.3765

0.7757 0.9330 0.4841 0.6797

0.7017 0.7110 0.4545 0.5447

0.6648 0.9552 0.6085 0.6905

0.7456 0.6475 0.6965 0.7671

0.7824 0.9952 0.5326 0.4508

0.9472 0.8147 0.4057 0.8381

0.9293 0.5945 0.7991 0.6985

0.5296 0.4216 0.9191 0.5995

0.6366 0.6123 0.4123 0.8192

0.3698 0.4383 0.3333 0.3333

Grey relational grade

0.6785 0.6519 0.5657 0.6235

 153

Table 7.13 The calculated grey relational coefficient and grey relational
grade for experimental factors to experimental result of the TWR

A0 B0 C0 D0

Grey relational coefficient

1.0000 1.0000 1.0000 1.0000

0.6890 0.5203 0.4797 0.5457

0.8779 0.5005 0.4239 0.4894

0.8035 0.6561 0.6659 0.9260

0.4302 0.4351 0.4899 0.7496

0.8966 0.6835 0.5414 0.7813

0.6669 0.4855 0.7099 0.5622

0.3664 0.3333 0.4747 0.4321

0.6817 0.8476 0.5050 0.8490

0.8227 0.8073 0.5042 0.5699

0.9487 0.7342 0.9458 0.9578

0.7898 0.4676 0.5296 0.5947

0.7241 0.8555 0.5657 0.4868

0.8477 0.7311 0.4630 0.9617

0.6674 0.4667 0.7774 0.5936

0.5306 0.4022 0.8256 0.6146

0.8267 0.7595 0.5858 0.7208

0.3333 0.3474 0.3333 0.3333

Grey relational grade

0.7168 0.6130 0.6012 0.6760

 154

r the reference sequences and comparability sequences

Table 7.14 The calculated grey relational coefficient and grey relational
grade for experimental factors to experimental result of the ROC

A0 B0 C0 D0
Grey relational coefficient
1.0000 1.0000 1.0000 1.0000
0.8946 0.6958 0.5444 0.6339
0.8822 0.5040 0.3569 0.4458
0.4381 0.3678 0.3333 0.5880
0.3357 0.3443 0.3463 0.6160
0.3333 0.4232 0.5183 0.3333
0.6337 0.4350 0.6549 0.4895
0.3632 0.3333 0.4963 0.4203
0.5431 0.6877 0.3520 0.9189
0.5710 0.5529 0.4931 0.5850
0.4591 0.5869 0.4040 0.4955
0.3971 0.6639 0.5216 0.6124
0.5324 0.4379 0.6627 0.7530
0.4836 0.5198 0.8079 0.4557
0.6656 0.8007 0.4943 0.9741
0.7590 0.4954 0.5607 0.8783
0.5869 0.5347 0.3747 0.8285
0.7357 0.5441 0.6073 0.8110
Grey relational grade
0.5897 0.5515 0.5294 0.6577
 155

Table 7.15 The calculated grey relational coefficient and grey relational
grade for experimental factors to experimental result of the T

A0 B0 C0 D0

Grey relational coefficient

1.0000 1.0000 1.0000 1.0000

0.8127 0.7360 0.6344 0.6919

0.7748 0.3651 0.3333 0.3929

0.6247 0.4506 0.4971 0.8805

0.3745 0.3333 0.4201 0.7039

0.3333 0.3602 0.5218 0.3333

0.9767 0.5439 0.8769 0.6504

0.5963 0.4868 0.8063 0.6847

0.5542 0.6429 0.3882 0.9305

0.8021 0.7490 0.3389 0.3989

0.6174 0.8547 0.5888 0.6495

0.7428 0.5338 0.6916 0.7438

0.6329 0.4564 0.8680 0.6542

0.9020 0.8919 0.4514 0.7391

0.7705 0.4264 0.9987 0.6072

0.5539 0.7764 0.3435 0.5463

0.9024 0.9876 0.6464 0.5349

0.5023 0.3468 0.4378 0.7430

Grey relational grade

0.6930 0.6079 0.6024 0.6603

 156

The most influential factors that affect the output variables are
determined by identifying the maximum values in each row. Hence, based on the
maximum values in the matrix of the Grey relational ( (MRR, A0), (TWR, A0),
(ROC, D) 0, (T, A0) = (0.6785, 0.7168, 0.6577, 0.6930) it can be found that
the factor A0, the voltage, has the most influence on the MRR, TWR and T with
the values of 0.6785, 0.7168 and 0.6930 simultaneously. The factor D0, the
Pulse-off time, is more influential factor for ROC with value of 0.6577.

Table 7.16 The calculated grey relational grade

A0 B0 C0 D0

MRR 0.6785 0.6519 0.5657 0.6235

TWR 0.7168 0.6130 0.6012 0.6760

ROC 0.5897 0.5515 0.5294 0.6577

T 0.6930 0.6079 0.6024 0.6603

Additionally, Table 7.17 gives the results of the ANOVA for the
MRR, TWR, ROC and T using the calculated values from the Grey relational
grade of Table 7.8 and the response table of Table 7.9. According to Table 7.17,
the factor A, the current with the contribution of 34.92% is the most significant
controlled parameter for the EDM operation followed by the voltage with the
contribution of 4.69%, the pulse-off time with the contribution of 3.26% and the
pulse-on time with the contribution of 1.58% with maximization of the MRR,
and minimization of the TWR, ROC, and T are considering simultaneously.
 157

Table 7.17 ANOVA results of grey relational grade

Degree Contribution
Machining Sum of Mean
of F value
parameter squares squares (%)
freedom

Current
2 0.05855 0.029275 2.828101 34.92
(A0)

Voltage
2 0.007873 0.003936 0.380276 4.69
(B0)

Pulse-on
2 0.002642 0.001321 0.127624 1.58
time (C0)

Pulse-off
2 0.005458 0.002729 0.263657 3.26
time (D0)

Error (E0) 9 0.0932 0.010351 55.56

Total 17 0.167687 0.009864 100

7.5 CONFIRMATION TEST

After identifying the most influential parameters, the final phase is to

verify the MRR, TWR, ROC, and T by conducting the confirmation experiments.
The A01B03C02D02 is an optimal parameter combination of the EDM process via
the GRA. Therefore, the condition A01B03C02D02 of the optimal parameter
combination of the EDM process is considered as a confirmation test. If the
optimal setting with the current of 10 A, voltage of 50 V, pulse-on time of 150 µs
and pulse-off time of 30 µs is used, the final workpiece gives the MRR of
0.0197g/min, TWR of 0.0113 g/min, ROC of 157.40 µm and T of 0.0088 which
are the results of experiment number 12 in Table 7.2.
 158

7.6 ANALYSIS BASED ON SEM MICROGRAPHS

The effect of parameters on micro-hole shape and thickness of the

recast layer which is formed during EDM can clearly be observed from the SEM
micrographs of micro-holes machined under different parametric combinations.
After analyzing all micrographs, the effect of each parameter has been identified
individually. Figure 7.2 shows the micrographs of machined micro-holes under
parametric combinations of a) 10A/ 30V/100 s/ 20 s, b) 12A/ 30V/100 s/ 30
s, c) 14A/50V/100 s/ 40 s. The figures clearly show the effect of peak current
at constant pulse on-time of 100 s. It shows that the white layer which is formed
on the wall of machined micro-holes gets increased as current is increased. Also,
it is seen from Figure 7.2 that the shape of the hole is better with low current as
compared to high current. Hence, for further investigations on the effect of pulse-
on time, current has been set to 10 A.

Figure 7.3 shows the micrographs of machined micro-holes under

parametric combinations as (a) 10 A/ 30V/100 s/ 20 s, (b) 10 A/50V/150 s/
30 s and (c) 10 A/30V/200 s/ 40 s. Since voltage and pulse-off time do not
have stronger effect on machining criteria, it can be concluded that with lower
pulse-on time (i.e., 100 s), the shape of micro-hole is better as shown in Figure
7.3(a) and as pulse-on time is increased, the shape of micro-holes gets
deteriorated and as shown in Figures 7.3(b) and 7.3(c), respectively. For higher
pulse-on time, the exit portion of the micro-hole is not uniformly machined. As
pulse-on time is increased, the tool wear at the front tip of the tool also gets
increased rapidly resulting eventually in a non-uniform or round shape at the
front of the tool and causes uneven machining at the exit or bottom portion of the
micro-hole.
 159

The enlarged views of the wall of the micro-holes have also been
shown in the Figure 7.4, which clearly depicts the white layer thickness at
different pulse-on time. The micrographs show that as pulse-on time increases,
thickness of the white layer gets increased considerably. As pulse-on time is
increased, the period of active machining time increases per cycle and the molten
material gets resolidified because of incomplete flushing. From the figures, it is
found that craters are produced at the recast layer and increases with both peak
current and pulse-on time. Also the surfaces of the micro-holes walls become
rougher with the increase in both current and pulse-on time.

Figure 7.4 shows SEM micrographs of internal surface features at

different machining conditions (a), (b), and (c), respectively, as mentioned in
Figure 7.3. It is observed from the figures that the oxides and carbon particles are
more in Figure 7.4 (b) and 7.4 (c) compared to Figure 7.4 (a), which corroborates
the fact that with this (10 A/ 30V/100 s/ 20 s) machining parameter, formation
of white layer is minimum.

Figure 7.2 SEM micrographs of micro-hole and recast layer at different

machining conditions. a) 10 A/ 30V/100 s/ 20 s, b) 12 A/ 30V/100 s/ 30 s,
c) 14 A/50V/100 s/ 40 s.
 160

Figure 7.3 SEM micrographs of micro-hole and recast layer at different

machining conditions. (a) 10 A/ 30V/100 s/ 20 s, (b) 10 A/50V/150 s/ 30 s
and (c) 10 A/30V/200 s/ 40 s.

Figure 7.4 SEM micrographs of internal features of micro-hole at different

machining conditions. (a) 10 A/ 30V/100 s/ 20 s, (b) 10 A/50V/150 s/ 30 s
and (c) 10 A/30V/200 s/ 40 s.

7.7 CONCLUDING REMARKS

The GRA based on an OA of the Taguchi method is a way of

optimizing the EDM operations. The following outcomes are observed during
micro-hole generation on SS 316.
 161

It is observed from the experimental results that the current of 10A, the
voltage of 50V, the pulse-on time of 200 µs, and the pulse-off time of 40
µs are the optimized values for the maximization of the MRR and
minimization of the TWR, ROC, and T which are simultaneously
considered. The most significant factor for the MRR, TWR and T is
observed to be current.

Based on the ANOVA, the sequence of the percentage of contribution to

the EDM process, in the order of important factor is the current, the
voltage, the pulse-off time and the pulse-on time. Hence, the current is the
most significant controlled factor for the EDM operation when the
maximization of MRR and minimizing TWR, the ROC and the T which
are simultaneously considered.

From the observation of SEM micrographs, it is found that at lower range

of current and pulse-on time, the topographical condition of micro-holes is
better and as current and pulse-on time are increased, the micro-hole
condition gets deteriorated. The thickness of the white layer formation on
the machined micro-holes gets increased sharply with increase in current
and pulse-on time.