Influence of Process Parameters on Wire EDM Process for AISI 316 Stainless Steel

Influence of Process Parameters on Wire EDM Process for AISI 316 Stainless Steel

1 K. R. Padmavathi, 2 S. Devaraj, 3 I. John Solomon, 4 A. Premkumar

1,2Asso. Prof., Dept. of Mechanical Engg., Panimalar Engg. College, Chennai-600123, India.

Abstract The optimization of wire electrical discharge machining (WEDM) parameters for AISI 316 stainless steel using Taguchis robust design approach is proposed. Experimentation is planned as per Taguchis L27 orthogonal array. Each experiment has to be performed under different process parameters of pulse on time, pulse off time, pulse peak current, wire tension and wire feed rate. Two responses namely material removal rate and surface roughness (Ra) were considered for each experiment. The optimum machining parameter combination is obtained by using the analysis of signal-to-noise (S/N) ratio. The level of importance of the machining parameters on the material removal rate and surface roughness is determined by using analysis of variance (ANOVA).

Keywords: WEDM, ANOVA, MRR, Surface roughness, Signal to noise ratio

1. INTRODUCTION

   As newer and more exotic materials have been developed in the past few decades, conventional machining operations tend to reach their limitations as relatively more complicated shaped jobs are required to be manufactured. The increased use of wire-cut electrical discharge machining (WEDM) in manufacturing has thus kept growing at a highly accelerated rate since its first industrial application more than 30 years ago. Its broad capabilities have allowed it to encompass production in aerospace and automotive industries and virtually all areas of conductive material machining. This is because wire EDM provides the best alternative, or sometimes the only alternative, for machining conductive, exotic and HSTR (high strength and temperature resistive) material with the scope of generating intricate shapes and profiles. Also, WEDM has proved to have incredible prospective in its applicability in the present day metal cutting industry for achieving a significant dimensional accuracy, surface finish and contour generation trait of products or parts. Wire EDM is a special form of the traditional EDM process in which the electrode is a continuously moving conductive wire. Material is worn from the work piece by a series of distinct sparks between the work piece and the wire electrode (tool) estranged by a thin film of dielectric fluid (deionised water) which is continuously force fed to the machining zone to flush away the eroded particles. The progress of the wire is proscribed numerically to attain the preferred three-dimensional shape and accuracy for the work piece. Although the average cutting speed, relative machining costs, accuracy and surface finish have been improved many times since the commercial inception of the machine, much more

   improvement is still required to meet the increasing demand of precision and accuracy by different industries.

2. LITERATURE REVIEW

   To improve the performance namely the material removal rate, cutting speed, dimensional accuracy and surface roughness of the WEDM process, several researches have attempted previously. Ramakrishnan et al. [7] used Multi Response optimization procedure for studying WEDM operations using Taguchis robust design of experiments. Experiments were planned as per Taguchis L16 Orthogonal array. Each experiment has been performed under different cutting conditions of pulse on time; wire tension, delay time, and wire feed speed and ignition current intensity. Three responses namely material removal rate, surface roughness and wire wear ratio (WWR) had been considered for each experiment. Liao et al [3] carried out the study on the machining parameters optimization of WEDM in SKD 11 alloy steel. They used 0.25mm diameter brass wire as electrode. According to Taguchi quality design concept L18 orthogonal arrays table was chosen for conducting experiments. A total of six machining parameters like pulse on time, pulse off time, table feed, wire tension, wire speed and flushing pressure were chosen for the controlling factors and each parameter have three levels and the designed machining conditions are evaluated in terms of the measured machining performances like gap width, material exclusion rate, surface coarseness, discharge frequency, gap voltage, normal discharge frequency ratio. Chiang et al [1] used grey relational analysis for their study on aluminium oxide particle reinforced material with multiple performance characteristics.

   Ho et al [2] reviewed the WEDM research concerning the

   optimization of the process parameters survey the influence of the various factors distressing the machining performance and yield. Mahabatra et al.[4] optimized the WEDM process parameters using Taguchi method. In this study the relationship between control factors like discharge current, pulse duration, pulse frequency, wire speed, wire tension and dielectric flow and responses like material removal rate, surface roughness and Kerf were established by means of non linear regression analysis. Shajan Kuriakose et al. [11] conducted multi objective optimization of WEDM process by Non Dominated Sorting Genetic Algorithm technique. Ozdemir et al. [5] performed exploration on machinability of nodular cast iron (GGG 40) by WEDM. Variation of the surface roughness and cutting rate with machining parameters were scientifically modeled by using the regression analysis method. Puri et al [6] made an attempt to

   model the white layer depth through response surface methodology. Tarang et al. [10] used feed forward neural network approach to determine the optimal cutting parameters while machining SUS-304 stainless steel in WEDM. Speeding et al [10] performed the modeling process through response surface methodology and artificial neural networks. A response surface model based on a central composite rotatable experimental design, and 4-16-3 size back propagation neural networks have been developed. Speeding et al [10] performed the modeling process through response surface methodology and artificial neural networks. A response surface model based on a central composite rotatable experimental design, and 4-16-3 size back propagation neural networks have been developed. Shajan kuriakose et al [11] conducted multi objective optimization of WEDM process by non dominated sorting genetic algorithm technique. They established that influence of the parameters on the cutting velocity and the surface finish are quiet contrary. Tarang et al [13] used feed forward neural network approach to determine the optimal cutting parameters while machining SUS 304 stainless steel in WEDM. The disparity of surface roughness and MRR with machining parameters and optimization of machining setting for minimum surface roughness and maximum MRR should be investigated experimentally and the obtained results should be interpreted and modeled statistically to understand closely the behavior of machining rate and accuracy in WEDM.

   In this cram, the upshot of the machining parameters and their level of consequence on the surface roughness and the MRR are statistically evaluated by using analysis of variance (ANOVA). Also, an optimization study is introduced for the case of low surface roughness and high MRR.

   Experiments were conducted underneath diverse machining parameters, namely, pulse on time, pulse off time, discharge current, wire tension and wire feed rate. The settings of machining parameters were determined by using Taguchi experimental design method.

3. EXPERIMENTAL DETAILS

   1. Machine tool, material and Measurement

      The exprimental studies were performed on a ELEKTRA OPTICUT 434 WEDM machine tool. Different settings of pulse on time, pulse off time, discharge current, wire tension and wire feed rate were used in the experiments as specified in Table

      TABLE 1 EXPERIMENTAL SETTINGS

      TABLE 2

      CHEMICAL COMPOSITION OF AISI 316 STAINLESS STEEL

      Elements	C	Cr	Ni	Mo	Mn	Si	P	Fe
      % (Wt.)	0.128	16.31	9.63	1.733	1.344	0.661	0.032	Balance

      The mean cutting speed data (Cs) was observed directly from the machine tool monitor. Generally during this process the wire diameter is kept constant. Therefore, the width of cut (W) remains constant. The MRR for the WEDM operations using Eqn. 1 which is shown below:

      MRR = Cs x L

      (1)

      Where Cs = cutting speed in mm/min., L = thickness of the material in mm. A Talysurf at 0.8mm cut-off value was applied to measure the surface roughness (Ra) of each specimen.

   2. Design of experiment based on Taguchi method

   To evaluate the effects of machining parameters on performance characteristics (surface roughness and MRR), and to identify the performance characteristics under the optimal machining parameters, a specially designed experimental procedure is required. Conventional investigational design methods are too complex and difficult to use. Additionally, large number of experiments has to be carried out when number of machining parameters increases. In this study, Taguchi method, a powerful tool for parameter design of performance characteristics, was used to determine optimal machining parameters for minimum surface roughness and maximum MRR in WEDM. In Taguchi method, process parameters which influence the products are separated into two main groups: control factors and noise factors. The control factors are used to select the best conditions for stability in design of manufacturing process, whereas the noise factors denote all factors that cause variation. Taguchi projected to get hold of the characteristic data by means of orthogonal array, and to explore the recital measure from the data to decide the optimal process parameters. This scheme uses an unusual plan of orthogonal arrays to cram the whole constraint space with petite number of experiments only. In this study, five machining parameters were used as control factors and each parameter was designed to have three levels, denoted 1, 2 and 3 [14, 15, 16] . According to the Taguchi quality design concept, a L27 orthogonal array table with 27 rows, analogous to the number of experiments as preferred for the experiments and the consequences are presented in Table 3.

   Sl.	Contro	Unit		Range			Level
   No .	l factors		Symbol	Actual	Deci ded	1	2	3
   1	Pulse on	µsec	A	1 – 10	4 6	4	5	6
   	time
   2	Pulse off	µsec	B	1 – 10	4 6	4	5	6
   	time
   3	Pulse	amps	C	0- 30	2 5	2	3.5	5
   	peak curren
   	t
   4	Wire	gm	D	200	600	60	800	1000

   tensio n

   1500

   0

   1000

   5 VoluWmiree 6, Issmu/em0in2

   feed

   rate

   E 0 10 3 5 3 4

   Pub5lished by, www.ijert.org

   TABLE 3

   ORTHOGONAL ARRAY FOR L27(313) TAGUCHI DESIGN

   td>

   5

   L27 (313 )	A Pulse on time µsec	B Pulse off time µsec	C Pulse peak Current amps	D Wire tension gm	E Wire feed rate m/min	M RR m m2/ mi n	SR (µm)
   1	4	4	2	600	3	17	3.20
   2	4	4	3.5	800	4	21	3.64
   3	4	4	5	1000	5	24	3.97
   4	4	5	3.5	800	4	15	3.42
   5	4	5	5	1000	5	24	3.77
   6	4	5	2	600	3	15	4.02
   7	4	6	5	1000	5	26	3.88
   8	4	6	2	600	3	11	3.43
   9	4	6	3.5	800	4	25. 5	3.76
   10	5	4	2	800	5	18	3.55
   11	5	4	3.5	1000	3	23	3.52
   12	5	4	5	600	4	27	3.49
   13	5	5	3.5	1000	3	23	3.89
   14	5	5	5	600	4	24	3.75
   15	5	5	2	800	5	16	3.26
   16	5	6	5	600	4	25	3.55
   17	5	6	2	800	5	19	3.38
   18	5	6	3.5	1000	3	22	3.28
   19	6	4	2	1000	4	24	3.75
   20	6	4	3.5	600	5	23	3.64
   21	6	4	800	3	27	3.68
   22	6	5	3.5	600	5	24	3.78
   23	6	5	5	800	3	19	3.76
   24	6	5	2	1000	4	17	3.74
   25	6	6	5	800	3	27	3.78
   26	6	6	2	1000	4	19	3.75
   27	6	6	3.5	600	5	21	3.76

   denotes the number of experiments. The S/N ratio can be calculated as a logarithmic conversion of the loss function as shown below:

   S/N ratio for MRR = -10 log 10 (LHB) (4) S/N ratio for SF = -10 log 10 (LLB) (5)

   Despite the consequences of the sort of the performance characteristics, a greater S/N value corresponds to a better performance. Therefore, the best possible level of the machining parameters is the level with the greatest S/N ratio value. By applying the Eqns. (2) (5), the S/N ratio values for each experiment of L27 (Table 3) was calculated (Table 4 and 5). Based on the analysis of S/N ratio, the optimal machining performance for the MRR was obtained at 6 µsec. pulse on time (Level 3), 4 µsec. pulse off time (Level 1), 5 amps. Pulse peak current (Level 3), 1000 g. wire tension (Level 3) and 4 m/min. wire feed rate (Level

   2) settings. Fig. 1 shows the effect of machining performance on the MRR. The optimum machining performance for the surface finish was obtained at 5 µsec. pulse on time (Level 2), 4 µsec. pulse off time (Level 1), 2 amps. Pulse peak current (Level 1), 800 g. wire tension (Level 2) and 3 m/min. wire feed rate (Level 1) settings. Fig. 2 shows the upshot of machining performance on the surface finish.

   TABLE 4

   S/N RATIO VALUES MRR

   Machining parameter	Mean S/N ratio by factor level (dB)
   	Level 1	Level 2	Level 3
   A	25.62	26.70	26.88
   B	27.01	25.71	26.48
   C	24.61	26.74	27.84
   D	26.05	26.19	26.96
   E	25.90	26.68	26.62

4. ANALYSIS AND DISCUSSION OF EXPERIMENTAL RESULTS

   The analysis of variance (ANOVA) was used to found statistically noteworthy machining parameters and the percent contribution of these parameters on the surface roughness and the MRR. In Taguchi method [8], a loss function is used to estimate the variation among the investigational value and the preferred value. This loss function is further malformed in to a signal-to-noise (S/N) ratio. There are numerous S/N ratios existing depending on the type of characteristics; lower the better (LB), nominal is the best (NB), and higher is better (HB). In WEDM, the lower surface roughness and higher MRR are the sign of better performance. For the HB and LB, the definition of the loss function (L) for machining performance results (MRR, surface roughness) of n repetitive number are :

   n

   Overall mean = 26.398 dB

   TABLE 5

   S/N RATIO VALUES FOR SURFACE FINISH

   Machining parameter	Mean S/N ratio by factor level (dB)
   	Level 1	Level 2	Level 3
   A	-11.2865	-10.9151	-11.4516
   B	-11.1236	-11.3720	-11.1580
   C	-11.0180	-11.1920	-11.4440
   D	-11.1680	-11.0690	-11.4160
   E	-11.1457	-11.2407	-11.2669

   Overall mean = -11.2272 dB

   A better feel for the relative effect of the different

   LHB

   LLB

   1/ n1/ Y 2 MRR i1

   n

   1/ nY 2 SF i1

   (2)

   (3)

   machining parameters on the MRR and the surface finish was obtained by disintegration of variance, which is called analysis of variance (ANOVA). The virtual significance of the machining parameters with respect to the MRR and the

   Where YMRR and YSF are the response for material removal rate and surface finish correspondingly and n

   surface finish was investigated to determine more

   accurately the optimum combinations of the machining

   parameters by using ANOVA. The results of ANOVA for the machining outputs are presentedin Tables 6 and 7. Statistically, F-test provides a verdict at some confidence level as to whether these estimates are significantly different. Larger F-value indicates that the variation of the process parameter makes a big change on the performance characteristics. F-values of the machining parameters are compared with the fitting confidence table.

   According to this analysis, the most effective parameter

   -9.5

   S/N ratio for Ra

   -10

   -10.5

   -11

   -11.5

   -12

   Pulse on time Off time

   Current Wire tension Wire feed speed Grand mean

   A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3

   Parameter level

   with respect to MRR is pulse peak current. Pulse on time and pulse off time are found as statistically significant factors, whereas the effect of wire tension and wire feed rate on the MRR was insignificant. Percent contribution indicates the relative power of a factor to reduce variation. For a factor with a high percent contribution, a small variation will have a great influence on the performance. The percent contributions of the machining parameters on the MRR are shown in Table 6. According to Table 6, pulse peak current was found to be the major factor affecting the MRR (67.089%), whereas pulse on time was found to be the second ranking factor (11.573%) and pulse off time was start to be the third grade reason (10.718). The percent involvement of wire tension and wire feed rate are minor, being 5.9797% and 4.642% correspondingly. According to F-test analysis, the momentous parameters on the MRR are open circuit voltage, pulse on time and pulse off time. During the above examination the optimal machining parameters identified for MRR are A3B1C3D3E2.

   The percent contributions of the machining parameters on the surface finish are shown in Table 7. According to Table 7, pulse on time is found to be the major factor affecting the surface finish (44.667%). The pulse peak current, wire tension and pulse off time are considered as significant factors with the percent contribution on the surface finish are 23.484, 18.73 and 10.723% respectively. The wire feed rate considered statistically insignificant, because it has a very low F value and percent contribution (2.3958%). Therefore this is included in the error term. Based on the above results the optimal cutting conditions obtained for surface finish are A2B1C1D2E1.

   Pulse on time off time Current Wire tension Wire feed speed Grand mean

   30

   S/N ratio for MRR

   28

   26

   24

   Figure 2. The effect of machining parameters on surface finish

   TABLE 6

   RESULTS OF ANOVA FOR MRR

   Machi nin Param eters	Degre es of freedom	Sum of square	Mean square	F test	% Contribution
   A	2	8.3491	4.1745	2.4 928	11.5725
   B	2	7.7324	3.8662	2.3 087	10.7177
   C	2	48.401	24.2009	14. 451	67.0886
   D	2	4.3135	2.1567	1.2 879	5.9788
   E	2	3.3491	1.6745	1	4.6422

   TABLE 7

   RESULTS OF ANOVA FOR SURFACE FINISH

   Machining Paramet ers	Degr ees of freedom	Sum of square		Mea n square	F test	% Contribution
   			1.482	0.74	18.6
   A	2	6		13	436	44.66714
   			0.355	0.17
   B	2	9		79	4.4757	10.72307
   			0.779	0.38
   C	2	5		97	9.8021	23.48443
   			0.621	0.31
   D	2	6		08	7.8175	18.72953
   			0.079	0.03
   E	2	5		97	1	2.395834

   22

   20

   A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3

   Parameter level

   Figure 1. The effect of machining parameters on MRR

5. CONFIRMATION EXPERIMENT

   The confirmation experiment is the final step of the Taguchis design of experiment process after selecting the optimal parameters. The purpose of the confirmation experiment is to validate the conclusions drawn during the analysis phase. In this learning, behind determining the most favorable situation and predicting the rejoinder under these conditions, a new experiment was planned and conducted with the optimum levels of the machining parameters. The concluding step is to envisage and validate the enhancement of the performance characteristic. The predicted S/N ratio opt using the best possible levels of the machining parameters can be calculated as:

   p

   TABLE 9

   RESULTS OF THE CONFIRMATION EXPERIMENT FOR SURFACE FINISH

   	Initial cutting parameters	Optimal cutting parameters
   		Prediction	Experiment
   Level	A2B2C2D 2E2	A2B1C1D 2E1	A2B1C1D 2E1
   Surface roughness (µm)	3.76		3.44
   S/N ratio (dB)	-11.5038		-10.7312

   opt

   m (i m )

   i1

   (6I)mprovement in S/N ratio for surface finish = 0.7726 dB

6. CONCLUSION

where m is total mean of S/N ratio, opt is the mean of S/N ratio at the optimal level, and p is the number of chief machining parameter that appreciably concern the performance. The results of experimental confirmation using optimal machining parameters are shown in Tables 8 and 9.

Table 8 shows the assessment of the predicted MRR with the authentic MRR using most favorable machining parameters. The improvement in S/N ratio from the starting machining parameters to the level of optimal machining parameters is 0.7903 dB. The MRR is increased by 1.0952 times. So, the MRR is greatly improved by using the approach. Table 9 shows the comparison of the predicted surface finish with the actual surface finish using the optimal machining parameters. The enhancement in S/N ratio from the starting machining parameters to the optimal machining parameters is 0.7726 dB. The surface roughness is decreased by 1.093 times. The experimental results inveterate the validity of used Taguchi method for enhancing the machining performance and optimizing the machining parameters. The MRR and the surface finish are significantly enhanced by using the approach.

TABLE 8

RESULTS OF THE CONFIRMATION EXPERIMENT FOR MRR

Improvement in S/n ratio for MRR = 0.7903 dB

This paper has offered an exploration on the optimization and the consequence of machining parameters on the MRR and the surface finish in WEDM operations. The outcome of various machining parameter such as pulse on time, pulse off time, pulse peak current, wire tension, and wire feed speed has been premeditated during the machining of AISI 316 stainless steel. The level of consequence of the machining parameters on the MRR and the surface finish is determined by using ANOVA. Based on ANOVA method, the greatly effectual parameters on both the MRR and the surface finish were found as pulse peak current and pulse on time other than the added parameters considered in this study. The results showed that pulse peak current was about six times imperative than the second graded factor (pulse on time) for controlling the MRR, whereas pulse on time for controlling the surface finish was about two times more significant than the second ranking factor (pulse peak current). An optimum parameter blend for the maximum MRR and minimum surface roughness was obtained by using the analysis of signal-to-noise(S/N) ratio. The confirmation tests indicated that it is probable to augment MRR and diminish surface roughness appreciably by using the projected statistical technique. The experimental results inveterate the validity of the used Taguchi method for enhancing the machining performance and optimizing the machining parameters in WEDM operations.

REFERENCES

1. Chiang Ko Ta, Fu-Ping Chang, Optimization of the WEDM process of particle reinforced material with multiple performance characteristics using grey relational analysis, J of Matrl. Proces. Tech., 2006, Vol 180, pp. 96-101.

2. Ho K H, Newman S.T, State of the art in wire electrical discharge machining, Int J of Mach tools and Mfg., 2004, Vol 44, pp 1247- 1259.

3. Liao Y.S., Huang J.T., Su H.C. A study on the machining- parameters optimization of wire electrical discharge machining Journal of Materials processing technology, 1997, Vol 71, pp.487- 493.

4. Mahapatra.S.S., Amar Patnaik, Optimization of wire electrical discharge machining process parameters using Taguchi method Int J Adv Manuf Technol., 2006, Vol.34, pp. 911925.

5. Ozdemir N, Cebli Ozek, n investigation on machinability of nodular cast iron by WEDM, Int J Advanced Mfg Tech., 2005, Vol 149, pp. 50-57.

6. Puri A B, Bhattacharya.B, Modeling and analysis of white layer depth in wire cut EDM process through response surface methodlogy, Int J of Advanced Mfg Technology, 2005, Vol. 25, pp. 301-307.

7. Ramakrishnan R, Karunamoorthy L, Multi response optimization of wire EDM operations using robust design of experiment, Int Journal of Adv Mfg. Tech., 2005, Vol 29, pp 105-112.

8. Ross J Phillip, Taguchi Technique for Quality Engineering, 1998, McGraw Hill Inc.

9. Sarkar S, Mitra S, Bhatacharyya B, Parametric optimization of wire electrical discharge machining of titanium aluminide alloy through an artificial neural network model, Int J of Adv Mfg Tech., 2005 ,Vol 27, pp 501-508.

10. Speeding T.A , Wang Z.G.Parametric optimization and surface characteristics of WEDM process, 1997 , Precision Engg., Vol 20, pp 5-15.

11. Shajan Kuriakose, Shanmugam M.S, Multi-objective optimization of wire electro discharge machining process by Non-dominated sorting genetic algorithm, J of Material process technology, 2005, Vol 170, pp. 133-141.

12. Shanmugam M S, P Hariharan, Gowri S, Optimization of wire electrical discharge machining based on Taguchi method, 2006, proceedings of AMPC.

13. Tarang Y S and Ma S.C, Determination of optimal cutting parameters in WEDM, Int j of Mach Tools and Mfg., 1995, Vol 35, pp. 1693-1701.

14. G.Ugrasena, H.V.Ravindra, G.V.NaveenPrakash, Y.N.Theertha Prasad, Optimization of Process Parameters in Wire EDM of HCHCr Material Using Taguchi's Technique, Materials Today: Proceedings, 2015, Vol.2, pp. 2443-2452.

15. Ravindranadh Bobbili, V.Madhu, A.K.Gogia, Multi response optimization of wire-EDM process parameters of ballistic grade aluminium alloy, Engineering Science and Technology, an International Journal, 2015, Vol. 18, pp. 720-726.

16. Sameh Habib, Optimization of machining parameters and wire vibration in wire electrical discharge machining process, Mechanics of Advanced Materials and Modern Processes, 2017, 3, pp. 1-9.