Analysis of Process Parameters in Wire EDM on D2 Tool Steel using Taguchi Method

Thispaper focuses on the effect of input parameters like pulse on time, pulse off time, servo voltage, and kerf width on the output characteristics of CNC wire EDM process such as material removal rate (MRR), surface roughness (SR), and kerf width (KW). The optimum process parameters and corresponding output responses are found out using Taguchi Method. In this research, High carbon high chromium D2 tool steel is used as the work piece with 0.25mm brass wire as tool. Keywords—WEDM, MRR, SR, KW, S/N Ratio.


INTRODUCTION
Achieving high accuracy and tighter tolerances during machining of materials is essential for many industries. Wire electric discharge machining (WEDM) helps to produce parts in economical way than traditional manufacturing process. In WEDM the material removal is through, electro erosion machining process , in which electric spark is generated between tool and work piece, flushed with de-ionised water. The material removal takes place due to repeated electric discharges between work piece and wire connected in an electrical circuit. The literatures related with experiments focussing on characteristic features of WEDM, it is found that parameters like pulse on time, pulse off time, voltage and wire feed have significant role in determining the performance characteristics like material removal rate, surface roughness and kerf width. Kumar

II.
DESIGN OF EXPERIMENTS According to the capability of machine tool, cutting tool and work piece, various process parameters and the levels for each parameters are selected and are listed in the Table 1. The designed combination of input parameters based on L9 orthogonal array are shown in Table 2and its corresponding material removal rate, surface roughness and kerf width are shown in the Table 3.  According to Taguchi method, the S/N ratio is the ratio of signal to noise, where signal represents the desired value and noise represents the undesired value. The output responses are used to calculate the S/N ratios given in Table 4.

Fig. 1: Main Effects Plot for S/N Ratio of MRR
The delta value is the variation of mean S/N ratio from first level to the third level, and thus shows how on each parameter affect the particular response.It can be seen that wire feed has the highest delta value and hence wire feed has the highest influence on MRR.From the main effects plot of MRR (Fig. 1) it is clear that the optimum process parameters for getting the optimum MRR is Pon = 125µs, P off = 48, µs V = 25V, wire feed = 2mm/min. The regression equation for MRR is found as follows; MRR = -2.46 + 0.04185 pulse on -0.0619 pulse off + 0.0907 servo v -0.3314 wire feed (1) The mean S/N ratio values for surface roughness are shown in Table 6 (smaller the better for SR).Here pulse off time has the highest delta value and hence influence the surface roughness the most.From the main effects plot

International Journal of Advanced Engineering Research and Science (IJAERS)
[ Vol-5, Issue-9, Sept-2018]  https://dx.doi.org/10.22161/ijaers.5.9.8  ISSN: 2349-6495(P) | 2456-1908(O) of SR (Fig. 2), the optimum process parameters are found to be pulse on time = 120µs, pulse off time = 48µs, servo voltage = 20v, wire feed = 2mm/min. The regression equation for SR is found as follows; SR = 12.94 + 0.0399 pulse on -0.2515 pulse off -0.0711 servo v -0.1731 wire feed (2)  Table 7 (smaller the better for KW).Here pulse on time has the highest delta value and hence influence the surface roughness the mostly. From the main effects plot of Kerf Width (Fig. 3) the optimum process parameters for kerf width are found to be, pulse on = 125µs, pulse off = 50µs, servo voltage = 20V, wire feed = 2 mm/min. The regression equation for KW is found as follows; KW = -2.073 + 0.02867 pulse on -0.0044 pulse off -0.02996 servo v-0.0161 wire feed(3)  The optimum output responses are found using regression analysis as shown in Table 9.

IV.
CONCLUSIONS Experimental investigation of D2 tool steel has been done on wire EDM and the following conclusions were made;  It was found thatmaterial removal rate was most influenced by wire feed, surface roughness by pulse off time and kerf width by pulse on time.  The optimum combination of proces s parameters for material removal rate, surface roughness and kerf width were found.  The optimum output responses were also found using regression analysis.