Evaluation of micro dry wire EDM of stainless steel on kerf accuracy

Abstract

Micro dry wire EDM (?DWEDM) is a process where gas is used as the dielectric fluid instead of a liquid. In this process certain modifications of wire EDM (WEDM) are needed during the machining operation to achieve stable machining. Smooth and stable machining operation as well as the kerf variation in ?DWEDM process remains as critical issues. Thus, the objectives of this research are to establish a stable ?DWEDM process and to develop kerf mathematical model. The investigation was performed on a stainless steel (SS304) with a tungsten wire as the electrode using integrated multi process machine tool, DT 110 (Mikrotools Inc., Singapore). This research consists of two main parts which are the process parameters selection and the mathematical modelling of kerf in ?DWEDM. For the process parameters selection, types of dielectric fluid, dielectric fluid pressure, polarity, threshold voltage, wire tension, wire feed rate, wire speed, gap voltage, and capacitance were the controlled parameters. The experimentation method used in this part was a conventional experimental method, one-factor-at-a-time (OFAT). The machining length of the microchannels were measured using scanning electron microscope (SEM). Stable and smooth machining operation of ?DWEDM was found to be with compressed air as the dielectric fluid, workpiece positive polarity, 24% threshold voltage, 0.0809 N wire tension, 0.2 ?m/sec wire feed rate, and 0.6 rpm wire speed. The best conditions in this part were proposed as the fixed parameters while the capacitance and gap voltage as the controlled parameters for the kerf investigation. For mathematical modelling of kerf, statistical analysis based on the response surface methodology (RSM) was employed. RSM employed consists of two main designs which were first-order design; Plackett-Burman design; and second-order design; central composite design (CCD). Plackett-Burman design was utilized in order to check the validity of the process parameter selection results. The validation results showed that the proposed parameters; capacitance (10.00-0.10 nF) and gap voltage (80-110 V); were the variables that should be used as the controlled parameters for kerf investigation in ?DWEDM using CCD. The results were obtained by measuring the kerf using SEM. The first-order design and the second-order design were analysed using ANOVA. The investigation of kerf was divided into two responses which were upper kerf and bottom kerf. Empirical models were developed for both of the responses. Both parameters; capacitance and gap voltage have high influence on both of the responses. The optimum parameters for both minimum upper and bottom kerf were found to be 0.1 nF capacitance, 91 V gap voltage, compressed air dielectric fluid, 0.0345 MPa dielectric fluid pressure, workpiece positive polarity, 24% threshold voltage, 0.0809 N wire tension, 0.2 ?m/sec wire feed rate, and 0.6 rpm wire speed. The developed models are found to be adequate since the percentage error were relatively small (< 3%). The main innovative contribution of this research is the identification of process parameters together with their level for stable machining and formulation of mathematical model for optimum kerf.