Experimental Investigation and Optimization of Rough EDM of High-Thermal-Conductivity Tool Steel with a Thin-Walled Electrode.

Abstract

The industrial application of electrical discharge machining (EDM) for manufacturing injection molding, in many cases, requires forming depth cavities with high length-to-width ratios, which is quite challenging. During slot EDM with thin-walled electrodes, short-circuits and arcing discharges occur, as a result of low efficiency in removing debris and bubble gas from the gap. Furthermore, unstable discharges can cause increases in tool wear and shape deviation of the machined parts. In order to characterize the influence of the type of electrode material and EDM parameters on the deep slot machining of high-thermal-conductivity tool steel (HTCS), experimental studies were conducted. An analytical and experimental investigation is carried out on the influence of EDM parameters on discharge current and pulse-on-time on the tool wear (TW), surface roughness (Ra), slot width (S)-dimension of the cavity, and material removal rate (MRR). The analyses of the EDS spectrum of the electrode indicate the occurrence of the additional carbon layer on the electrode. Carbon deposition on the anode surface can provide an additional thermal barrier that reduces electrode wear in the case of the copper electrode but for graphite electrodes, uneven deposition of carbon on the electrode leads to unstable discharges and leads to increase tool wear. The response surface methodology (RSM) was used to build empirical models of the influence of the discharge current I and pulse-on-time ton on Ra, S, TW, and MRR. Analysis of variance (ANOVA) was used to establish the statistical significance parameters. The calculated contribution indicated that the discharge current had the most influence (over 70%) on the Ra, S, TW, and MRR, followed by the discharge time. Multicriteria optimization with Derringer's function was then used to minimize the surface roughness, slot width, and TW, while maximizing MRR. A validation test confirms that the maximal error between the predicted and obtained values did not exceed 7%.