Simulating the Electrical and Thermal Conductivity in EDM Die sinking of Cu-TaC Compact Electrodes using neural network Muataz Hazza F. AL Hazza, Mohammed Baba Ndaliman, Mohammad Yeakub Ali, Ahsan Ali Khan Department of Manufacturing and Materials Engineering, Kulliyyah of Engineering International Islamic University Malaysia, Malaysia muataz@iium.edu.my Abstract— electrical and thermal conductivity of electrode materials in Electrical Discharge Machining (EDM) is the main factors that determine the suitability of these electrodes. These electrodes are usually made from graphite, copper and copper alloys because these materials have high electrical and thermal conductivity in additional to high melting temperature. The objective of this paper is to simulate the electrical conductivity and thermal conductivity for new compact electrodes made from Cupper- Tantalum Carbide (Cu-TaC). This simulation based on experimental work. Materials have been used in the experiment were Cupper and Tantalum Carbide powders. The compositions of the Cu-TaC are made of 30 % and 55 % wt of TaC, while the compacting pressures are 1, 500 psi and 3,000 psi. The thermal and electrical conductivities have been measured. The Neural Network (NN) technique has been used to simulate the electrical conductivity and thermal conductivity of the green compacted electrodes. They were subjected to sintering at temperatures of 450°C and 850 °C. Keywords-component; Simulation, Neural Network, EDM, Green Compact Electrode, Cu-TaC electrode, thermal conductivity, electrical conductivity I. INTRODUCTION Electrical discharge machining (EDM) has been an important manufacturing process for the machining of hard materials which are difficult to machine by conventional methods [1]. It is recommended for hard materials and micro machined features. One of the principal components of EDM is the tool electrode. It provides the appropriate shape or profile to the finished work piece surface. The principle of EDM machining is based on the removal of conductive materials by discharges. High frequency discharge pulse takes a place between the tool and the workpiece so as to remove the material from the work piece [2]. The electrodes for EDM are usually made of graphite, brass, copper, or copper-tungsten alloys. However, Different researchers used different materials and different methods to produce these electrodes. In EDM, the removal of material is based upon the electrodischarge erosion (EDE) effect of electric sparks occurring between two electrodes that are separated by a dielectric liquid. The pair of the workpiece and the cutting tool acts as electrodes. One of these electrodes acts as the cathode while the other is the anode. However, the cutting tool is more commonly referred to as the electrode. The electrode has been a significant component of the EDM mechanism. It has been fabricated by quite a number of methods including conventional machining, rapid tooling [3] and powder metallurgy (PM). Ndaliman et al. [4] introduced new electrodes made from Cu-TaC. They used the Powder Metallurgy (PM) method to fabricate the electrode by two different methods. Zhang et al. [5] studied the feasibility of TiN/Cu-based composite materials as (EDM) electrodes. Many researchers worked on PM fabricated electrodes. Samuel and Philip [6] examined in their research the basic properties of the green compacts electrodes produced from electrolytic copper powder. They found the electrodes suitable for use in EDM. Some of the electrodes are being used for conventional EDM with the focused on material removal rate (MRR), and surface roughness [7,8]. Tsai et al. [9] observed that higher MRR was realizable with the use of Cu-Cr PM electrode when they compared its performance with that of metallic copper electrode. Beri et al. [10] proved that better surface finish is also attainable with the use of these electrodes. PM compacted electrodes have been also produced and used in pre-sintered or sintered conditions [11, 12], while some are being used for machining in green compacted forms [4]. This paper therefore focuses on simulating the thermal conductivity, electrical conductivity and density of the new EDM electrodes produced from Tantalum Carbide (TaC) and Copper (Cu) powders in two different forms- the green compacted electrodes (GCE) and sintered electrodes (SE) by using PM method. These are important properties of EDM electrodes which will not only determine their suitability in machining, but also the level of surface integrity to be attained by the workpiece. The electrodes produced from these powders have earlier been investigated for suitability in EDM as well as capability in surface modification of workpiece [4,13]. Mukherjee and Ray [14] classified the modeling techniques used in metal cutting into three main parts: statistical regression (single, multi and multivariate), fuzzy set theory and artificial neural network. Some of the researchers used statistical regression [15-17] Many researchers used the neural network in predicting and estimating the machining outputs for different machining process [18-23]. However, mathematical models also can be used in machining. 2013 International Conference on Advanced Computer Science Applications and Technologies 978-1-4799-2758-6/13 $31.00 © 2013 IEEE DOI 10.1109/ACSAT.2013.22 74