Finite element model for topography prediction of electrical discharge textured surfaces considering multi-discharge phenomenon

•Expanding Gaussian heat flux and variable plasma flushing efficiency incorporated.•Stochastic multi-spark distribution in location, energy and chronological order.•User-defined subroutines used for dynamic simulation of spark erosions in ABAQUS.•Evolution of surface texture in EDT simulation with increasing crater overlaps.•Simulated topographies validated using error function and distribution methods. [Display omitted] The prediction of surface topography obtained with electrical discharge texturing (EDT) process is very challenging, due to stochastic nature and complex physics in occurrence of sparks. Though there are a few models for surface texture prediction in the literature, there are still ample opportunities to improve upon by adopting more realistic assumptions and formulations for simulation. This paper presents a numerical simulation of progressive development of surface topography due to sequential application of spark discharges on an instantaneous textured surface formed by all previous spark discharges. The model incorporates major single-spark features such as, Gaussian distribution of heat flux, temperature-dependent work material properties, latent heat of melting, and operating parameter dependent variation of factors such as, cathode energy fraction, spark radius, and plasma flushing efficiency; besides the multi-spark features such as, stochastic distribution of sparks with respect to: i) location, ii) energy level, and iii) chronological order. Finite element simulation of EDT is performed in ABAQUS with the help of user-defined subroutines for the dynamic simulation of spark discharges, flagging of elements attaining evaporation temperature from further heat flux application and storage of the highest temperatures attained by each element. An evolution of surface texture through the progress of spark erosions has been studied. As EDT continues, the number of craters overlapping generally tend to increase, whereas the center-to-center distance between a crater with its nearest neighbour tends to decrease. The simulated surface textures are validated with the experimental ones using two methods: error function method and Ra distribution method, and were found to be in good agreement.