Modelling melt pool characteristics to predict process parameters for selective laser melting of titanium alloys

Abstract

The time-consuming experimental process required to optimize process parameters for the selective laser melting (SLM) of new materials is a major hurdle in adopting this technology. For this reason, various methods in the literature attempt to shorten the time taken to optimise the process parameters for a specific material. One possible solution to this problem is the use of accurate numerical modelling to predict the required process parameters. Currently, melt pool modelling on part-scale is computationally expensive and not a feasible method for reducing the time required for the optimisation of process parameters. However, some studies have indicated that it is possible to effectively optimise process parameters through experimental investigation of the cross-sectional geometry of the molten region of single tracks created at various process parameters. It follows, therefore, that the accurate modelling of single tracks can be an effective way of determining optimal process parameters. This study reports on efforts to set up a numerical model of a single track of the SLM process that is accurate enough to be effectively used to optimise process parameters. Experimental and simulation results were obtained for a broad range of laser powers and scanning speeds to investigate the accuracy of the model. Thereafter, the effects of changing various simulation parameters on the accuracy of the model with respect to simulated melt pool dimensions were investigated to find parameters that could be responsible for inaccuracies in the developed model. These parameters included the temperature-dependent absorptivity of the material, the evaporation pressure coefficient, and to a lesser extent, the effect of temperature on the surface tension. It was concluded that provided that accurate values could be obtained for the simulation parameters discussed in this document, single-track SLM simulations could be a very powerful tool for quickly and effectively determining process parameters for any material.