Efficient process parameter optimisation procedure in Laser Powder Bed Fusion

Abstract

Important properties such as porosity, microstructure and surface roughness of metal parts produced by Laser Powder Bed Fusion (L-PBF) are largely determined by process parameters including laser power, scan speed, distance between scan lines and layer thickness. L-PBF parameter optimisation methods are now generally based on the production of an array of samples where the laser power and scan speed are varied. A large number of samples needs to be analysed for selecting a suitable combination of parameters. This makes parameter optimisation for L-PBF complicated and time consuming. In this work, an efficient procedure is proposed for optimising the key parameters of the laser scan pattern. Build files for samples are generated, which cover a predefined parameter range. Scripts are used for automated evaluation of optical microscope images of cross-sections. This procedure allows variation of parameters within a sample and thereby strongly reduces the required number of samples and time for parameter optimisation.