Statistical modelling and optimization of surface roughness in the selective laser sintering process

Selective laser sintering (SLS) is a layered manufacturing process that builds prototypes by selective sintering of materials in powder form, like thermoplastic polymer powder (Poll amide 2200), using a CO2 laser. Prototypes made by SLS are widely used in product development as they can be used for product testing. SLS prototypes, therefore, should have a very good surface finish for functional performance as well as aesthetics. However, prototypes made by the SLS process have comparatively high surface roughness due to the stair stepping effect. Surface roughness of the prototypes also depends on the various process parameters. This paper attempts to study the effect of process parameters, namely build orientation, laser power, layer thickness, beam speed, and hatch spacing, on surface roughness. Central rotatable composite design (CCD) of experiments was used to plan the experiments. Analysis of variance (ANOVA) was used to study the significance of process variables on surface roughness. In the case of upward-facing surfaces, build orientation and layer thickness have been found to be significant parameters. In downward-facing surfaces, other than build orientation and layer thickness, laser power has also been found to be significant. Empirical models have been developed for estimating the surface roughness of the parts. A trust-region-based optimization method (standard module of MATLAB) has been employed to obtain a set of process parameters for obtaining the best surface finish. A confirmation experiment has been carried out at an optimum set of parameters and predicted results were found to be in good agreement with experimental findings. A case study of a standard part 'Truncheon' is also presented.