A numerical and experimental study of pulsed laser polishing of metals at the meso/micro scale.

摘要

Pulsed Laser Micro Polishing (PLmuP) was investigated numerically and experimentally as a method to reduce the surface roughness of parts at the meso/micro-scale having poor relative surface roughness. A model of the PLmuP process was created based on analytic fluid flow equations for the amplitude decay of capillary waves. The model also included results from a one-dimensional finite element (FEM) simulation that was used to estimate the melt depths and durations caused by laser pulses ranging from 50-1000 ns in duration. The critical spatial frequency, fcr, was developed to predict the spatial frequency above which significant polishing would be observed.;Initial experiments on electroplated and micromilled nickel samples yielded positive results; sample surface roughness (Ra) was reduced by as much as a factor of seven on an electroplated sample. Polishing on the micromilled sample was found to be limited by the low spatial frequency content present in the surface that was the result of the micromilling process used to create it. Also, the fcr value was found to be a good indicator of the spatial frequency above which significant amplitude reduction was seen in the spatial frequency domain.;Further line polishing studies were performed on four metals, nickel, Ti6A14V, SS 316 and Al 6061-T6 using an improved experimental apparatus that included a galvanometer-based optical scan head. Results from these experiments showed that the surface roughness (Ra) of all four metals could be reduced by approximately a factor of two. Additionally, the fcr predictions made using the model were found to have the correct general trend, but the experimentally observed fcr values were higher than those predicted by the model.;Experiments on the processing parameters of the PLmuP process were also performed. These showed that increasing the laser spot overlap was beneficial for polishing. It was observed that longer laser pulses were more beneficial for polishing; a prediction consistent with the model that was developed. Two-dimensional polishing was also investigated using a traditional zig-zag trajectory as well as trajectories generated using an attractive field-based algorithm.