Roughness prediction on laser polished surfaces

The paper presents a methodology to predict the surface topography on laser polished surfaces. The method is based on the thermal field prediction generated by the laser. The thermal field is obtained through the differential equation of heat transfer by conduction. The prediction of the resulting topography is based on the filtering of the spatial frequency spectrum of the initial topography which is obtained by FFT analysis. Once the temperature field is obtained, a critical frequency is calculated and a low-pass filter is applied to cut the frequencies higher than the critical one. The resulting surface is reconstructed using the remaining frequencies. The methodology has been tested on a DIN 1.2379 tool steel and the experimental validation shows a relatively good agreement between the predicted and measured values of mean roughness (Ra) and mean roughness depth (Rz), with errors lower than 15% in all the cases.