Sensitivity of power spectral density (PSD) analysis for measuring conditions

Nowadays power spectral density (PSD) analysis is one of the leading characterisation techniques of surface topography. Fractal dimension obtained from PSD plays a significant role in recent friction, adhesion and wear models.
Knowledge of the sensitivity of PSD for the analysis and adjustment and measurement of parameters is relevant in extensive and reliable applications. In this study, a three-dimensional (3D) PSD analysis of engineering surfaces is carried out, and its sensitivity for frequency sampling, for line fitting – required to calculate the fractal dimension – and for sampling the distance of measurement are investigated.
Based the on results, it can be established that 3D PSD analysis provides stable results when using only few frequencies. Line fitting has a considerable impact on fractal dimension results, so the application of PSD in the calculation of fractal dimension needs circumspection. A huge amount of fractal information is in the height frequency range, so the sampling distance greatly influences the fractal dimension of surface. Thus, measuring conditions are required to be specified in order to get proper information about surface self-affinity.