Statistical Analysis of Mechanical Interaction of Rough Surfaces

Abstract

The contact problem of a pair of mating rough surfaces has been investigated. The problem is formulated analytically, in terms of junction deformation models and surface topography described by statistical distribution of heights and curvatures of the peaks. Statistical analysis of a surface profile was possible by suitable analog to digital conversion. For most surfaces the distribution of all heights and peak heights closely approach a Gaussian distribution. The data on distribution of radius of curvature at the peaks are best fitted by a log-normal function. Statistical correlation between peak heights and radii is found to be very low and hence heights and radii are considered to be independent. With suitable junction deformation models for a pair of asperities in sliding interaction, friction behavior of a pair of mating rough surfaces is estimated. It is shown that surface roughness does not appreciably change the friction behavior under conditions of insignificant adhesion. In cases of strong adhesion, the analytical results show that an improvement in surface finish can substantially reduce the friction coefficient.