Uncertainty in measurement of surface topography

The 2.5-dimensional (2.5D) roughness parameters were standardized in 2012. With their increasing use in science and industry, the request for traceability and uncertainty evaluation for these parameters follows logically. This paper gives an overview of the problems and possibilities that appear when uncertainties have to be associated with values that are derived from a measured surface topography, such as the Ra-value of a periodic specimen, the RSm value of a type-D standard, and the Sa-value of a single cutoff length of a type D standard. It is shown that straightforward implementation of the methods described in the 'Guide to the Expression of Uncertainty in Measurement' (GUM) leads to impossible and impracticable equations because of the correlations between some millions of measurement points. A practical solution is found by considering the main aspects of uncertainty, as these are given in the recent ISO 25178 standards series, and applying these to a measured surface topography as a whole.