How to select the most relevant 3D roughness parameters of a surface

TLDR

Comprehensive roughness analysis uses about sixty 3D parameters to describe surface morphology for various functions and properties. The study proposes a multiscale surface topography decomposition method for stainless steel processed by rolling and electrical discharge machining. The authors compute 56 ISO, EUR, and ASME 3D roughness parameters and use MesRug software to statistically identify the most relevant parameters and optimal analysis scale for rolling and EDM processes. For rolling, Vmc measured at a 3 µm scale best characterizes the cold‑rolling process, while for EDM, SPD at an 8 µm scale most accurately represents the surface. © 2013 Wiley Periodicals, Inc.; SCANNING 36:150–160, 2014.

Abstract

Summary In order to conduct a comprehensive roughness analysis, around sixty 3D roughness parameters are created to describe most of the surface morphology with regard to specific functions, properties or applications. In this paper, a multiscale surface topography decomposition method is proposed with application to stainless steel (AISI 304), which is processed by rolling at different fabrication stages and by electrical discharge tool machining. Fifty‐six 3D‐roughness parameters defined in ISO, EUR, and ASME standards are calculated for the measured surfaces. Then, expert software “MesRug” is employed to perform statistical analysis on acquired data in order to find the most relevant parameters characterizing the effect of both processes (rolling and machining), and to determine the most appropriate scale of analysis. For the rolling process: The parameter V mc (the Core Material Volume—defined as volume of material comprising the texture between heights corresponding to the material ratio values of p = 10% and q = 80%) computed at the scale of 3 µm is the most relevant parameter to characterize the cold rolling process. For the EDM Process, the best roughness parameter is S PD that represents the number of peaks per unit area after segmentation of a surface into motifs computed at the scale of 8 µm. SCANNING 36:150–160, 2014. © 2013 Wiley Periodicals, Inc.