Traceable characterization of areal topography of structured surfaces using confocal microscopy

Abstract

This PhD thesis addresses, from an analytical and experimental point of view, the characterization of structured surfaces through 3D roughness parameters. The thesis includes novel models for the determination of outliers and develops an original method to ensure the traceability of the measurements made by confocal microscopy in the X, Y, Z axes. The experimental validation of the models and the proposed methodology has been made from different files with synthetic data provided by the National Institute of Standards and Technology (NIST), available on its Website, and also through the use of physical standards calibrated by external laboratories. The uncertainties associated with the different parameters evaluated, such as Sa, Sq, Ssk, Sku, St, Sp, Sv..., have been estimated by applying the Monte Carlo Method to surface finish measurements. For this, the most significant sources of uncertainty: kinematic, geometric, optical, environmental... and those due to the mathematical models used, have been taken into account. The calculation algorithms have been developed in Matlab and their validation have been carried out using the supercomputer “Magerit” of the Centro de Supercomputación y Visualización of the Universidad Politécnica de Madrid.