Highly accurate non-contact characterization of engineering surfaces using confocal microscopy

TLDR

Optical non‑contact techniques are attractive for 3‑D characterization of complex engineering surfaces, yet their use has been limited to moderate height and slope variations because high‑frequency artefacts make interpretation difficult and have led to over‑estimates of optical resolution compared with tactile methods. This study introduces two confocal microscopy approaches for highly accurate surface characterization. The first method operates on sub‑centimetre fields and matches the precision of mechanical stylus instruments even on rough surfaces, while the second method covers measurement fields up to square centimetres. Results on PTB roughness standards show that the first method yields surface statistics and raw topographic data that closely agree with tactile measurements.

Abstract

Optical non-contact techniques are very interesting for 3D characterization of sensitive and complex engineering surfaces. Unfortunately, the application of optical techniques was for many years restricted to selected types of surfaces which have only moderate variations of height and surface slope relative to their lateral resolution and measurement field. Owing to the fact that artefacts and form deviations occur with high spatial frequencies in optically measured topographs, there were some difficulties in interpreting the results and comparing them with the tactile standard techniques for surface characterization. Furthermore, artefacts in optically measured profiles have often been misinterpreted in terms of the resolution of optical techniques being higher than that of the tactile techniques. This paper presents two optical methods of confocal microscopy for highly accurate characterization of surfaces. The first method works on measurement fields of less than 1 and is in practice absolutely comparable to the mechanical stylus instrument, even on rough surfaces. For this method results compare very well not only in surface statistics but also in topographic raw data, as will be demonstrated for the PTB roughness standards. The second method works on measurement fields up to square centimetres.