Chromatic confocal precision measurement accuracy analysis with surface roughness based on a scalar diffraction and raytracing hybrid model

Abstract

Chromatic confocal sensors are widely used in surface inspection steps in precision machining. However, it is still a challenge to achieve optimum accuracy when measuring rough surfaces. Rough surfaces can lead to a deviation of several hundred nanometers in the measurement results, which is approximately twice the nominal measurement accuracy of the sensor. The wave optical model is commonly used to analyze the impact of surface roughness characteristics on the accuracy of chromatic confocal sensors. The model is computationally complex and highly dependent on the surface profile. In this paper, an improved model is proposed by combining the scalar diffraction and raytracing. Based on the generalized Harvey-shack (GHS) scattering theory and the masking effect function, the model reveals the relationship between the statistical parameters of rough surface and the error of spectral confocal signal. Based on this model, we find that the rough surface with the root mean square (RMS) roughness 0.01–0.3 μm introduces about 0.05–1 μm deviation in profile measurement. We further experimentally verified the effect of surface roughness on the spectral signals and proved the reliability of the analytical model. The model can be used as a theoretical basis for analyzing scattering phenomena caused by surface roughness and as a reference for sensor selection, design and data compensation for measuring rough surfaces using chromatic confocal sensors.