Error Separation for Wide Area Film Measurement

We wanted to use a multiple probes and white light interferometer to measure the surface profile of thin film. However, this system, as assessed with a scanning method, suffers from the presence of a moving stage and systematic sensor errors. In this paper, in order to separate measurement error caused by the moving stage and systematic sensor errors, the least squares analysis is applied to achieve self-calibration in the measurement process. The modeling principle and resolution process of the least squares analysis with multiple probes and autocollimator are introduced and the corresponding theory uncertainty calculation method is also given. Using this method, we analysis the experimental data and obtain a shape close to the real file. Contrasting with the actual value, the bias and uncertainty in the case of different number of probes are discussed. The results demonstrated the feasibility of the constructed multi-ball cantilever system with the autocollimator for measuring thin film with high accuracy. Copyright © 2014 IFSA Publishing, S. L.