Research on technique of wavefront retrieval based on Foucault test

Abstract

During finely grinding the best fit sphere and initial stage of polishing, surface error of large aperture aspheric mirrors is too big to test using common interferometer. Foucault test is widely used in fabricating large aperture mirrors. However, the optical path is disturbed seriously by air turbulence, and changes of light and dark zones can not be identified, which often lowers people's judging ability and results in making mistake to diagnose surface error of the whole mirror. To solve the problem, the research presents wavefront retrieval based on Foucault test through digital image processing and quantitative calculation. Firstly, real Foucault image can be gained through collecting a variety of images by CCD, and then average these image to eliminate air turbulence. Secondly, gray values are converted into surface error values through principle derivation, mathematical modeling, and software programming. Thirdly, linear deviation brought by defocus should be removed by least-square method to get real surface error. At last, according to real surface error, plot wavefront map, gray contour map and corresponding pseudo color contour map. The experimental results indicates that the three-dimensional wavefront map and two-dimensional contour map are able to accurately and intuitively show surface error on the whole mirrors under test, and they are beneficial to grasp surface error as a whole. The technique can be used to guide the fabrication of large aperture and long focal mirrors during grinding and initial stage of polishing the aspheric surface, which improves fabricating efficiency and precision greatly.