Research On Absolute Testing Of The Optical Flat Mirror

Optical system calls for an increasingly higher surface accuracy of an optical flat and hence stricter requirements for related measurement techniques.In optical flat measurement,the tradition method of measuring directly with an interferometer belongs to relative measurement,whose accuracy is subject to a large degree to the standard plane surface.In general,its measurement accuracy falls in ?/20—?/50.To ensure a higher accuracy,absolute measurement is needed.Absolute measurement can effectively eliminate the influence of the standard plane surface via multiple measurements,and reproduce the measured flat surface,thus a high-accuracy measurement.Three-flat test based on Zernike polynomials fitting is an absolute measurement designed on the foundation of traditional three-flat test.For it is easy to operate and has a high accuracy,the Zernike method is commonly used as an absolute measurement.Theoretically,the flat surface reproduced in this way can meet the nano-level absolute measurement requirements.In actual measurement,there are random errors and system errors.This paper mainly focuses on the three-flat test based on Zernike polynomials fitting and conducts a taxonomic study of these two types of errors in measurement.Of them,random errors include impact of the humidity and temperature of measure environment on cavity length and the measured flat;while system errors are consisted of influence of optical axis deviation of the standard flat and the measured flat,rotation angle deviation of the measured flat and area intercept of relative measurement results.For the former,this paper calibrates,on the basis of research,the humidity and temperature of experimental environment needed in a nano-level absolute measurement,as well as the cavity length.For the latter,it analyzes the effect of different deviations on the results of absolute measurement.Furthermore,after the calibration of measure environment and absolute measurement algorithm,it proves through experiment the impact of system errors on this absolute measurement technique.Via both theoretical analysis and experimental verification,the paper finds out that optical axis deviation of the standard flat and the measured flat exerts the biggest influence on the accuracy of the three-flat test based on Zernike polynomials fitting.Besides,temperature of measure environment,cavity length,rotation angle deyiation and area intercept ratio of relative measurement resultscan also cause errors in measurement results.To satisfy the nano-level accuracy demand,cavity length should be set to 0.02m under the condition of less than 0.2 ? temperature error;optical axis deviation should be within 2 pixels and rotation angle deviation within 1°;at the same time,area intercept ratio should be adjusted to 95%.