Search

A novel magnetorheological finishing (MRF) method, which is named Lap-MRF, is proposed in this paper. The magnetorheological fluid (MR fluid) in the polishing zones can be renewed continuously so that the determinacy is ensured. A lap, instead of a large polishing wheel, is used to expand the polishing area, which improves the material removal rate largely. Lap-MRF uses flexible MR fluid as polishing pad to match the surface well. Moreover, the polishing pad executes planetary motion so as to obtain smooth surface. In this paper, the principle of Lap-MRF and the theoretical model of material removal rate are presented. Using the finite element analysis method, the permanent magnet unit is simulated and a multi-parameter optimization is conducted to improve the performance of Lap-MRF. Finally, a series of polishing experiments and simulation process are carried out. For K9 sample, the volume removal rate is up to 0.76 mm3/min and its relative change rate is less than 5.5%. For silicon modification layer sample, the surface roughness is improved to 0.788 nm RMS (root mean square) from 1.610 nm RMS. There is no deep pit and the polishing ripple is not apparent on the surface. For Φ1000 mm flat mirror, the convergence efficiency of simulation process is up to 97.2%. These results verify the validity of the proposed method, which makes Lap-MRF to be a promising finishing technology for large aperture mirrors. [ABSTRACT FROM AUTHOR]

Abstract: We present a method for optical wavefront measurement using phase retrieval with subaperture stitching method in situations where detected intensity patterns are undersampled or where the limit of aberrations is determined by the largest spot size overfilling the detector array. In this technique we use a smaller moving subaperture to produce a number of adequately sampled intensity patterns captured by the detector located near the focus plane. These patterns are then used to retrieve the full-aperture wavefront through an optimization phase retrieval algorithm. The relationship between the optimization parameters of moving subaperture and the support information is given. Then expressions for the gradient of an error metric with respect to the optimization parameters are presented. An experimental setup used to measure the reflected wavefront of a concave mirror by employing this method is arranged. The experimental results indicate that the phase retrieval results are in excellent agreement with that obtained with interferometer, which confirms the validity of this method. [Copyright &y& Elsevier]

Subaperture stitching test in combination of the reconfigurable optical null we proposed recently provides flexible solutions to various surfaces including convex aspheres and even aspheres of large aperture. However it is challenging for the stitching optimization to get the real surface error because the surface error is strongly coupled with misalignment-induced aberrations in near-null subaperture measurements. Aiming at this challenge, we first figure out the property of aberrations induced by misalignment of optical null or test surface. It shows that identical misalignment of the optical null introduces nearly identical aberrations to subapertures with different off-axis distances, while misalignment of the test surface introduces little aberrations to the central subaperture. The stitching algorithm is then proposed with focus on decoupling surface error and induced aberrations. The major step is to calibrate out the effect of misaligned near-null optics before stitching optimization by using the central subaperture measurement. We also present the through-the-back null test for the purpose of cross test. The axial distance is precisely monitored by a low coherence interferometer, which enables accurate determination of the spherical aberration component of surface error. Final experimental results show consistent spherical aberration obtained by stitching test and by the through-the-back null test. It is a big step towards instrumentation of subaperture stitching test for aspheres with rather big amount of misalignments in surface metrology practice. [ABSTRACT FROM AUTHOR]

Abstract: Subaperture stitching interferometry combined with a test plate is attractive for testing large convex aspheres, but the stitching algorithm is challenging because the aberrations induced by misaligned test surface or test plate are coupled with the surface figure. By relating the subaperture configuration to the overlapping deviations through ray trace and coordinate transformation, the subaperture misalignment is optimally recognized and corrected to give a minimal overlapping inconsistency in an iterative way. Allowing for misaligned test plate, we decompose the induced aberrations into three parts which are corrected by the stitching algorithm, removed in the form of the Zernike polynomials and left uncorrected as residuals. Finally we present simulation results of testing a convex aspheric mirror with a computer generated hologram which shows the algorithm successfully retrieves the surface figure with the test mirror or the hologram misaligned. [Copyright &y& Elsevier]