Reference-free beam-sampling (RFBS) methodology and system for optical freeform surface measurements

• Main Author: Guo, Wenjiang
• Other Authors: Chen I-Ming
• Format: Theses and Dissertations
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Manufacturing::Metrology
• Online Access: http://hdl.handle.net/10356/62336
• Institution: Nanyang Technological University

Freeform surfaces of optical quality are of huge demand in the precision and optical industry. Although six degree-of-freedom manufacturing technologies are available to fabricate such surfaces, the metrology infrastructure cannot simultaneously perform accurate, universal and non-contact large area measurement in a short time. The main challenges are the varying curvature across the surface together with the large dynamic range. The Shack-Hartmann wavefront sensing (SHWS) technique samples the wavefront by capturing the focal spot image of the surface, which is compared with the reference image to measure the wavefront slope. The wavefront is then revealed through reconstruction. SHWS technique has merits of area-based, non-contact, high accuracy, and insensitivity to vibration. It is used for flatness and aberration measurements, optical alignment, and adaptive optics. If applying the SHWS technique to measure freeform surfaces, the sampling is an issue as the focal spot image obtained is of poor quality, which degrades the measurement performance and many of the times even make it impossible to reveal the surface profile. Also, the reference image from a flat mirror is no more valid due to out of measurement range. Furthermore, the lateral resolution is too limited to detect the varying curvatures. Last but not least, the wavefront slope data is difficult, or even impossible, to be extracted because of image distortions. This dissertation designed and developed the reference-free beam-sampling (RFBS) methodology and system for freeform surface measurements. In the RFBS, the beam-sampling and the reference-free techniques have been devised and developed as the sensing mechanism. The method uses the beam itself as the optical probe to sample the surface. Through introducing a lateral disturbance to the modulated beam, the second order derivative of the surface is measured. The digital scanning technique is proposed and integrated into the sensing mechanism to enhance the lateral resolution. The dynamic windowing and the adaptive centroiding algorithms have been devised to extract the measurement raw data from the captured images, which is then reconstructed into the surface through the proposed 3D reconstruction technique. Freeform surfaces of various forms were measured. It is shown that tens nanometer height accuracy is achieved when measuring surfaces with over 1mm peak-to-valley value. In addition to the advantages of SHWS, no reference image is needed, no mechanical scanning is involved, and no pre-knowledge of the surface is required for the proposed method. Therefore, the RFBS system shows high potential for in-situ nm-level freeform surface measurements.