Development of in-situ measurement systems for capturing free-form and mirror finish surface information

• Main Author: Fu, Shaowei
• Other Authors: Tegoeh Tjahjowidodo
• Format: Theses and Dissertations
• Language: English
• Published: 2019
• Subjects: Engineering::Mechanical engineering
• Online Access: https://hdl.handle.net/10356/103297; http://hdl.handle.net/10220/49983
• Institution: Nanyang Technological University

Surface finish usually affects the quality, performance and lifetime of engineering products. With the emerging advanced manufacturing technologies developed in last decades, it is also critical to evaluate and control surface characteristics of various products in engineering and manufacturing fields. Currently, the surface texture of a workpiece is inspected by post-process methods where the finished workpiece is removed from the machine and sent to a metrology laboratory for measurement. The measurement process requires precise alignment which takes considerable time to achieve. These off-line instruments are also difficult to be integrated into an automatic inspection system. In order to improve the manufacturing accuracy and quality control of the advanced manufactured products, an accurate and in-situ surface measurement has become increasingly important. This research aims to develop in-situ surface inspection solutions for various advanced manufacturing surfaces such as mirror finishing, additive manufacturing and free-form surfaces. Firstly, an automated surface inspection system is developed using fringe pattern illumination method, the method allows for an in-situ surface roughness measurement and through a dedicated surface roughness calculation algorithm, an on-line data acquisition and processing on mirror finish surface can be carried out. In addition, the developed surface inspection system is also able to detect the directions of machining mark and measure the surface uniformity. For defect detection on mirror finish surfaces, a developed system based on light scattering model is able to perform in-situ surface defect detection. Secondly, a displacement measurement system based on image grating technique is developed to improve the robustness and flexibility for displacement measurement using conventional optical encoder and can be integrated into various surface metrology instruments. The performance of the proposed image grating system is compared with a high-accuracy heterodyne laser interferometer. A measurement error compensation method based on optical distortion model is evaluated both theoretically and experimentally. Then, the error compensation method is implemented in the developed image grating system to reduce the optical distortion error. The experimental result shows that the measurement error is less than 0.35 µm within 50 mm measurement range. In addition, a repeatability study of the proposed system is conducted, and the residual errors are within ±0.15 µm. Finally, a surface finish measurement system based on laser confocal method is developed to measure the surface roughness of the different machined and additive manufacturing surfaces. To reduce the inherent disadvantages of confocal sensor such as scattering noise at sharp peaks and background noise caused by specular reflection from the optical elements, the developed system has been calibrated and a linear correction factor has been applied. The proposed system was integrated into a robotic system with the measurement distance and angle adjusted during measurement based on a CAD model of the workpiece in question. Experimental data confirms the capability of this system to measure the surface roughness within the Ra range of 0.2–7 μm, with a relative accuracy of 5%. To overcome the vertical scanning limitation of the laser confocal sensor, an adaptive surface tracing algorithm is developed and tested to measure free-form surfaces. The developed image grating system is also integrated in to the adaptive surface tracing system for validation. Experimental result demonstrates the capability of surface tracing system to measure the free-form surface with surface height deviation more than 8 mm and the surface texture and profile measurement errors can be controlled within 0.03 µm.