Comparison of contact and non-contact reverse engineering techniques for free-form surfaces with coordinate measuring machine and 3D laser

• Main Author: Abd. Manaf, Ismanizam
• Format: Thesis
• Language: English
• Published: 2009
• Online Access: http://psasir.upm.edu.my/id/eprint/51548/1/FK%202009%20117RR.pdf; http://psasir.upm.edu.my/id/eprint/51548/
• Institution: Universiti Putra Malaysia

Application of reverse engineering (RE) technology enables creation of Computer Aided Design (CAD) models of existing product/part by capturing the surfaces data using either contact or non-contact methods. RE process consists of four main stages from part digitization, data editing, surface generation to CAD model construction. Many researches in RE applications show CAD model of an existing part was successfully reconstructed by using CMM or 3D laser scanner, but the researches need additional task in editing stage such as developed extra programme to reduce noises and errors. Additional task in editing stage can be eliminated if part digitization stage is carried out properly since surface measurement is the first priority in RE process. In this research, a free-form (FF) surface of an automotive component was generated using contact and non contact RE methods. Four scanning techniques used with two types of CMM namely Global Status Brown and Sharpe for point to point scanning and UMESS UMM550 Carl Zeiss for line scanning and two types of 3D laser scanners such as Cyberware 3D Rapid Digitizer for automated scanning and Minolta 3D Laser Scanner for manual scanning. The digitization strategies, scanned result and CAD models generated using both methods are then analyzed and compared. Digitization parameters for contact methods are tips radius,minimum angle, step width and scanning speed while for non-contact method are rotation sense, distance, laser density, rotation view angle,scanning mode, width, pitch and scanning angles. Point to point scanning technique with large amount of scanned data volume is most time consuming as measured while conducting an experiment that is more than 10 hours need to complete scanning time from CMM measurement to data processing. In line scanning, although small data volume gathered, it takes 3 hours 25 minutes to perform complete scanned part surface. In automated scanning required 40 minutes inclusive of processing time while manual scanning takes 1 hour 30 minutes to perform a complete scanned. It means in term of time taken, laser scan technique may reduce the RE processing time compared to tactile probing. However, while comparing to editing and modeling task, also the modeling skill required, data captured from manual scanning technique is most demanding. Point to point scanning produced best quality result compared to other technique since a prototype model is successfully produced with 1% percentage error. Most unacceptable result is manual scanning because too many missing surfaces to the surface scanned result.