Volumetric defect analysis in friction stir welding based on three dimensional reconstructed images

Abstract

Friction Stir Welding (FSW) is a relatively new technology in welding. The ultimate aim of any welding process is to achieve a defect-free weld. Several defects during FSW can be attributed to various factors and are required to be examined and eliminated if required. The exact severity of a surface defect in FSW cannot be determined by a 2D image, which does not provide any information about the depth of the defect. Thus, this paper aims towards the comprehensive analysis of degradation caused over the weld strength due to surface defect, based on its 3D reconstructed depth map obtained from 3D Optical Microscope. This research work proposed a novel algorithm to determine certain 3D geometrical features like effective length, effective width, average depth and effective volume for giving the severity of surface defects with a precision of around micrometres scale. The depth data were also used to determine the surface textural features like average roughness and RMS roughness. The results obtained by this technique were also experimentally verified by the conventional surface roughness tester machine.