The Open UniversitySkip to content

Advanced experimental methods for the characterization of welded structures

Kartal, Mehmet Emin (2008). Advanced experimental methods for the characterization of welded structures. PhD thesis The Open University.

Full text available as:
PDF (Version of Record) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (21MB) | Preview
Google Scholar: Look up in Google Scholar


Welding is one of the most prevalent techniques for mechanical fastening of metals. Recent developments in welding technology have led to welding techniques being more readily employed in safety-critical engineering structures. Since the existence of residual stresses and material property variation around welds affects the mechanical performance and thereby structural integrity, it is essential to improve our knowledge in understanding and modelling the mechanical response of the welded structures. The present work focuses on mechanical characterizations of such structures. This work can be broadly classified into two parts; the first part investigates the residual stress distribution in welded specimens of different metals and the second part presents investigations of mechanical properties in welded specimen using full field optical techniques.

A newly invented destructive technique for residual stress measurement, the contour method, was used for the investigations of the residual stress in welded joints in this study. The principle of the contour method is based on a variation of Bueckner's superposition theory. By means of a single straight cut, the 2D residual stress component normal to the region of interest can be determined. In this work, first the numerical simulations of the contour method using two and three dimensional bodies have been demonstrated. The contour method was then applied to one-pass and three-pass groove weld specimens and the results obtained from the contour method were compared to those obtained by the neutron diffraction strain measurement technique.

The capability of the contour method to measure multiple residual stress components was also investigated in this project. A recent development of the contour method of stress measurement, the multi-axial contour method, permits measurement of the 3D residual stress distribution in a body, based on the assumption that the residual stresses are due to an inelastic misfit strain (eigenstrain) that does not change when a sample containing residual stresses is sectioned. The eigenstrain is derived from measured displacements due to residual stress relaxation when the specimen is sectioned. By carrying out multiple cuts, the full residual stress tensor in a continuously processed body can then be determined. In this study, finite element simulations of the technique were carried out to verify the method numerically. The method was then used to determine the residual stresses in a VPP A-welded sample, and the results were validated by neutron diffraction measurements.

As part of the characterization of the welded structures, a study was undertaken to develop a method of extracting local mechanical properties from weld metal by strain mapping using the digital image correlation (DIC) technique. The feasibility of determining local stress-strain behaviour in the weld zone of a 316H stainless steel pipe with a girth weld was investigated by tensile tests on miniature and standard tensile test specimens. In addition, electron speckle pattern interferometry (ESPI) was utilized to obtain the full-field strain maps of a standard tensile specimen during loading and compared to those obtained in the same specimen by digital image correlation in order to verify the DIC measurements.

Item Type: Thesis (PhD)
Copyright Holders: 2008 The Author
Academic Unit/School: Faculty of Science, Technology, Engineering and Mathematics (STEM) > Engineering and Innovation
Item ID: 59992
Depositing User: ORO Import
Date Deposited: 22 Mar 2019 15:56
Last Modified: 30 Mar 2019 20:36
Share this page:

Download history for this item

These details should be considered as only a guide to the number of downloads performed manually. Algorithmic methods have been applied in an attempt to remove automated downloads from the displayed statistics but no guarantee can be made as to the accuracy of the figures.

Actions (login may be required)

Policies | Disclaimer

© The Open University   contact the OU