| Summary: | This thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University London Crack tip constraint is a significant issue in engineering components' design and repair decisions. The main reason is that the use of plane strain fracture toughness derived from deeply cracked and thick section specimens in structural integrity assessments is generally considered conservative. Generally, real components contain shallow cracks and thin sections that lead to significant variability in effective toughness due to loss of crack tip constraint. The overall objective of this research was therefore to develop test methods used in fracture assessment of high strength steel components operating in extreme low temperatures, by incorporating a reduction in the inherent conservatisms in assessment procedures, particularly BS 7910. The work contained both experimental and numerical analyses of pin-loaded single edge notched tension (SENT) and three-point single edge notched bend (SENB) specimens at room and low (-120℃) temperatures under different constraint conditions. Finite element analyses (FEA) of steel pipelines containing surface flaws subjected to both internal pressure and bending were also conducted. Further, a method was proposed, based on the combined use of digital image correlation (DIC) to measure full-field displacements at room temperature and a finite element approach to extract the strain energy release rate of shallow cracks. A finite element model with imported DIC-measured full-field displacements acting as boundary conditions is solved and the J-integral was computed. Additional preliminary testing was carried out on aluminium 5083 coupons using X-ray computed tomography, intended for digital volume correlation (DVC) analysis. The experimental and numerical results showed that a decrease in temperature leads to a reduction in fracture toughness and therefore, susceptibility to brittle failure. The numerical analyses also showed that loss of constraint in shallow and thin ...
|
|---|