A new experimental method to study the influence of welding residual stresses on fatigue crack propagation

Abstract

This paper presents a study on the influence of welding residual stresses (RS) on fatigue crack propagation rate (FCPR) in mode I. The objective of this work is to develop a novel methodology that allows a variation of a RS field in the studied specimen while keeping constant all other variables influencing FCPR. This led to the development of a novel specimen geometry, named CT-RES, in which RS are introduced by weld bead deposition far from the region in which fatigue crack propagation (FCP) occurs. As a consequence, the effect of factors influencing FCPR other than RS such as microstructural changes or plastic deformation, often introduced by welding processes, can be avoided. The welding RS introduced in the CT-RES specimen were determined by the contour method and the weight functions method was used to calculate the stress intensity factor (SIF), Kres, resulting from the RS as the fatigue crack propagates into the specimen. The evolution of cyclic stress ratio at the crack tip, Rlocal, was then computed from Kres to quantify the influence of RS on the cyclic stress ratio. The results show that for a given stress intensity range, dK, the FCPR of the welded geometry with fixed externally low R ratio (R = 0.1), but constantly increasing Rlocal, is the same as for the as-machined geometry without RS, solicited at high cyclic stress ratio (R = 0.7). These observations partially validate the BS7910 standard philosophy in which the remaining life of a flawed structure in presence of tensile RS is calculated from a high cyclic stress ratio (R > 0.5) crack propagation curve to eliminate crack closure effects.