Development of Advanced Methods for Quantifying Fracture Toughness Properties in the Presence of Residual Stresses

  • Robert Hurlston

    Student thesis: Doctor of Engineering


    Welding is an essential process in many industries for both the production and repair of engineering plant, notably pressure vessels and piping. However, welding processes cause large magnitudes of residual stress to be induced within the structure. Residual stress can be defined as a stress that exists in a material when it is under no primary loading. Whilst residual stresses can be reduced by post weld heat treatment, such treatments are not always possible, and so high residual stresses can remain in serviceThe current methodology for evaluating fracture toughness from specimens, particularly if these contain weld residual stresses is presented in BS7448-1997. This method relies on the assumption that the effect of residual stress on fracture toughness measurements can be negated by the application of a local compression, to the ligament ahead of the pre-crack in the test specimen. Recent research has investigated the validity of this assumption. The results suggest that, far from being removed, the residual stresses are modified or even enhanced via local compression. This can reduce the value of measured fracture toughness below its true value. In order to ensure the validity of fracture toughness measurements in materials that contain residual stress, a more robust method for its quantification is developed.The aim of this project was to extend current understanding regarding the magnitude and distribution of residual stresses retained in standard fracture mechanics specimens removed from welds and the consequent effects of these stresses on measured fracture toughness, both in terms of the crack driving force and crack-tip constraint. Furthermore, the project aimed to derive improved methods for the quantification of valid values of fracture toughness from laboratory specimens containing residual stresses. This was achieved via a combination of analytical and experimental work.The effect of specimen extraction on the level of retained residual stress in specimens extracted from non stress-relieved welds was investigated using parametric finite element analyses. Simplified methods to quantify the levels of residual stresses in fracture mechanics specimens removed from welds and their significance, in terms of contribution to crack driving force, are proposed.The influence of residual stresses on the measured fracture toughness properties of ferritic pressure vessel steel, tested in the cleavage fracture regime, has also been studied. A refined method of out-of-plane compression was devised and used to generate significant residual stresses in three-point bend specimens. This method was then used experimentally, alongside supporting elastic-plastic analyses, to quantify the effects of the residual stresses on fracture toughness in terms of both crack driving force and crack-tip constraint in geometrically high and low constraint specimens. A method whereby fracture toughness data, obtained from specimens containing residual stresses, can be corrected to provide valid fracture toughness properties using constraint based fracture mechanics alongside a simple fracture model has been proposed. The main conclusions from the work are as follows.Significant weld residual stresses have been shown to be retained in certain laboratory specimens post extraction from non stress-relieved welds. The magnitude and distribution of retained residual stress has been shown to be dependant on: • Material yield and flow properties• Specimen size; where larger specimens are more likely to retain significant levels of residual stress than smaller specimens• Specimen type; either compact tension (CT) or single edge notched bend (SENB), where there is a tendency for specimens to retain higher relative levels of residual stress in the directions of their largest dimensions; i.e. bend specimens retain more residual stress along their length than CT specimens and CT specimens retain more residual stress across their width than bend specimens• Extra
    Date of Award1 Aug 2012
    Original languageEnglish
    Awarding Institution
    • The University of Manchester
    SupervisorAndrew Sherry (Supervisor)


    • Cleavage Fracture
    • Residual Stress
    • Welding

    Cite this