Application of Fracture Mechanics Method to Offshore Structural Crack Instability Analysis

Abstract

ABSTRACT A comprehensive comparison study has been performed to establish the limits of applicability and quantify the levels of conservatism in two crack instability analysis methods. These are the COD design curve and the R6 failure assessment diagram methods. The J-integral crack instability analysis solutions were used as the reference. The results of the study show that the R6 failure assessment diagram method with and without crack extension provides well-behaving solutions. The results using the British Standard Institution's published document procedures for the COD design curve method show a significant amount of un conservative solutions with a large scatter band. A new approach to improve the COD design curve method solution is developed, and its validity is demonstrated. 1. INTRODUCTION Fracture mechanics methods are becoming an important tool to perform an engineering critical assessment (EGA) of defects in offshore structures. The main objective of the EGA analysis is to establish the maximum tolerable defect size which would not compromise the service requirement. This can be achieved by ensuring that the fatigue life of joints with defects is greater than the required service life and that any defect does not attain a size to cause either crack instability or plastic collapse. There are several fracture mechanics methods that have been proposed to establish conditions of crack instability. These methods are the crack opening displacement (COD), failure assessment diagram (R6), and J-integral crack instability procedures. The COD method is the one that is extensively used by the offshore industry. In a previous paper by the authors,1 comparisons between the three methods were performed to establish the limits of applicability and quantify the levels of conservatism in both the COD and R6 methods. The paper reported several cases where apparent lack of conservatism was observed in the COD results. This paper presents an extension of the previous study by extensive analysis of different load cases. Analysis of some of these cases was only possible by the recent publication of their J-integra1 solutions.2 Also, a recently developed R6 curve for hardening materials that were made available to the authors allowed the extension of the previous R6 solutions. The paper also developed several modifications to the current COD analysis approach to ensure its reliability for EGA analysis. In this study, the J-integral results were used as referee because, as was discussed in the previous work, it represents the one with the most rigorous analysis procedures. 2. GEOMETRY AND MATERIAL For fair comparison between the three considered fracture mechanics methods, only flaw geometries that have known J-integral solutions, plastic collapse loads, and stress intensity factor solutions were selected. Also, consistency was maintained among the used material properties for all the analyzed cases. The material properties were selected from a data base3,4 created for ductile tearing instability studies for A508 Class A steel, which has similar characteristics to offshore structural steels.