An engineering approach for examining crack growth and stability in flawed structures

Abstract

This paper summarises progress made in two research programmes, sponsored by the Electric Power Research Institute (EPRI), to identify viable parameters for characterising crack initiation and continued extension. An engineering/design methodology, based on these parameters, for the assessment of crack growth and instability in engineering structures which are stressed beyond the regime of applicability of linear elastic fracture mechanics is developed in this paper. The ultimate goal in the development of such a methodology is to establish an improved basis for analysing the effect of flaws (postulated or detected) on the safety margins of pressure boundary components of light water-cooled type nuclear steam supply systems. The methodology can also be employed for structural integrity analyses of other engineering components. Extensive experimental and analytical investigations undertaken to evaluate potential criteria for crack initiation and growth and the selection of the final criteria for analysing crack growth and stability in flawed structures are summarised. The experimental and analytical results obtained to date suggest that parameters based on the J- integral and the crack tip opening displacement, δ, are the most promising. This is not surprising since, from a theoretical basis, the two approaches are similar if certain conditions are met. An engineering/design approach for the assessment of crack growth and instability in flawed structures is outlined. The approach exploits the consequences of J- controlled crack growth—the J resistance ( J R ) curve is a material property and crack driving forces can be determined from deformation plasticity analyses. Crack driving forces for the complete range of elastic-plastic deformation are obtained from a simple estimation scheme. The basic elements of the estimation scheme are the linear elastic solutions and the fully plastic solutions for the relevant crack configuration; the latter solutions are catalogued in a plastic fracture handbook. Crack-driving force diagrams, together with the J R curve, are employed to construct stability diagrams and predict the load deformation and crack growth behaviour of several crack geometries. The predictions are in good agreement with experimental data and full-blown numerical crack growth calculations. Using the driving force computed from the estimation scheme failure assessment diagrams can also be constructed from J- controlled growth. The shape and position of the failure line depend on the crack configuration and material properties. Relative merits and difficulties associated with the J or δ resistance curve approach for treating stable crack growth and fracture instability in structural components are also discussed. In applications involving relatively small amounts of crack extension, the J resistance approach assuming J- controlled growth has significant advantages, and the fracture predictions are in good agreement with actual test data. For larger amounts of crack extension, the situation is less certain. The limited studies carried out under non-J- controlled conditions suggest that the predictions from the engineering/design methodology will be conservative.