The viability of the KIa procedure for predicting crack arrest under displacement controlled loadings

The ASME Code procedure ( 1 ) for predicting crack arrest in nuclear reactor pressure vessels is based on a static linear elastic fracture mechanics (LEFM) analysis: a crack arrests when the crack tip static stress intensity factor K I ST falls below a value K Ia , which is assumed to be a material property, and is referred to as the arrest toughness. This approach makes exact predictions for the case where a semi-infinite crack propagates in an unbounded solid due to the application of time independent loads ( 2 ) for K Ia can be then equated with K Id = K Id (v → 0), where K Id (v) is the dynamic fracture toughness for a crack velocity v. However, for other more general situations, the usefulness of the K Ia procedure depends on its ability to make arrest predictions that are sufficiently accurate for practical purposes. It is generally recognized that the procedure is more appropriate for use in situations where (a) wave reflection effects are minimal and (b) when the crack jump length is small. As regards this second point, by analysing the case where an edge crack propagates in a semi-infinite solid due to constant forces being applied at the crack mouth, the author has shown ( 3 ) that the K Ia procedure should be fairly accurate (to within ~20%) in predicting the crack jump length, if the crack jump length is less than twice the initial crack size. The present paper extends the earlier work to the case where fixed displacements are applied to the faces of a crack. It is shown that the K Ia procedure can seriously underpredict the crack jump length, and the results are therefore in sharp contrast to those for the fixed load situation. (3)