It has been identified that in-service carburization of austenitic stainless steels in the UK Advanced Gas Cooled Reactor (AGR) fleet can impact upon creep-fatigue crack initiation, which is assessed using the R5 Volume 2/3 assessment methodology.
Material properties for the carburized layer have been derived through testing of preconditioned specimens. These properties have been used to propose a simplified assessment methodology for the treatment of creep-fatigue crack initiation in carburized specimens. The methodology accounts for the negative impacts of carburization, such as the reduced creep ductility, but does not benefit from the potentially positive impacts, such as the increased creep deformation resistance.
At relatively low strain ranges, typically seen by plant components, these simplifications generally result in a very conservative creep-fatigue lifetime prediction. However, at short dwell times (in load control) at high strain ranges (in strain control) the methodology has the potential to be non-conservative when the stress mismatch between the carburised layer and bulk material is severe due to differing cyclic stress-strain properties. Therefore pragmatic limits of application must be applied to the simplified approach. This paper explores these limitations of the simplified assessment advice, the impact on plant assessments and how current research is looking at how less conservative assessment methodology can be developed.