Creep-fatigue interaction rules for P91: MATTER – Deliverable D4.5

A number of methods for predicting the creep-fatigue life of P91 steel are presented and discussed. Beside the interaction diagram method established in the design codes, ASME Subsection NH and RCC-MR, and the ductility exhaustion modification (DE) established in BS-R5, a set of simpler models are considered. Application of different rules to creep-fatigue experiments on available P91 data with relatively short hold times showed that some of the interaction diagram methods are overly conservative and predicted extensive creep damage for all tests. The simplified models showed excellent performance and the two modified ductility based interaction diagram models (SMDE and EMDE) performed satisfactorily. The relative in-sensitivity to the creep deformation rate displayed by the modified ductility exhaustion models demonstrate a general advantage of these models against purely stress-based approach, particularly when considering the variability of material property and other uncertainties encountered when assessing actual components. However, because the creep damage is low in most tests with fatigue damage dominating and the expected negative impact of cyclic softening on creep damage, it is not possible to draw solid conclusions as to the likely accuracy and conservatism of these models for increased creep damage during longer dwells that will likely occur in plant. Also, a thorough description is given in the application of the RCC-MR and the ASME Subsection NH codes in the design calculations for Korean SFR heat exchangers.