MODELS OF THE TEMPERATURE DEPENDENCE AND SCATTER IN J-R AND J_(0.1) FOR FERRITIC REACTOR PRESSURE VESSEL STEELS

摘要

At the 2014 ASME Pressure Vessel and Piping Conference, these authors and others presented a paper that drew together a number of models describing the fracture toughness of ferritic reactor pressure vessel (RPV) steels. That paper summarized models of both the temperature dependence and scatter in a number of fracture toughness metrics (i.e., K_(Jc), K_(Ia), J_(Ic), and J_(0.1)). That paper also provided equations that quantify the interrelationships between these toughness metrics, and how these interrelationships are affected by hardening. Significantly, all of these models and interrelationships are linked via a single parameter: the Master Curve index temperature, T_e, which can be measured as described in ASTM Standard Test Method E1921. Work is currently underway within the ASME Section Ⅺ Working Group on Flaw Evaluation (WGFE) to develop a revision to Code Case N-830 that incorporates all of these models, and provides information on how to apply them in a flaw evaluation. As part of that work, an effort was initiated to augment these models by the addition of a model that can be used to predict the temperature variation of, and the scatter in, J-R curve behavior. A J-R curve model is also expected to support on-going WGFE efforts to in development of acceptance criteria for flaws in ferritic components operating in the upper shelf temperature range. The work presented in this paper provides a model of the J-R behavior of ferritic RPV steels. When combined with other fracture toughness models to be published in Code Case N-830-1, this model allows prediction of the mean J-R curve, confidence bounds on the mean, and the temperature dependence of J-R all based only on input of T_o. The J-R model described herein has equivalent or better accuracy to other models described in the literature, and generally has fewer fitting parameters than those other models. Because the full J-R curve is predicted, this model is also useful for prediction of J_(0.1).