Comments on: “Understanding the Larson-Miller parameter”, by F. T. Furillo, S. Purushothaman and J. K. Tien (1)

Abstract

The Larson--Miller parameter has been a useful tool in the handling of creep rupture data. Essentially, this is because, as an empirical scheme, it has so often been successful at correlating such data for a wide range of experimental conditions. Its use has proved invaluable to studies of nuclear fuel cladding embrittlement in reactor environments under diverse loading conditions. The kind of master plot that is generated through the use of this parameter is very helpful for predicting low stress-long time rupture behavior. The basic premise behind the success of the Larson--Miller method lies in the fortuitous elimination of one of the two independent variables that can be controlled during a creep test: the absolute temperature T and the applied stress sigma. When a test sample is stressed at temperature, it will fail after a time t/sub r/, depending upon the values of some unspecified, and usually unknown, material parameters. Larson and Miller (2) presented a simple expression involving these creep rupture variables which is only a function of the applied stress: LMP = T(C+log tr) = f(sigma). The quantity C is specified as a material constant, but is roughly 20 for a wide variety of commercial alloys. The eliminationmore » of the temperature dependence allows one to construct master rupture curves. This is the power of the Larson--Miller method. Once the function F(sigma) is determined graphically, regardless of its algebraic form, the stress is needed in order to know LMP. Knowing LMP, one can calculate t/sub r/ as a function of temperature. The paper under discussion does not serve to further an understanding of this useful empirical parameter. In fact, their derivation leads to a parameter similar in form to Larson and Miller's, except that it is a function of both stress and temperature, thereby defeating the purpose of the method.« less