Creep life predicition of P91 Steel from stress relaxation tests

Abstract

Reliable creep life prediction of engineering materials needs long hours of testing over a range of stress and temperature. There have been considerable efforts to develop alternative methods to cut down the test duration. Short term stress relaxation test (SRT) is one of the possible options. Recent works suggest that if stress relaxation tests are done at a stress a little higher than the yield stress of the material the stress drops within a span of 10 to 20 hours to a level where most of the creep tests are done. From such tests material parameters describing the stress/temperature dependence of the minimum creep rate of the material can be established. Since the minimum creep rate has a definite correlation with the time to rupture this could possibly be a method worth exploring for quicker estimation of creep life of engineering materials. However there is very little experimental data to establish the correlation between the parameters like stress exponent (n) and activation energy (Q) derived from stress relaxation tests with those obtained directly from the conventional creep tests. The work presents creep as well as stress relaxation data on P91 steel obtained from tests performed on specimens made from the same stock of steel. The Q & n obtained from stress relaxation tests were found to be a function of the total strain at which the tests were done. These were found to be 355kJ/mol and 7.3 respectively at a total strain of 1.8%. The Q & n obtained from the creep tests are 597kJ/mol and 9.9 respectively. This suggests that stress relaxation is likely to overestimate the creep rate at a given stress and temperature.