Comparison of creep strength and intrinsic ductility for serviced and reheat treated T91 steel based on stress relaxation testing

Abstract

High precision stress relaxation tests (SRT) at four temperatures were conducted on T91 (9%Cr) steel after extended boiler service, and also after re-heat treatment. Relative differences in creep strength, measured over five decades in strain rate were dependent on test temperature. Using an established correlation between strain rate sensitivity and elongation at failure, intrinsic ductility values as a function of stress and test temperature were determined. The general trend of a minimum in ductility in terms of stress or strain rate was consistent with long term creep rupture data on T91, and with literature data on alloy steels. However, the precision and repeatability of the SRT analysis contrasted with the appreciable scatter and heat to heat variation in traditional testing. It is argued that the current creep strength evaluation based on the nearly constant state measurement from the SRT test is superior to the measurement of stress dependence of minimum creep rate in traditional creep rupture testing. The complexity of primary creep in laboratory testing, which may not be significant at operating stresses where loading strains may be fully recoverable (anelastic), does not apply to the SRT. Since very low strain rates are achieved in a one day test, the procedures for setting of design allowables and design analysis based on the SRT data should not be significantly different from current practice. This technique offers accelerated alloy development and optimisation for creep strength and also ductility, and hence resistance to notch sensitivity.