senior engineer, Westinghouse Research and Development Center, Pittsburgh, Pa.
For quenched and tempered SAE 4340 steel, the inelastic strains measured during complex stress-strain histories are observed to be larger than would be estimated from constant amplitude test results. Ignoring this effect can result in nonconservative fatigue life predictions. The effect is accounted for by basing fatigue life calculations on inelastic strain. A stress versus inelastic strain relationship is determined from incremental step tests. The inelastic strain versus cycles to failure relationship is employed, and at short lives is determined from constant amplitude tests. At long lives, either the short life data can be extrapolated, or initial and periodic overstrains should be imposed during the constant amplitude tests. Actual and calculated lives are compared for a variety of complex history fatigue tests on unnotched axially loaded specimens.
Paper ID: JTE10016J