Available creep and creep-fatigue data on type 304 stainless steel are re-examined in the light of recently generated basic cavitation data on the same material. This basic study has shown creep damage to be a highly inhomogeneous phenomenon, both in space and in time. A small fraction of boundaries are so intensely cavitated by about 10–25 percent of life that for all practical purposes they can be considered cracked. Other similar boundaries, that are similarly oriented with respect to the tensile stress direction, however, are almost devoid of cavities at the end of life. The early nucleation of cracks by creep mechanism has a significant influence on creep-fatigue interactions, particularly for tests at low strain ranges. A simple ductility exhaustion model appears to be able to account for the early crack initiation. A hardness-modified ductility approach appears to provide an upper bound to the ductility displayed by all tests and correlates data for long hold-time fatigue tests at low strain range. Tests at higher strain ranges and/or shorter hold times do not achieve their full potential ductility because their lives are cut short by crack propagation. Their lives can be predicted empirically by an enhanced fatigue crack growth equation in the presence of cavitation damage.
Creep-fatigue life prediction for different heats of Type 304 stainless steel by linear-damage rule, strain-range partitioning method, and damage-rate approach