Creep-fatigue damage accumulation and interaction diagram based on metallographic interpretation of mechanisms

The deleterious effect of creep damage upon fatigue endurance in elevated temperature low cycle fatigue is well established and has become known as the 'creep-fatigue interaction'. This has been identified as a narrow region where a transgranular crack becomes intergranular as creep damage develops. In other cases transgranular fatigue cracks and intergranular creep cavitation damage to some extent can take place separately ('competitive' mode) so that interaction is less pronounced. These phenomena have been incorporated into 'damage interaction diagrams' which feature in many high temperature Code cases (ASME, R5, RCC-MR etc.). Fractional creep and fatigue damage are considered separately, and failure is conceded when their sum is unity ('additive' mode). For some materials the sum is set to a value <1. It appears that purely empirical arguments have been used to arrive at total damage values <1. It is shown that such deviations from linearity can be simply predicted, assuming that creep damage accumulates uniformly during a given test. For maximum interaction, the resulting assessment curve is shown to pass through the points (total fatigue damage = 0.4; total creep damage = 0.4), or very nearly. The influence of fatigue upon creep damage is less well known. Examples are provided, and when included in the calculation, produce a more damaging curve passing through the point (0.33, 0.33). By use of 'interaction coefficients' less damaging curves for the 'competitive' model can be produced. The calculation may be simplified by use of an effective fatigue damage cycle number for use as a reference. Based on metallographic examples from power plant operation and laboratory testing, the 'competitive', 'additive' and 'true interaction' modes of creep and fatigue damage are identified. These can be related to practical applications of base load operation, fast start-up and shut-down procedures, slow start-up and shut-down procedures etc. Components might operate under base load conditions (creep) followed by temperature cycling (fatigue). These sequential events may not be interactive (as has been demonstrated in the laboratory) and thus use of the current assessment diagram would be pessimistic. Comment is made on a proposed damage diagram for the advanced ferritic steels, which is highly conservative and does not reflect their good performance in service and laboratory tests.