Small cracks can grow at ΔK levels previously thought safe (that is, ΔK < ΔKΔKth). It follows that ΔKth values, measured according to ASTM Test Method for Measurements of Fatigue Crack Growth Rates (E 647) procedures and associated with long-crack test specimens, may lead to nonconservative lifetime estimates of a component that contains very small cracks. One major reason for this discrepancy may be traced to the fact that different closure levels are found at the same applied ΔK level for long and short cracks, respectively. Much effort has been given to the generation of conservative fatigue crack propagation (FCP) data through the development of large quantities of short-crack data. Unfortunately, the generation of these data are time-consuming, tedious, and subject to considerable amounts of scatter.
An alternative test method, based on the use of standard-sized samples and employing established automated data acquisition procedures, has been identified. This method is based on maintaining a constant maximum stress intensity level (Kcmax) during the ΔK-decreasing test procedure. By maintaining a constant Kmax value, mean stress and associated R levels are found to increase markedly as ΔK decreases. The development of these high levels of mean stress and R ratios produces a long crack with negligible crack closure, which provides an upper bound estimate for the behavior of short cracks. Kcmax FCP data for several aluminum, iron, and nickel-based alloys have been generated and demonstrate the utility of the Kcmax test methodology as a simple and convenient means to obtain upper bound estimates of short-crack behavior in these structural alloy systems.