In many structural flaw evaluations, frequently it is necessary to predict the maximum load where a ductile tearing failure mode is expected. In such a calculation, large ductile crack growth needs to be accounted for to predict the maximum load, especially for large structures and/or lower toughness materials. One difficulty in such an evaluation is that the amount of crack growth obtainable in C(T) specimens that can be machined from the material (e.g., from a pipe) is relatively small. Because of this, it is necessary to extrapolate the J-R curves to the larger amount of crack growth. No generally accepted method exists for such extrapolations.
This paper discusses methods used to evaluate J-R curves extrapolation techniques. As part of this study, several materials were evaluated using C(T) specimens of various sizes but with the same thickness. This is frequently the case for piping, whereas for a pressure vessel or other heavy-wall structure the specimens will frequently not even be the same thickness as the structure. The results in this study showed more geometry effects with the JM-R curve than expected.