A life prediction method based on an analysis of crack coalescence in low-cycle fatigue of a smooth specimen.

The quantitative analysis of crack coalescence is one of the significant problems in low-cycle fatigue of a smooth specimen, since the coalescence of many distributed cracks reduces the fatigue life extremely. In this paper, an analytical procedure is proposed to estimate the crack propagation life for material with a considerable frequency of crack coalescences. Two-dimensional modeling is employed to describe the distribution of cracks and the conditions of crack coalescence. When the initial distribution of cracks is given, based on the modeling, the subsequent crack propagation life is calculated successively by using the growth law of long through-thickness cracks through an energy density parameter related to the J-integral range. The proposed method is applied to a practical problem in pure copper and the predicted results are in good agreement with the experimental data.