Prediction of high-cycle fatigue life reliability of aluminum cast alloy from statistical characteristics of defects at meso-scale

In this paper, a growth rate law of a small fatigue crack of an arbitrary metal and a method for predicting a size distribution of spheroidal defects cut by a specimen are proposed. And then, by combining the crack growth rate law and the prediction method for the surface defect size distribution, a method for predicting a high-cycle fatigue life reliability of a metal with defects is proposed. Rotating bending fatigue tests of plain specimens with/without a small blind hole (0.3 mm in diameter, 0.15 mm in depth) are carried out on an aluminum cast alloy JIS AC4B-T6 with eutectic Si, Fe compounds and porosities. By comparing analyzed results with experimental ones, the validity of the present method is examined.