Fatigue-Crack Propagation Through a Measured Residual Stress Field in Alloy Steel

Fatigue crack-propagation tests were performed using 5 by 30 mm cross-section bend specimens of a nickel-chromium-molybdenum steel. The fatigue crack-propagation rate was determined from a group of stress-free specimens by measuring crack length on the specimen surfaces at intervals during cycling. Residual stress was produced in a second group of specimens by using a localized plastic deformation process. Resistance strain gages were first applied near one edge of each specimen along the line of intended crack growth. A series of 1-mm-deep plastic indentations was then made along the opposite edge of the specimen using a 25-mm-diameter pin. The strain gages provided a direct, accurate measure of the elastic, residual stress produced on one side of the specimen due to the local plastic deformation on the opposite side.

Measured crack-propagation rates in the specimens with residual stress are compared with rates in residual stress-free specimens. Crack-propagation rates are lower, as expected, near the edge of the specimen where the initial residual stress is compressive. Propagation rates remain lower even as the crack grows deeper into the specimen where the initial residual stress is tensile, which is not what would be expected from a simple superposition of stresses. However, an analysis involving the combination of the applied stress-intensity factor with that estimated from a redistribution of the residual stress in the specimens can account for the lower crack-propagation rates.