Fatigue-crack growth in two aluminum alloys and crack-closure analyses under constant-amplitude and spectrum loading

Fatigue-crack-growth tests were conducted on compact, C(T), and middle-crack tension, M(T), specimens made of 7075-T6 and 7249-T76511 aluminum alloys. The 7075 tests were conducted on thin-sheet (2-mm thick) material; whereas, the 7249 tests were conducted on three thicknesses (2, 3.2 and 6.35-mm). The thin-sheet 7075 alloy did not have residual stresses, but the 7249 alloy (B = 2 mm) had low-level ( < 10 MPa) residual stresses from measurements. However, the other two 7249 alloy thicknesses were residual-stress free based on measurements. These tests were conducted to generate fatigue-crack-growth-rate data from threshold to near fracture at two load ratios (R = 0.1 and 0.7). Two methods were used to generate near threshold data on C(T) specimens: (1) compression pre-cracking constant-amplitude (CPCA) and (2) compression pre-cracking load-reduction (CPLR). Constant-amplitude tests were conducted on the M(T) specimens at the same two load ratios. Crack-closure measurements using a compliance method were made on some of the low R tests and a crack-closure model were used to develop an effective stress-intensity factor range against rate relation using a constant constraint factor (alpha = 1.85). The constraint factor accounts for three-dimensional stress-state effects on plastic yielding at crack tips and crack-closure behavior. Simulated aircraft wing spectrum tests were conducted on the M(T) specimens using two modified full-scale-fatigue-test (FSFT) spectra used on the P-3C aircraft. Tensile pre-cracking was used on all of the M(T) specimens, except one 3.2-mm thick 7249 specimen, before the spectra were applied. The same spectrum sequence was applied, but one had tension-compression loads, whereas, the other only tension-tension loads. The spectrum tests were used to calibrate the constraint-loss regime (plane-strain to plane-stress; alpha = 1.85-1.0) behavior for the FASTRAN strip-yield model. Comparisons were made between the spectrum tests and calculations made with the FASTRAN life-prediction code; and the calculated crack-growth lives were generally within +/- 20% of the test results.