The influence of high R ratio on notched fatigue behaviour of 1045 steel with three different heat treatments

Three 1045 steels with hardness levels of Rc=10, 37, and 50 were tested under cyclic axial load control conditions using mildly notched specimens, Kt=1.65, at high R ratios of 0.8 and 0.9. The notched ultimate tensile strengths, Sun, for the three steels were greater than the unnotched ultimate tensile strengths, Su. This allowed values of nominal maximum stress, Smax, and mean stress, Sm, to exceed Su in most tests. Thus, fatigue limits based on Smax were higher than Su in 5 of the 6 test conditions. S–Nf curves were very flat in 5 of the 6 test conditions with appreciable scatter. With Smax and Sm>Su in most tests, usual S–Nf fatigue life models involving Su could not be used. Replacing Su with Sun, allowed calculations, but these were completely inaccurate. Local strain-life, ε–Nf, models were also completely inaccurate for these high R ratios. Rc=10 specimens failed by cyclic creep/ratcheting from internal microvoid coelescence and not from fatigue. Rc=37 specimens failed by fatigue from surface thumbnail cracks, but were influenced by cyclic creep/ratcheting. Rc=50 specimens failed by brittle fracture from minute surface fatigue cracking without cyclic creep/ratcheting. In design situations at long life, usual S–Nf models with these materials and high R ratios would restrict Smax to levels well below the experimental fatigue limits resulting in very conservative results.