| Induction hardened components rely heavily on both the hard martensite formed in the case during processing and the corresponding compressive residual stresses to improve fatigue performance of a component. In this study, 1045, 4145, and 1060 steel alloys were characterized and fatigue tested with cantilever bending in the as hot-rolled, normalized, or quenched and tempered conditions. The alloys were induction hardened to achieve either "low" or "high" case depths. Post-induction hardening characterization included residual stress depth profiles, microhardness traverses, and microstructure analysis in the as-processed and run-out at the endurance limit (1.5x107 cycles) conditions. The experimentally determined endurance limits were found to scale first by core microstructure with ferrite/pearlite microstructures exhibiting the lowest endurance limits followed by bainitic microstructures and finally tempered martensite microstructures. Within the tempered martensite core microstructures of the 1045 and 4145 conditions, the endurance limit was found to increase by increasing the core hardness. Fatigue testing elucidated the competing crack nucleation mechanisms (i.e. microstructural feature versus inclusion controlled crack nucleation). Fatigue cracks only nucleated from microstructural features in the 1045 normalized low and high case depth condition as well as the 4145 hot rolled high case depth condition. However, fatigue cracks nucleated from both inclusions and microstructural features in the 1045 quenched and tempered low and high case depth, 4145 hot rolled low case depth, 4145 quenched and tempered low case depth, and 1060 normalized conditions. The microstructural feature responsible for crack nucleation in some conditions was characterized with advanced microscopy utilizing a focused ion beam (FIB) and electron backscattered diffraction (EBSD). Characterization of the 1045 normalized condition revealed a ferritic region with fatigue induced deformation substructure in the crack nucleation area. The crack nucleation region in the 4145 hot rolled condition was associated with the bainitic microstructure. Because of the prominence of inclusions nucleating fatigue cracks in some conditions, extreme value statistical (EVS) analysis was undertaken based on an inclusion survey of the 1045, 4145, and 1060 alloys. By combining EVS analysis with a fracture mechanics methodology, the endurance limit of the induction hardened samples was conservatively estimated within 5-15 pct. of the experimentally determined endurance limits. A Woodvine analysis, modified by incorporating residual stress, was also performed by comparing the local fatigue resistance based on the local microhardness to the effective stress (applied stress plus the residual stress). The modified Woodvine model successfully predicts the nucleation locations of fatigue cracks observed during fatigue testing and provides insight on the effects of case depth, selection of core microstructures, and residual stress on fatigue resistance. |
