Application of local concepts to evaluate the fatigue strength of cavitation-induced defects on cast components

The global push for sustainable energy has intensified the deployment of alternative fuels, such as hydrogen and ammonia, to decarbonize combustion processes. However, this transition introduces new challenges, particularly in critical engine components. These components are prone to cavitation-induced defects due to vibrations and coolant interactions, potentially leading to engine failure. This work investigates the applicability of the Theory of Critical Distance (TCD) and the Total Strain Energy Density (T-SED) for evaluating cavitation-induced defects in a casting material. The research begins with a material characterization for a stress ratio of R = 0, including testing of both notched and unnotched specimens with varying notch radius. Cyclic fracture mechanics parameters are derived from Single Edge Notched Bending (SENB) specimen testing, and the El-Haddad length is calculated using the threshold and fatigue strength of the unnotched specimen for the Kitagawa-Takahashi diagram. Based on these results, TCD and T-SED reference curves are evaluated. Planar simulations are conducted using geometries derived from cavitation defects, with material properties and boundary conditions applied to replicate simplified real-world conditions. The TCD and T-SED approaches are used to evaluate the stress and energy distribution around the defect. The TCD involves calculating the critical distance from notched and unnotched S-N curves using a statistical approach, while the T-SED approach evaluates the energy within a control volume that depends on the critical distance and the notch geometry. Both methods are validated using specimens with artificially induced defects, and the results are compared using TCD and T-SED reference curves. Based on the results of the simulations and experimental work, both methods are suitable to evaluate the influence of 2D defects on the achievable number of load cycles of the component.