PSI - Issue 75

Marco Bonato et al. / Procedia Structural Integrity 75 (2025) 677–690 / Structural Integrity Procedia (2025)

681

5

associated with each parameter is indeed very small (coefficient of variation < 2%) therefore it has been considered that such variability would not affect the results of the vibration fatigue time-to-failure.

Figure 1. CAE representation of the prototype model (left) and default material card for the Al-alloy of the prototype (right).

2.2. FEA Fatigue Simulation Settings The simulations are performed by considering a complex vibration signal type sweep-sine-on-random. The combination of random and sine amplitudes were selected so that the expected nominal time to failure would not exceed 1 hours. A finite element model of the prototype specimen was constructed using Ansys. A 2nd order hexahedral mesh with refined sizing is used at the critical location and 2nd order tetrahedons is used at other locations. Glue contact is applied between the cantilever beam and zig with a patch size 7.5mm (1.5 times the screw head). The contact assumption is validated through eigenfrequency correlation between test and simulation.

Figure

2.

CAE

Mesh

(left)

and

Contact

Status

(right)

Fatigue analyses were performed using nCode DesignLife. The FEA model is shown in Figure 1. Initially, all simulations were carried out for the original SSoR computed via the transient analysis method. Three iterations were run by adjusting the simulation parameters, from generic ones (typical case of simulation performed on a certain material that has not been fully characterized) to accurate parameters extrapolated from the full characterization. Only during the fourth iteration, the initial original sine sweep on random signal was transformed into purely random PSD and a purely sine sweep signal (via the fatigue equivalence method illustrated in [Bonato (2015), Bonato (2023) and Yang (2025)]. The analysis of these results will be the subject of a further publication.