PSI - Issue 57

Marco Bonato et al. / Procedia Structural Integrity 57 (2024) 799–809 / Structural Integrity Procedia 00 (2019) 000 – 000

807

9

are found on the externalcasing of the components (Figure 7b). The vibration fatigue loadings concentrate the stress on the mechanical part, leaving less damage on the electronic components (the rotor, the stator and the diodes.).

Fig. 7. (a) Damage contour map for the HVCH bracket (b) Damage contour map for the alternator bracket

3. Results and Discussion The cumulative fatigue damage results are focused on those nodes that show high vibration stress during the modal shape simulation. The damage value has been calculated by simulating the input vibration load on axial direction at the time. The fatigue value is obtained from the cumulative stress on the h ot spots, i.e. the nodes that show high fatigue or lower useful life. For the purpose of the comparative analysis, only those nodes that show higher stress concentration were considered for the vibration damage analysis. Also, only the axial direction that generates the highest damage (in all cases analyzed, it was found that the maximum damage is provoked by one particular axial direction of the acceleration load). No FEA fatigue simulation has been performed from the original signals (shock plus random and sine on random, since that would require a time series approach not standardized yet for design validation). The preliminary modal analysis of the components permits to immediately identify the hot spots on the components, those nodes that show the highest stress concentration, therefore can be considered the weakest point of the design. It is interesting to see that in both components, the stress concentration appears on the previously defined “mechanical parts”, i.e. on the brackets of the HCVH and on the external casing of the alternator. Both elements are composed of aluminum alloy. No significant level of stress concentration was found on the so- called “electronic part”, i.e. on the component of the PCBs or the heatingelement (in case of the HVCH). This is a posteriori knowledge that should be used to address the appropriate choice of the two parameters for the simplification of the original validation tests, the SoR and SpR, and have indeed be used to generate equivalent signals with tailored combinations of Q and b (see Section 2.1). The results clearly show the important difference on the predicted fatigue life on the hot zone if the simplified PSD or Sine sweep are generated from the FDS with inappropriate parameters, i.e. those suitable for the equivalent fatigue failure located on the electronic parts. The effect of the parameters and the resulting useful life at the hot spots of the HVCH are shown in Figure 19. The fatigue damage calculated at the same node of the bracket has been normalized to the reference PSD, i.e. the one obtained with b=8 and Q=25. It can be seen that the PSD generated with the settings that target the electronic failure (b4 and Q25) slightly under-estimate the vibration damage recorded on the brackets. On the other hand, the PSD for mechanical failure generated with “typical” settings for aluminum components (b8 and Q10) highly over estimate the damage by three orders of magnitude.