PSI - Issue 57

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

800

2

1. Introduction With the automotive market becoming more and more globalized and competitive, the importance of performing faster product development has driven our company to find new ways to make product validation more efficient, and in parallel to search for tools that guarantee a more robust and flexible approach to predictive reliability. In the framework of a global program focused on the improvement of the reliability of our components, particular attention has been given to improve the prediction of the mechanical endurance of engine cooling components undergoing vibration loadings. Finite Element Analysis is generally the first step of the design validation process. The numericalsimulations offer the advantage of validating the design on the basis of the predicted stress, induced on that part by the accelerated validation signal. In the case of automotive components, the nature of the signal depends on the location of the component within the vehicle. For parts mounted on the body and cabin of the car, a random signal better represents the vibration loadings which are transmitted during driving on harsh conditions from the suspensions to the chassis. In case of engine mounted components, the type of vibration is more sinusoidal in nature, since it better replicates the harmonic vibration from a rotating machine such as the electric or combustion engine. In reality, the vibrations load are never purely random nor purely sinusoidal. For this reason, various carmakers (original equipment manufacturers, OEMs) are required to validate the design of their components according to more representative “mixed” signals. Sine-Sweep-on-Random (SSoR) vibration specifications are used to validate the mechanical endurance of the components mounted on the engine. Random vibrations with addition of shock pulses, called Shocks-plus-Random (SpR) are the typical vibration signal for design validation of components mounted on the body of the car. The SSoR signal attempts to include those vibrations generated by a rotating machine which are not purely harmonic but random (the so-called leftover) (Bonato and Goge (2018)). In the case of SpR, one would include those highly transient events that cannot be fully reproduced by a power spectrum density signal (PSD) which is ergodic and stationary, such as potholes or small shocks. In both cases the dynamic response of the component to be validated is in the elastic domain, so these specifications can in principle be further accelerated or transformed into other signal types based on the fatigue damage equivalence. The acceleration of vibration signal is the main objective of developing shorter validation tests, since one would cumulate the total vibration usage life of the components to , for example; 30 hours monoaxial accelerated shaker tests (Delaux and Kihm (2010)). The same criteria can be used to create “simpler” vibration signals, i.e. accumulatingthe same vibration fatigue damage of a sine on random signal (or a shock plus random) into an equivalent PSD or sine sweep. The advantage of this approach is that the new signal can be used easily for FEA based simulations in the frequency domain, whereas more complex signals require a more computationally more challenging time series approach. There are two main parameters that need to be tailored in order to perform the fatigue equivalent simplification of complex signals: the damping ratio Q and the Basquin fatigue coefficient b. These two parameters are strictly linked to the component under investigation, and in particular to the zone of failure. Therefore one must apply a Physics of Failure (PoF) approach to predict the main failure mode. In this paper, we investigated the effect of the parameters considered for the generation of the iso-damage cumulative signals, and the effect of the predicted life to failure obtained from the FEA simulations.

Nomenclature HVCH High Voltage Coolant Heater FEA Finite Element Analysis FDS Fatigue Damage Spectrum g Gravity (m/s²) PSD Power Spectrum Density PoF Physics of Failure Q Damping ratio SSoR Sine Sweep on Random  Damping factor