Empirical characterisation of friction parameters for non-linear stick-slip simulation to predict the severity of squeak sounds
Artikel i vetenskaplig tidskrift, 2022
Squeak and rattle (S&R) are nonstationary annoying sounds inside the car cabin that impose high warranty costs on car manufacturers. The need for taking S&R preventing measures and the maturity level and cost considerations of the physical prototypes during the pre-design-freeze stages justifies the use of virtual simulation methods. Squeak is a friction-induced high-frequency sound that is attributed to the stick-slip friction phenomenon. The importance of the friction parameters in the squeak severity prognosis is analytically and experimentally mentioned in the literature. However, studying the variation of these parameters respecting the changes in loading and driving conditions with the aim of application in virtual simulations has remained limited or too simplistic. In this work, the rate weakening effect of the friction coefficient curve was involved in the nonlinear finite element simulation of stick-slip events by an exponential decay formulation. The approximated squeak severity by the virtual simulations for selected material pairs agreed with the empirical results from a flexural stick-slip test bench. From the empirical stick-slip data, the dependence of the squeak severity on the friction decay coefficient and the difference of static and kinetic friction coefficients at low and high normal loads were observed, respectively. The relativity of friction parameters on the test conditions demands a dynamic updating of the friction model that can be achieved by polynomial or exponential approximations. Also, the observed polynomial relationship between the squeak severity and the operational conditions can be used to estimate the squeak severity from the linear dynamic simulation results. The outcome of this work can help to better understand the influence of the friction modelling parameters and their variation respecting the operational conditions. This can facilitate a more accurate prediction of squeak risk by employing virtual simulation tools in the pre-design-freeze stages of car development.