Search Results

The development process of automotive brakes is known to be challenging and time-consuming. It is an iterative process consisting of interplay between brake squeal simulation and extensive experimental investigations of the brake system at the test rig and in the vehicle. In this context, the complex eigenvalue analysis (CEA) of linearized FE models is a part of standard development process of brake systems. Nevertheless this linear analysis has not reached the status of a predictive tool yet, remaining a tool accompanying experimental investigations of the brake system only. Possible reasons may be inadequate simplifications of frictional contact, damping effects and friction material modeling on one hand and insufficiencies of the mathematical mechanical models themselves, i.e. linear vs. nonlinear stability analyses on the other hand. The extensive experimental investigations apply time consuming standard test procedures and need efficiency improvement.

Brake squeal is usually investigated using linearized models and the eigenvalues of the linear equations of motion. Eigenvalues with positive real parts are interpreted as the onset of squeal. Nonlinearities are commonly neglected due to the high effort associated with the corresponding calculations. Following the linear theory, the vibration amplitude should increase exponentially. On the other hand experimental results and overall experience show, that brake squeal is a stationary or quasi-stationary vibration phenomenon with approximately constant amplitude. This can only be explained by introducing nonlinearities into the model. These nonlinearities are limiting the increasing vibration amplitudes to a stationary limit cycle. Considering experimentally identified material properties of the brake lining as the main source of nonlinearities in the system a nonlinear disk brake model is introduced.