Abstract
KEYWORDS - CAD-based robustness evaluation using stochastic analysis, robust design, CAE-based quality management
ABSTRACT – The quality requirements of brake systems for passenger cars constantly increase. As part of the quality requirements OEM’s ask for lower probabilities for brake noise. To fulfill that requirement, the quality management needs to address and include as well as the brake design and the brake production process. The control of production and process tolerances is one necessary part of the quality management. However, since significant sources of brake noise are controlled by variations of environmental conditions or alterations of brake systems, the brake system also need “in build” robustness to minimize noise during the life circle of brakes. In the past, fine tuning and proof of brake noise quality was mainly based on tests. Today there is a mix of simulation and tests to balance the brake system and reduce brake noise probability. Of course, with the given time and cost pressure, improvement cycles late in the development process need to be reduced. That is only possible with an increase of CAE-based robustness evaluation taking into account all relevant sources of variation which may have an influence on brake noise occurrence. Therefore a CAE-process in virtual prototyping having sufficient forecast quality to check and quantify robustness becomes the key for a better integration of simulation and test based quality management. With such process in place quality management of brake systems will have the best possible balance between CAE-based simulation and real world testing to ensure robustness of the brake system in regard to rising brake noise quality requirements. The paper will discuss the challenges for software tools, CAE-modelling and CAE-processes to successfully apply a CAE-based robustness evaluation for brake noise application in virtual prototyping.
Robustness evaluation is a methodology to investigate how input scatter affects response variation and help to understand how causes connect to response variation. It should be noted that the introduced software tool optiSLang (11) is a general purpose tool for CAE-based robustness evaluation and the company Dynardo is a general purpose engineering consultant for CAE-based robustness evaluation, either not specialized to brake squeal simulation. Therefore the paper does not contribute to the necessary discussion what is the appropriate CAE modelling to reflect the underlying physics of brake squeal accurate enough. But because simulation is used today to investigate brake squeal and simulation models are successfully validated against hardware tests we state that appropriate CAE-models are available and can be successfully used to perform CAE-based robustness evaluation.