Correlation Work on Shaker Rig Tests and Simulations - An investigation of damper, bushing, friction and tire modeling with respect to vertical vibration insulation
Examensarbete för masterexamen
Automotive engineering (MPAUT), MSc
In the Automotive Industry it becomes more and more important to reduce costs and cut down development time. It is one of the main challenges in order to stay competitive, also for the Volvo Car Cooperation, where this thesis work was carried out in cooperation with Chalmers University of Technology. With regard to Vehicle Dynamics, a large amount of the cost is generated while prototype cars need to be set-up and changed to meet requirements during the testing phase. With the increase in computational power and the availability of advanced software for vehicle dynamics simulations in recent years, it is now possible to do some of this work earlier in the design phase. The future goal is to use prototype testing mainly for validation of having met the requirements, rather than changing the car to achieve them. For this to be possible, it needs to be assured that the CAE methods are stable and show good correlation to physical testing. This thesis focuses on correlation of vertical vibration insulation simulations. The starting point was that shaker rig tests showed different results from shaker rig simulations in Adams Car. A thorough pre-study was performed to understand the theory of key areas for vertical vehicle dynamics, such as damper, bushing, tire and friction modeling. The technical background section of this work should therefore provide all the needed information to understand the carried out work, also for a reader new to this field. As a next step, simulation models are developed ranging from linear quarter car state-space models in MATLAB to more advanced non-linear quarter car models in Dymola. The most advanced Dymola model features a self-developed low-pass filter damper model, hydro engine mount and top mount bushings as well as different friction implementations. Friction modeling was also investigated with a full car model in Adams Car. Shaker rig, damper and suspension parameter measurement machine tests where conducted, to parametrize models and gain more insight into the physical phenomenas that are tried to be replicated. Friction was identified as the largest factor why simulations and tests do not correlate. The implementation of friction improved results significantly, especially at low frequencies. More accurate damper modeling that tries to capture the hysteresis loop in the Force-Velocity Diagram due to compressibility, cavitation and backlash, seems to have a smaller influence on shaker rig simulation correlation, especially in the low frequency range (0 3 Hz). It could be verified that accurate bushing and tire modeling is needed to obtain good correlation at higher frequencies (3 20 Hz). Convergence of simulations was found as a main issue for models that included friction. Validation of bushing and tire modeling, as well as accurate parametrization of friction and improved solve-ability of models with friction remain as future tasks.
Farkostteknik , Transport , Vehicle Engineering , Transport