This content is not included in
your SAE MOBILUS subscription, or you are not logged in.
Dynamic Analysis of the Audi Valvelift System
Abstract:
Fully variable valve trains provide comprehensive means of adjustment in terms of variable valve timing and valve lift. The efficiency of the engine is improved in the operating range and in return, an increasing complexness of the mechanical design and control engineering must be handled.
For optimization and design of these kinds of complex systems, detailed simulation models covering different physical domains, i.e. mechanics, hydraulics, electrodynamics and control are needed. Topic of this work is the variable valve train named Audi valvelift system (AVS) e.g. used in the Audi 2.8l V6 FSI engine. The idea of AVS is to use different cam lobes at different operating points. Each intake valve can be actuated by a large and a small cam. For full load, the two inlet valves are opened by the large cam profile - ideal for high charge volumes and flow speeds in the combustion chamber. Under partial load, the small cam profiles are used. As a result, the gas exchange improves, throttling losses are minimized and fuel consumption is reduced.
To investigate the dynamical behavior and interactions between all subsystems, e.g. between chain vibrations and shifting events, an overall simulation model comprising the timing chain drive, chain tensioner, hydraulic cam phasing system, valve train and the AVS system including the electromagnetic actuator has been derived.
For modeling and simulation of the different subsystems, several highly specialized and efficient programs are used. To couple these subsystems co-simulation techniques are applied with the advantage that all subsystems can be computed in parallel on multi-core architectures to speed up integration time making optimization possible.
In this paper a dynamical analysis of the entire AVS valve train including the electromagnetic actuators combining experiments and simulation is presented. The simulation model is described and validated with experimental results using different test rigs. Furthermore, interactions between the subsystems and optimization results are presented.