Cebulla, T., Grundl, K., Schindler, T., Ulbrich, H. et al., "Spatial Dynamics of Pushbelt CVTs: Model Enhancements," SAE Technical Paper 2012-01-0307, 2012, doi:10.4271/2012-01-0307.
Apart from performance, comfort, cost and fun to drive, the reduction of fuel consumption has become a primary driver in the world market of the automotive industry. As continuously variable transmissions based on the pushbelt principle can be operated in an optimal state at any time, they are very suitable to meet the mentioned requirements. However, the power transmission in the system is very complicated. Both detailed measurements and simulations are necessary to understand and to optimize the physical mechanisms, power density and shift characteristics.The current paper presents a spatial simulation model for transient analysis at different levels of detail. An initial model based on non-smooth multi-body theory is outlined. It consists of two rigid pulleys each with one tilting loose sheave. The pushbelt comprises two ring packages based on the co-rotational approach. These guide approximately 400 rigid elements by bilateral friction laws, with possibility for Coulomb and Stribeck curves. Flexible decoupled contact laws with spatial friction between sheaves and elements represent the axial stiffness of the elements. In-between the elements, flexible and frictionless contact laws are chosen. The comparison with local and global measurements validates this approach in realistic load cases.In the main part, improvement possibilities are presented. In particular, the influence of curved strand shapes between the pulleys is discussed. Planar pre-integration strategies are added to reduce the simulation time. The effect of unilateral contact laws between elements and ring packages is examined. Finally, the stiffness of the sheaves is added by coupled quasi-static contact laws.Overall, an efficient framework is shown to investigate the dynamics of individual components, pushbelt misalignment, elastic slip as well as local push and tension forces. This is very important for the economic optimization of the real system without carrying out costly experiments.