Efficiency optimization of the push-belt CVT by variator slip control

Abstract

Continuously Variable Transmissions (CVT) are becoming increasingly popular in automotive applications. What makes them attractive is the ability to vary the transmission ratio in a stepless manner without interrupting the torque transfer. This increases comfort by eliminating the discrete shifting events and increases performance by choosing the most suitable transmission ratio for every driving situation. Using a CVT could potentially save more than 15% of fuel consumption compared to manually shifted vehicles. This figure however is never met, because of the internal losses in the CVTs in production today. If the losses in a CVT can be lowered, then the overall fuel economy of a CVT equipped vehicle will be improved with the same amount. With current CVTs ranging around 80% efficiency, an improvement of around 10% is possible compared to currently available CVTs if an optimal actuation and control system is used. This thesis is about the optimization of the control system of the CVT by using slip as the control variable. This is part of a larger project focussing on the entire actuation and control system. Also a CVT with Electro-Mechanically Pulley Actuation (EMPAct) is developed aiming to reduce the power consumption of the CVT actuation system. Combined, these two projects aim to improve the fuel economy of the CK2 transmission from Jatco with 10%. Models for the clamping forces and traction in the variator are compared. The continuous belt model is compared with a pushbelt model. A parameter study shows the influence of the model parameters on the outcome of the models. The output of the models are also compared to measured values. A nonlinear dynamic model for slip in the variator is derived. This model can be linearized in certain operating points. This model can be used for the design of a control system, simulation of slip in the variator or for analysis. Measurement of slip directly is not possible, therefore a good estimation method is needed. Several estimations of slip in the variator are compared. The position measurement of the pulley is used in the measurements shown in this thesis. Measurements on a beltbox testrig are given that clearly show a relation between slip and efficiency and slip and traction. This relation changes as a function of other parameters like speed, ratio, clamping force etc. Estimation of the efficiency potential of the pushbelt variator shows that a potential of between 5% for high torques and 20% for low torques exists. A slip control system is developed to show the possible efficiency improvement. First, a beltbox setup is used to test a simplified slip controlled variator. Ratio changing is not taken into account in this setup. After successful tests with this setup another setup is used that incorporates a Jatco CK2 transmission and an internal combustion engine. This test setup is more realistic, but therefore also more complicated to control. A gain scheduled approach is used to compensate for the slower actuation system. This system is then also applied to a testing vehicle.