Modelling the influence of velocity on wet friction-element friction in clutches

Abstract

Purpose This study aims to establish a friction coefficient model relative to the rotation speed of a wet clutch engagement, which can predict friction coefficient under different stages of slipping velocity and different load pressures. In particular, the model has been improved by accounting the speed effect for the perdition of wet friction-element boundary friction, which is significant for understanding the friction mechanisms and for supporting the development of more efficient and related products. Design/methodology/approach This research investigated the mechanism of wet friction in a wet clutch engagement. A mixed friction model is established based on the asperity model and Newton’s law of viscosity. To obtain a friction coefficient computed by the model, the normal load shared by both asperities and lubrication fluid needs to be determined. Therefore, rough surface contact mechanism is analysed; a surface topography model is established; and surface parameters are obtained by means of surface topography measurement and reconstruction. Finally, verification of the mixed friction model is achieved. Findings Friction will be generated by both the asperity contact and the lubrication film shear relative to the rotation speed. And, the higher the relative speed, the larger the shearing power of lubrication film. It is caused by decrease in contact area of asperity. Surface morphology of a sintered bronze friction disk was obtained by a Laser-Micro-Test. The predicted results by the established model show that the total friction coefficient slightly reduced and then increased suddenly with speed. The surface topography model is responsible for the nonlinear behaviour of the asperity friction. Results of the simulation model are in agreement with those of the wet clutch engagement experiments. Originality/value This research is original and it is supported by the national defence project. The wet friction element which is applied on tracked vehicles is analysed for the first time. Through the model, the trend of the friction coefficient can be more accurately predicted. The problem of the wet friction plate modelling difficult is solved by using the mixed friction model.

Effect of ATF temperature on the friction-wear characteristic of a wet multi-disc clutch during the running-in process

“…As the ATF is usually applied to lubricate the friction surfaces and take away the friction heat, the ATF temperature is critical to the friction and wear performance of clutch components (Wu et al, 2018a(Wu et al, , 2018bKugimiya et al, 1998). During a clutch engaging process, the ATF film at the friction components is squeezed, so sliding contact of the opposing surfaces occurs (Wang et al, 2018). The temperature will change the viscosity of ATF, thus altering the actual pressure of the contacting surface asperity summit (Zhang et al, 2016;Holgerson and Lundberg, 1999;Zhao et al, 2017aZhao et al, , 2017b.…”

Influence of applied pressure on the tribological behaviors of a wet multi-disc clutch during the running-in process

“…(Jenson et al , 2018; Cheng et al , 2015; Zhou et al , 2019). Changes in friction and wear that occur during the running-in process are more than a consequence of surface roughness alterations alone, but also the operational parameters, such as temperature, rotation speed and applied load (Blau, 2005; Wu et al , 2018; Wang et al , 2018).…”

Identification and Validation of Linear Friction Models Using ANOVA and Stepwise Regression

“…However, the application of EHL methods to wet clutches is not appropriate due to low contact pressure and very small hydrodynamic film buildup [12]. Nevertheless, there are physical models from the application of dynamic pressure lubrication theory and Hertz contact theory, which make it possible to estimate the CoF in wet clutches and help to understand the friction mechanisms [13,14]. However, their practical relevance is limited due to the assumptions and simplifications made during model derivation.…”