Rubber Friction on Ice: Experiments and Modeling

Abstract

Rubber friction on ice is studied both experimentally and theoretically. The friction tests involve three different rubber tread compounds and four ice surfaces exhibiting different roughness characteristics. Tests are carried out at four different ambient air temperatures ranging from À5 to À13 C, under three different nominal pressures ranging from 0.15 to 0:45 MPa, and at the sliding speed 0.65 m/s. The viscoelastic properties of all the rubber compounds are characterized using dynamic mechanical analysis. The surface topography of all ice surfaces is measured optically. This provides access to standard roughness quantities and to the surface roughness power spectra. As for modeling, we consider two important contributions to rubber friction on ice: (1) a contribution from the viscoelasticity of the rubber activated by ice asperities scratching the rubber surface and (2) an adhesive contribution from shearing the area of real contact between rubber and ice. At first, a macroscopic empirical formula for the friction coefficient is fitted to our test results, yielding a satisfactory correlation. In order to get insight into microscopic features of rubber friction on ice, we also apply the Persson rubber friction and contact mechanics theory. We discuss the role of temperature-dependent plastic smoothing of the ice surfaces and of frictional heating-induced formation of a meltwater film between rubber and ice. The elaborate model exhibits very satisfactory predictive capabilities. The study shows the importance of combining advanced testing and state-ofthe-art modeling regarding rubber friction on ice.

Multiscale physics of rubber-ice friction

“…During steady sliding the friction coefficient decreases with an increasing load, which may arise from the same effect as discussed in Ref. 12 or may be related to the smoothing of the ice surface by the melting of the ice, which is at its greatest at the highest load.…”

Study on tire‐ice traction using a combined neural network and secondary development finite element modeling method

“…Makkonen and Tikanmäki presented a theoretical model of kinetic ice friction that can also be applied to melting ice. However, the friction traction models established all show a high level of error . Specifically, the results of models are not in good agreement with the experimental results.…”

“…However, the friction traction models established all show a high level of error. 19 Specifically, the results of models are not in good agreement with the experimental results. This is partially attributable to the failure to realize efficiency and accuracy of friction characteristics between the tire and the ice surface.…”

Optimization of the tire ice traction using combined Levenberg–Marquardt (LM) algorithm and neural network

“…Lastly, Klapproth [19] proposed an innovative mathematical equation to describe rough rubber's viscous friction when on smoothened ice. However, the friction traction models that had been established show a high error [20]. In other words, the model built cannot effectively use the experimental data samples to accurately predict the value of friction coefficient under different practical environments.…”