About the mechanical properties of bicycle tyres

Abstract

A lot of research has been done on the behaviour of pneumatic tyres and this has led to various tyre models and a lot of measurement data. However, in the specific field of bicycle tyres, not so much measurement data is available. However, in 2013 Andrew Dressel received the Degree of Doctor of Philosophy in Engineering at the University of Wisconsin-Milwaukee by presenting his research: Measuring and modeling the mechanical properties of bicycle tires. In this research he did a lot of measurements with multiple bicycle tyre brands and models under different conditions. The results from these measurements are interesting to use for modelling purposes. The goal of this research is to find out if it is possible to estimate the bicycle tyre behaviour in terms of vertical stiffness, cornering stiffness and camber stiffness based on known parameters like the inflation pressure, tyre width, rim width, vertical load and the rubber compound using a tyre model. An important part of this thesis are tyre models. The measurement data will be analysed using various tyre models. The first used tyre model is the brush model. This model uses a single material parameter and it turns out that this is too less to be able to extract clear relations between the tyre behaviour and the known parameters like inflation pressure, tyre width and rim width. The second tyre model that is used is the enhanced string model. This model is an extension of the brush model. This model has three material parameters for the vertical direction and also three material parameters for the lateral direction. Investigating this model shows that there is reasonable suspicion to assume that one of the model parameters represents the inflation pressure. In order to find the material parameters for every tyre, inflation pressure and normal load combination a parameter optimisation is required. The results from this parameter optimisation show that the suspected model parameter is actually not related to the inflation pressure. When the obtained model parameters from the optimisation are put back into the tyre model interesting results are acquired, because the model output agrees fairly well with the measurement data. This is unexpected because the model parameters were differing a lot depending on how they were obtained, e.g. obtaining the parameters from cornering stiffness or from camber stiffness. The answer to the research question is no. In order to extract relations between the model parameters and the tyres behaviour depending on their measurable properties, a lot more complete measurements are needed. The whole idea was to be able to estimate tyre behaviour without the need of extensive testing. This still might be possible, but in order to to that a lot more measurements are needed first. These measurements should create a baseline of model parameters which can be used to estimate the model parameters of unmeasured tyres.