Wheel squeal is a loud, tonal noise generated when a train negotiates the curve of a rail. Previous field tests and laboratory tests found that the sound pressure level of squeal noise increases with angle of attack and rolling speed. Also, it was found that the probability of squeal increases with relative humidity and that the wheel squeal still exists after the application of friction modifiers. These phenomena, however, have not been explained in terms of the generation mechanism of wheel squeal.
In this research, these problems are explored in three ways: theoretical modelling, numerical simulation and experimental investigation. The theoretical modelling is performed by integrating the model of contact mechanics in rolling contact and the model of wheel vibration. The integrated model in the time domain can predict the lateral creepage and wheel vibration, so as to illustrate the generation mechanism of wheel squeal. A rolling contact two disc test rig is used for experimental investigation. A new method was developed for the test rig to measure the contact forces, and a series of tests under various conditions were conducted to investigate the effect of angle of attack, rolling speed, relative humidity and friction modifiers on the rolling contact mechanics and wheel squeal.
The theoretical model is utilised to illustrate how the nonlinear friction creep behaviour interacts with the wheel vibration. In addition, the causes of wheel squeal sound pressure level increasing with angle of attack and rolling speed are explained, respectively. Furthermore, the vibration velocity amplitudes of the wheel at different rolling speeds and angles of attack are simulated, and the results correlated well with experimental measurements. The test results show that the adhesion ratio decreases with relative humidity, and the reason why relative humidity affects the occurrences of squeal is explained. The test results show that friction modifiers can greatly reduce or eliminate the negative slope of the measured creep curve, but squeal still exists at high rolling speed and angle of attack. Based on the experimental results and the theoretical model of wheel squeal, it is concluded that this may be because the test rig measurements are quasistatic while the real instantaneous friction creep curves still have a negative slope.
This thesis makes some substantial novel contributions to the existing literature. Firstly, the mechanisms that cause the sound pressure levels of wheel squeal to increase with angle of attack and rolling speed are determined theoretically. Furthermore, wheel vibrations at various angles of attack and rolling speeds are simulated and correlated reasonably well with experimental results. The effect of negative damping on squeal occurrence is simulated to explain why the squeal tends to occur in high relative humidity. The integration between the friction-creepage curve and wheel vibration is simulated to explain why squeal still exists after the application of friction modifiers. The novel experimental investigations include the developments of new methods to measure contact forces and angle of attack. New laboratory investigations determining the effect of rolling speed on frequency divergence of squeal noise and the effects of relative humidity and oil-based friction modifiers on traction in rolling contact and sound pressure level of wheel squeal are also contributed.