On the relationship between wheel and rail surface roughness and rolling noise

Theoretical models linking rolling noise and surface roughness have been available for some 20 years. For even longer, the qualitative link has been acknowledged between the presence of visible corrugation on rail or wheel surfaces and increased noise generation. This roughness, or undulation in the surface profile, has wavelengths of most importance between about 10 and 300 mm, and amplitudes from between about 0.1 mu m and 30 mu m-greater for severe corrugation. An important open question in relation to roughness excitation is the validity of the assumed linear relation between roughness and noise. In part, the answer to this question depends on the quality and completeness of the roughness measurements. This is illustrated by comparisons of different rail roughness systems. Special care must also be taken with the processing of roughness data before a representative excitation spectrum can be derived. In this paper experimental evidence from a number of sources is presented to confirm the linear relation between roughness and noise, at least for roughness amplitudes which are not too severe. Attention is also given to the validity of the assumption made in the models that the wheel and rail roughness spectra may be added on an equal basis to give the total roughness spectrum. (C) 1996 Academic Press Limited