Elsevier

Wear

Volume 265, Issues 9–10, 30 October 2008, Pages 1238-1248
Wear

Calculation of wear on a corrugated rail using a three-dimensional contact model

https://doi.org/10.1016/j.wear.2008.01.026Get rights and content

Abstract

This paper presents the results from simulations of railhead wear using a three-dimensional contact model. The wheel/rail contact is modelled as non-Hertzian and non-steady based on the Variational Method [J.J. Kalker, Three-Dimensional Elastic Bodies in Rolling Contact, Kluwer Academic Publishers, 1990] and the wear is assumed to be proportional to the frictional work. A wheel rolling over initial sinusoidal roughnesses is considered and when constant values of normal force and creepage are used, it is found that the maximum wear occurs at positions close to the crest of the initial wavelength and therefore roughness is not predicted to grow. A range of prescribed dynamic normal forces were then used to investigate the relationship between wear and dynamic force but again, no corrugation growth mechanism was revealed. Wear calculations using both Hertzian and non-Hertzian contact are compared and the significance of the non-Hertzian effects on the wear calculation is shown.

Introduction

Rail corrugation and roughness have long been concerns in the railway industry. Rail roughness consists of broad band irregularities covering wavelengths from 10 to 500 mm on the railhead while corrugation is a quasi-periodic irregularity. Six types of corrugation have been classified by Grassie and Kalousek [2], according to different wavelength-fixing and damage mechanisms but, although much research work has been carried out, there is still inadequate knowledge of how roughness develops over time and why corrugation appears on some types of trackform in some place and not in others. Rail grinding is widely used to treat rail corrugation and roughness.

A large number of parameters influence the dynamic system of train/track interaction and thus influence the wear of the wheel and rail. The main factors are: contact geometry; normal force; traction (or creepage) and coefficient of friction. It is therefore desirable to perform wear calculations with a contact model that includes all these factors and to determine the relevance of each factor. In this paper, the wheel/rail contact is modelled in three-dimensions based on the variational method by Kalker [1]. Sinusoidal corrugations on the railhead are considered for different wavelengths and amplitudes and the normal and tangential problems are solved in a transient way using a time-stepping integration technique, taking account of non-Hertzian normal contact and non-steady tangential contact. The wear in the contact patch is calculated based on the friction work and a constant normal force and creepage are initially used to investigate any possible wavelength-fixing mechanism. A range of prescribed dynamic forces are then used to investigate the relationship between wear and force and a comparison is also carried out between Hertzian and non-Hertizan contact modelling.

Section snippets

Background

Most research work has been focused on short-pitch corrugation [3], [4], [5], [6]. The most common wavelengths of short-pitch corrugation are in the range of 20–100 mm. The main damage mechanism according to Grassie [2] for this type corrugation is wear of troughs mainly from longitudinal slip. Although different wear mechanisms exist, abrasive wear is the only mechanism accounted for and material loss due to this type of wear is assumed to be proportional to the friction work. To investigate

Wheel/rail system

Consider a free wheel rolling over a corrugated rail, as shown in Fig. 1. A global coordinate system is defined to have an x-axis corresponding to the rolling direction and y-axis corresponding to the lateral direction and a local coordinate system which moves as the wheel centre moves along the rail is defined for the potential contact area. Parameters for the wheel/rail system are given in Table 1. For a nominal load of 70 kN, the wheel/rail geometry gives an elliptical contact patch having a

Wear calculation using a constant normal force

Wear after one wheel passage was first calculated for sinusoidal roughness of wavelengths λ = 25, 30, 40, 60, 80 and 100 mm. These cover the range of typical short-pitch corrugation wavelengths. A corrugation amplitude of 10 μm was chosen to be sufficiently small to avoid the loss of the contact in the corrugation trough. Nielsen [16] gives a two-dimensional plot for characteristic wavelengths for which the corrugation is most likely to develop. It was shown that, for a given creepage, only

Wear calculation using prescribed dynamic normal forces

The wear calculation using a constant normal force in Section 4 does not reveal any mechanism for corrugation growth so in this section a dynamic normal force will be considered. A dynamic normal force is normally obtained from a vehicle/track interaction model and various such models have been developed, either in the time domain or the frequency domain. For a sinusoidal roughness, the dynamic normal force obtained from an interaction model is normally close to sinusoidal, but with certain

Conclusions

The contact problem of a wheel rolling over a rough rail has been modelled in three-dimensions using the variational method. The rolling process has been treated as transient and the wear calculated for various sinusoidal roughnesses. The contact geometry due to the roughness has significant influences on solutions of the normal and tangential wheel/rail contact. For the case where a constant normal force and creepage are present, it has been demonstrated that the maximum wear occurs near the

Acknowledgement

This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) through Rail Research UK (RRUK).

References (24)

  • S.L. Grassie et al.

    Rail corrugation: characteristics, caused and treatment, proceedings of the Institute of Mechanical Engineers Part F

    Journal of Rail and Rapid Transit

    (1993)
  • P.J. Remington

    Wheel rail rolling noise.1. Theoretical-analysis

    Journal of The Acoustical Society of America

    (1987)
  • Cited by (44)

    • Simulation of rail roughness growth on small radius curves using a non-Hertzian and non-steady wheel-rail contact model

      2014, Wear
      Citation Excerpt :

      In numerical studies of long-term corrugation growth, approximate contact models are often applied both for the computationally cumbersome time-integration of the dynamic vehicle–track interaction and for the calculation of wear. In recent years, several researchers have presented models that instead use non-Hertzian and non-steady contact models for the wear calculation [11–15]. It has been shown that, both for single wavelength and broadband rail roughness excitation, these models do not predict corrugation growth.

    • Analysis of Unilateral Rail Corrugation Mechanism Based on Friction Self-Excited Theory

      2022, Xinan Jiaotong Daxue Xuebao/Journal of Southwest Jiaotong University
    View all citing articles on Scopus
    View full text