Technical Reports

Control of Wheel/Rail Noise and Vibration

  • 01
  • Apr
  • 1983
AUTHOR: P. J. Remington, N. R. Dixon, L. G. Kurzweil, C. W. Menge, J. D. Stahr, and L. E. Wittig
OFFICE: RPD
REPORT NUMBER: DOT-TSC-UMTA-82-S7
SUBJECT: Track/Train Interactions, Best Practices
KEYWORDS: Rail Grinding; Wheel Truing
ABSTRACT: This report presents the results of a program to develop and evaluate techniques for the control of wheel/rail noise in urban rail transit systems. The first part of the program included a literature review and a cost-benefit analysis to select, for further study, the most cost-effective wheel/rail control treatments. In this document, the focus is on the development, improvement, and validation of the analytical tools to be used in the design, development, and testing of those treatments. An analytical model of the generation of wheel/rail noise was developed and validated through an extensive series of field tests carried out at the Transportation Test Center in Pueblo, Colorado, using the State-of-the-Art-Car. A sensitivity analysis was performed using this analytical model. The analysis showed that wheel/rail noise is relatively insensitive to changes in most system parameter values, except wheel and rail roughness, contact area, and contact stiffness. The surface finish produced by most wheel truing and rail grinding machines was measured. A belt grinder used by the Toronto Transit Commission for wheel truing and a rail grinding block car used by the Chicago Transit Authority to grind rails were found to produce the quietest surface finishes, giving an estimated 12 dBA of noise reduction when compared with typical rapid transit wheels and rails in revenue service. A scale model of a new concept wheel employing a resilient tread was built and tested. Noise reductions of up to 8 dBA were achieved with tread stresses in the manageable range. The analytical model developed under this program was shown to agree reasonably well with field measurements of wheel/rail noise.
Pdf Download PDF Document [11.3 MB]

####