Wavelength fixing mechanism

Published: 12 April 2020| Version 1 | DOI: 10.17632/8k8k4b4cvj.1
Bingo Balekwa


The data is collected to establish the wavelength fixing mechanism for short-pitch rail corrugation. Average wavelengths are measured on each of the heavily corrugated rail track curves using a ruler. Train speeds are downloaded from a few locomotive black boxes. The speeds data is extracted into an excel spreadsheet. The blackbox downloads data every minute and captures Global Positioning System (GPS) coordinates for the location where the train speeds were captured. Together with train speeds, corrugation wavelengths are used to calculate corrugation frequencies for each track curve. The average corrugation frequencies were saved for correlation to a locomotive traction wheelset natural frequencies. Experimental Modal Analysis was conducted to obtain the point Frequency Response Functions (FRF) of the class D39200 locomotive wheelset. The wheelset resonance modes were obtained and a fundamental resonance mode at 108 Hz is found have a more closer correlation to corrugation frequencies for each track curve. The GPS coordinates were used on Google Earth Pro to view the terrain and gradients of the track curves. The blackbox data downloaded also recorded whether the train was in traction or braking during recording of the data, together with that information, the gradient on Google Earth Pro was used to determine the direction of the train. The excel spreadsheets named with locomotive numbers contain the GPS data downloaded in decimal form, whilst the files with a CSV format shows the GPS data that has been converted to degrees, minutes and seconds. A Finite Element Model was developed to validate the resonance modes of the traction wheelset and also to investigate the mode shapes of the wheels and axle on vibrational excitation. The axle and wheel drawings for the D39200 wheelset used for the model are attached. Modala Analysis was conducted on the rail to capture the vibration phase midspan between two sleepers. The vibration phase was captured for rail on FY and PY-type concrete sleepers; and also on steel sleepers. This exercise was done to investigate the correlation of the vibration phase of the wheels to that of the rail on different types of sleepers. Two excel files named "Rail on steel sleeper vs PY or FY type concrete sleeper". This data also has the captured vibration phase data for the wheels.


Steps to reproduce

Steps to reproduce are in chronological order and as follows: 1. Measure complete wavelengths for short pitch corrugation along a track curve and calculate the average wavelength for that particular curve. 2. Download a black box data for train speeds per GPS coordinate. 3. Together with train speeds, use the average wavelengths to calculate the corrugation frequencies for each track curve. 4. Suspend a locomotive bogie away from the ground using a crane so no external damping occurs on the wheels. 5. Use a PXI equipment installed with LabVIEW or SignalExpress, a triaxial accelerometer, impact hammer keyboard and monitor to conduct Modal Analysis on a D39200 locomotive class wheelset or similar. Test setup should be for a sample rate of 4 kHz with 30 averages. 6. Obtain the Frequency Response Functions (FRF) in a text file and extract the data to a spreadsheet. 7. Develop FRF plots. 8. Develop an FEA model to conduct modal analysis and validate the experimental modal analysis results. 9. Correlate the modal analysis FRF results to calculated corrugation frequencies. 10. Use the modal analysis equipment in point number 5 on the field side of the rail crown and excite the rail with the vertical force on the running surface. Place the accelerometer midspan on the field side of the rail crown. 11. Conduct modal analysis on the rail and obtain vibration phase results. 12. Correlate the vibration phases of the rails with those of the locomotive wheels to investigate any correlation.


University of Johannesburg - Doornfontein Campus


Railtrack Modeling