Defining ballast properties in light railway and tram networks

Purpose and goal

The aim with this project was to develop an cost effective and relatively simple to use method for determination of ballast stiffness for light rail. This would signal when maintenance is necessary or when packing ballast is sufficient. The method has worked technically well and shows significant difference between average stiffness for good and bad ballast. The issue is that the variation in stiffness between measurement points at the same track segment is too large to get sufficient confidence and cost efficiency since too many measurements have to be performed and analysed.

Results and expected effects

Extensive tests have shown that the method used, measurement of vertical dynamic stiffness between 10-100 Hz for rails with impact excitation works well for the determination of local ballast quality. The method itself could be rationalized with lighter equipment, cheaper instrumentation and automated evaluation software. The large variability obtained between points on a short track segment for ´good´ ballast makes tests at a large number of points necessary for sufficient confidence. The risk to false alarm of ´bad´ ballast becomes otherwisw too large.

Approach and implementation

Extensive verification tests and analysis initially. additional, comprehensive measurements to understand better why a large variation in stiffness is obtained especially for tracks with good ballast quality. A MSc thesis student were engaged in the project during the first half of 2014 with the initial aim to define a simplified equipment and automated software for the field tests. The thesis work was changed to a rigorous, independent statistical analysis of test data. It showed that one can not predict god ballast with sufficient confidence unless many points are measured.