In recent years there has been an increase in the knowledge of, and need for, non-invasive monitoring of ballast in order to identify the problematic sections of track and decrease the maintenance cost. Various technologies such as Ground Penetrating Radar (GPR) are becoming accepted for investigating the condition of ballast. However since these techniques were not originally developed for engineering applications, their applicability in ballast evaluations can be sometimes uncertain. Continued empirical studies and condition specific calibrations are needed to demonstrate repeatable and quantifiable results. In this study large-scale track models with trapezoidal section area were constructed at the University of Massachusetts to investigate the effects of breakdown fouling, and the effects of changing geotechnical properties on GPR traces. This paper presents the design and construction of large scale track models, and methods used for GPR data collection. GPR data are presented in this paper that demonstrate sensitivity to the track model properties and variables. In particular, the experiments are being used to evaluate changes in GPR data with changing geotechnical properties of the ballast such as density, water content, grain size distribution (GSD), and fouling percentage.

This content is only available via PDF.
You do not currently have access to this content.