Abstract
This paper presents findings of a railroad ballast study using the discrete element method (DEM) focused on mesoscale performance modeling of ballast layer under different tie support conditions. The simulation assembles ballast gradation that met the requirements of both American Railway Engineering and Maintenance-of-Way Association (AREMA) No. 3 and No. 4A specifications with polyhedral particle shapes created similar to the field-collected ballast samples. A full-track model was generated as a basic model, on which five different support conditions were studied in the DEM simulation. Static rail seat loads of 10 kips (44.5 kN) were applied until the DEM model became stable. The pressure distribution along the tie-ballast interface predicted by DEM simulations was in good agreement with previously published results backcalculated from laboratory testing. Static rail seat loads of 20 kips (89 kN) were then applied in the calibrated DEM model to evaluate in-track performance. Results from the validated full-track DEM simulations indicated that only a small portion of ballast particles participated in load distribution under static loading. Particles on the shoulders and particles in the areas with poor support conditions often experience no or very low contact forces. Load transfer mechanisms investigated through a contact force network varied greatly among different support conditions: lack of rail seat support, full support, and lack of center support had wider force distribution angles than the high center binding and severe center binding conditions. The severe center binding scenario was found to be the most critical support condition in terms of causing the highest tie-ballast contact pressure exceeding 30% of the AREMA allowable pressure.