Investigating Morphology and Breakage Evolution Characteristics of Railroad Ballasts over Distinct Supports Subjected to Impact Loading

The ballast bed constantly degrades under the repeated applications of impact loading exerted by passing trains in terms of the particle size, shape, breakage, fouling, etc., thus significantly jeopardizing the in-service performance and operational safety of ballasted tracks. In this study, the morphology and breakage evolution characteristics of railroad ballasts of single- and multiple-size ranges were investigated from laboratory impact-load tests. Both a concrete block and sand layer were placed to mimic the distinct under-ballast supports. The degradation trends of the typical shape and breakage indices were comparatively quantified for different combinations of ballast particle sizes and shapes, under-ballast supports, impact energies, and number of impact-load applications (N). The results show that both shape and size affect ballast particle breakage, with shape being more influential. The breakage severity of flake-like particles is about 1.5–1.66 times and 1.25–1.5 times higher than those of regular and needle-like particles, respectively. Under impact loading, large and small single-size ballasts degrade mainly by breakage and abrasion, respectively. The modified fouling index (FI) of flake-like particles within 31.5–40 mm is about 3.6 times that of regular particles within 50–63 mm. The shape indices of the ballast particles within 31.5–40 mm exhibit the most profound changes. The severities of the ballast breakage and fines generation (or modified FI) increased by 50% and 74%, respectively, due to the increase in the under-ballast support stiffness by 100 times and the drop height of 80 cm, respectively. The convexity and ballast breakage index (BBI) are promising for quantifying particle-degradation trends, and their statistical correlation found herein is potentially useful for the transition of ballast-bed-maintenance management from the current plan-based scheduling to condition-based upgrading.