Measuring the contact stiffness at the grain scale of fresh and used granite ballast

Abstract

Faster, heavier and more frequent use is placing greater demands than ever before on railways. These increased demands reduce the time available for maintenance without costly line closures. There is a need to improve the understanding of the performance of key materials used in railway construction to improve their specification and achieve greater durability. A key material is the source rock for rail-way ballast. It is known that different rock sources perform very differently as ballasts and there is a general consensus that igneous rock types perform the best. However, there can still be great variability in performance between alternate igneous rock ballasts and performance may vary with age/use. To better understand ballast behaviour modern numerical tools such as the discrete element method (DEM) are able to model particles at the grain scale and can be used to assess how changes in ballast material and/or changes in performance with age/use may influence railway performance. However, a lack of understanding of inter-particle ballast contact mechanics mean that current DEM simulations lack the necessary input data to provide a high level of confidence in their outputs. To begin to address this, experiments were carried out to understand the normal contact behavior of fresh and used granite type ballast grains under cycles of load at magnitudes representative of in-service use. The grains were tested using a uniaxial loading machine and the deflections measured using micro-photogrammetry and digital image correlation. The tests indicate significantly different load-deformation characteristics for fresh and used granite ballast and present a method for modelling the contact mechanics modified from Hertzian contact theory. Data also shows that used ballast shows less plastic deformation on first loading and could therefore be potentially re-used in track rather than down-cycled as is currently the case in some countries.

More information

Identifiers

ORCID for Geoff Watson: orcid.org/0000-0003-3074-5196

ORCID for Louis Le Pen: orcid.org/0000-0002-4362-3895

Catalogue record

Export record