DEM modelling of adhesion in sanded wheel-rail contacts

In railways, in the contact between the wheel and the rail forces according to 10 tons wheel-load and higher are transferred on an area of a fingernail. This results in extremely high contact stresses, both perpendicular to the rail surface (normal direction) and tangentially to it (tangential direction), e.g. caused by braking or curving forces. The maximum tangential force that can be transmitted is limited by the maximal adhesion coefficient (AC), which is approx. 0.35 under dry conditions. Contamination of the wheel-rail contact, e.g. by rainwater or fallen leaves in autumn, can reduce the AC, sometimes below 0.1, causing massive problems during braking and, in extreme cases, safety issues. To prevent this and to increase the AC, sanding systems have been used for decades, whereby grains of sand are blown into the wheel-rail contact. Sanding of the wheel-rail contact has been well studied experimentally. Both in the laboratory and on the track, different contact conditions such as dry, wet or different contaminates have been investigated, as well as different types of sand. Under wet conditions, different types of sand can leave the AC in the worst case unchanged or increase it, some almost to the range of dry conditions. The physical mechanisms for this are still poorly understood. Entering the contact, the sand grains are partially crushed, and plastic deformation occurs on the surfaces of the wheel and rail. The AC is possibly increased by form closure effects, when parts of the sand grains are pressed into the metal surfaces, or the ground sand solidifies under the high pressure and thus increases the effective contact area in the wheel-rail interface. The interaction between sand and water in wet contacts is also unclear. This lack of understanding is caused by (todays) inability to experimentally monitor any of the before mentioned mechanisms in the contact zone during roll-over. In order to understand more precisely which mechanisms increase the AC, a simulation model will be developed in the project, using the so-called "Discrete Element Method". This model will take into account the different mechanisms that occur during rollover in the sanded wheel-rail contact. A detailed parameterisation and validation of the model with experimental data is planned. In this way, the model can contribute to a deeper understanding of the adhesion-increasing mechanisms in sanded wheel-rail contacts.

Railway safety and performance depend critically on the frictional contact between train wheels and rails. Under wet or contaminated conditions, so-called low adhesion conditions can occur, meaning that trains have reduced breaking and traction capability. To overcome low adhesion, small amounts of sand are sprayed into the wheel-rail contact. Although sanding has been used for decades, the physical mechanisms that make it effective were not well understood. This knowledge gap limits the ability to optimise sanding strategies and minimise unwanted side effects such as surface damage and excessive wear. A recent research project has now provided detailed insight into how sanding increases adhesion under wet contact conditions. Combining specially designed experiments and advanced simulations using the discrete element method (DEM), the project investigated how sand grains break, deform, and interact with steel surfaces under the extremely high pressures typical of railway operation, comparable to the stress induced by 10 tons weight on the area of a fingernail. The research focused on two commonly used types of rail sand from Great Britain and Austria, with different performance in adhesion tests. In single grain crushing tests under high pressure, repeated sand grain fracture was accompanied by formation of clusters of crushed sand, which then indented into the steel surfaces - with notable differences in in fragment spread between the sand types. Building on these experiments, a novel DEM model was developed - representing, for the first time, all key physical processes occurring during sanding under wet conditions. These include grain breakage, varying fragments spread, cluster formation, steel surface indentation, and shearing of sand fragments. Each mechanism was carefully parametrised using dedicated laboratory tests and the resulting DEM model was validated qualitatively using adhesion tests for both types of sand. Finally, the DEM model was used to investigate why sanding is particularly effective under wet conditions. Results for single sand grains showed that higher adhesion occurs when more of the normal load is carried through sand-steel contacts instead of direct steel-steel contact. This factor is positively influenced by larger sand grain size and greater sand fragment spread. Adhesion also increases with higher steel surface hardness, resulting in less indentation of forming sand clusters, and with higher sand-steel friction. These findings confirm that sand characteristics, especially grain size and wet-condition fragmentation behaviour, play a decisive role in real-world sanding performance. This project marks the first time that the physics of sanded wheel-rail contacts have been understood in such detail. Its outcomes support more effective sanding strategies, help rail operators improve traction under challenging conditions, and contribute directly to safer, more reliable railway networks.