Traveling by train is becoming increasingly important, and, especially for long-distance travels, the current societal pressure for decarbonization is accelerating a modal shift from air travel to train travel.
With the introduction of faster and more comfortable trains, roll
...
Traveling by train is becoming increasingly important, and, especially for long-distance travels, the current societal pressure for decarbonization is accelerating a modal shift from air travel to train travel.
With the introduction of faster and more comfortable trains, rolling contact fatigue (RCF) damage in rails started to spread over the networks, already in the 1990s. The RCF is controllable with newly introduced railhead profiles and with rails made of steels with improved wear resistant properties. Rails nevertheless need frequent maintenance, and for that purpose, high-powered precision grinding trains were developed. In Chapter 1 the backgrounds and motivation to understand the mechanisms for damage development, in particular the relationship between rail grinding and the durability of the resulting rail contact surface, are presented.
Despite the proven rail life extension due to grinding maintenance, grinding-related rail damage may occur. The fixed mutual distance between initiations is a specific characteristic of this damage type. A second characteristic is the complete removal of the running surface during ongoing crack propagation. Grinding-related damage is mainly detected in the newly introduced rail steels with improved wear resistance.
Limited empirical evidence is available to explain the initiation of damage from the surface condition of the ground rail. The empirical evidence that is available points to an initiation mechanism involving deep grinding grooves and surface heating and rapid cooling during grinding, which produces hard and brittle white etching layers (WEL).
With grinding being an irreplaceable maintenance activity, the objective of the research is to obtain metallurgical understanding of the relation between the characteristic features of the freshly ground surface, the formation of the contact surface, and damage initiation. This understanding will support decisions on rail grinding design and further research into rail grinding and the development of applications, thereby avoiding the introduction of preferential locations for damage initiation.
Chapter 2 presents the results of a track test. In this test the processes acting on the freshly ground rail are studied in a medium-wide curve of a live track. This experiment is designed as follows. A new rail is installed before the start of the experiment and subsequently in-situ ground according to the ProRail specifications for preventive rail grinding. Inspections are performed during eight months to evaluate the contact surface formation and four rail samples were extracted for detailed tribological and metallographic evaluation.
First, the freshly ground rail surface is studied, and the surface condition is characterized. The grinding introduced facets on the rail head, and grinding roughness is present on these facets. This roughness has a normally distributed variation in height profile. Friction between abrasives and the rail surface caused deformation beneath the surface and surface temperatures to rise to above AC3 leading to fresh WEL formation. Ploughing, typical for grinding processes, has resulted in the formation of ridges and slivers, which partially cover the freshly ground surface.
Inspection results and micrographs have revealed that roughness asperities are deformed and extruded under recurring wheel contacts, a process that proceeds fastest at the facet transitions. Most of the grinding-related WEL is fractured during the deformation processes and removed from the surface and fresh WEL is formed due to frictional heat generation during deformation.
The observed roughness-reduction mechanism contributes to the fast formation of the contact surface but carries the risk that detrimental particles cannot escape and become trapped. The results further show that especially deep grinding grooves containing WEL facilitate the initiation of damage. The characteristic ground surface features, successive steps in contact surface formation and damage initiation are captured in a novel schematic 5-stage wear model for ground rail surfaces.
The case study in Chapter 3 is performed on rail samples extracted from the railway line Zutphen-Hengelo in The Netherlands. The objective of the study is to evaluate damage formation in rails subjected to light bi-directional traffic after grinding. For that purpose rail surface conditions, subsurface deformation patterns, and representative surface breaking damages are studied.
The rail surface shows, three years after the grinding maintenance, that grinding roughness is still present on the running surface, while on the gauge corner the roughness has worn away. These differences in wear are explained by the bi-directional use of the railway line. Tangential shear stresses acting on the gauge corner are independent of the direction of travel, causing locally a ‘normal’ wear rate whereas tractive shear stress reversal reduces the wear rate on the running surface.
The S-shaped deformation patterns of the lamellae of the pearlitic rail steel beneath the contact surface are explained by the tractive shear stress reversal and cracks propagate symmetrically in both running directions.
Finally, the damage initiation is explained by two, line-specific conditions that occur
simultaneously. First, when traffic is light, rail wear is low and this wear rate is further reduced by the rolling direction reversal. Observations of the rail surface showed that these specific loading conditions prevent the complete removal of the ground surface features and thus the formation of a durable load-carrying surface. Second, rail corrugation is not fully removed by grinding, causing variations in the wheel contact conditions.
The case study shows that the sustainable maintenance of rails in lightly loaded railway lines requires a distinct specification, with special attention to the surface roughness
after grinding and to the removal of short-pitch corrugations.
Chapter 4 presents the results of an experiment using a twin-disc test setup in which five representative surface conditions that may occur after rail maintenance, are studied. The objective of the experimental design was to evaluate the contact surface formation, the number of load cycles until the coefficient of friction between the two surfaces increases, and the level that is reached.
Periodic rail maintenance removes damage in a timely manner, but all characteristic surface features must also be worn away again and again by the passing wheels in order to obtain a smooth rail surface.
The experimental results show that the initial coefficient of friction is low. This is a known characteristic of freshly machined surfaces, but it is not reflected in the considerations in rail maintenance as found in the literature.
The results also reveal that when the contact surface is formed by asperity deformation, the coefficient of friction rises already after a low number of load cycles and a high level is reached. When deformation is insufficient and wear have to take place, the process takes more load cycles.
The third observation worth mentioning is that when rail maintenance leads to subsurface deformation and strain hardening, it can take longer for breaking-in processes to be completed.
The experimental results contribute to the determination of specific requirements for, for example, tool management during the performance of rail maintenance.
The railway industry is constantly striving for performance improvements, for example by developing rail steels with improved wear resistance, which can extend its service life. The increased wear resistance may result from controlled accelerated cooling after hot rolling. Another method is to alter the chemical composition. Chapter 5 presents the results of an experimental study to determine the mechanical properties of a novel air-cooled, vanadium-alloyed, hypereutectoid rail steel. Fractographic evaluations are made to study the crack paths. The experimental results are compared with the performance of standard rail steels and a controlled accelerated cooled rail steel.
The air-cooled hypereutectoid rail steel and the hypo-eutectoid controlled accelerated cooled rail steel that are compared, exhibit comparable hardness, the material property that is traditionally used for the classification of rail steel grades. The mechanical behaviors of both rail steels are different. For example, the strain-hardening capacity and the crack growth rate of the air-cooled steel is higher compared to the controlled accelerated cooled rail steel.
The results of the linear elastic fracture mechanics tests are explained by means of a detailed characterization of the microstructure. Therefore the investigation contributes to the development of knowledge on the microstructure-mechanical relationships of pearlitic rail steels. In addition, the results also contribute to considerations in the selection of railway steel for specific applications.
Finally, Chapter 6 discusses the main conclusions of the research presented in this thesis and the consequences for the performance of rail maintenance. It is concluded during the project that research on damage that may initiate on the freshly ground surface is a green field and, as a result, extensive research is necessary into various aspects.
Recommendations for further research comprise the quantification of detrimental features of the freshly ground surface condition that were identified, the determination of limits of acceptable presence of these features, and the development of maintenance applications that are robust against the variations encountered in track.
