Damage mechanisms such as surface cracks and wear on a rail can reduce the service life of a railway track. The purpose of this investigation was to study the development of these two damage mechanisms on new and 3-year-old rails in a commuter railway track over a period of 2 years. Four curves were studied with radius between 303 and 616 m. In two of the curves, two different kinds of rail steel grade (UIC 900A grade with ultimate strength 900 N/mm² and UIC 1100 grade with ultimate strength 1100 N/mm²) were used in each curve. In the other two curves, only the lower-strength rail was used. Four pieces of new rail, each 20 m long, were inserted in the two curves with both UIC 900A and UIC 1100 grade rail. Lubrication was applied on the high rail of one of the curves with both UIC 900A and UIC 1100 grade rail and on one of the curves with only UIC 900A grade rail. The two remaining curves were not lubricated. Surface cracks in the form of headchecks could be noted on the surface of the new 1100 grade rails after 1 month of traffic. By contrast, the surface of the UIC 900A grade rails showed visible surface cracks in only two of four curves and that after approximately 2 years of traffic. Both materials seemed to be similarly sensitive to crack initiation but the 1100 grade rail was more sensitive to crack propagation and also more sensitive to the formation of headcheck cracks. Lubrication, as expected, reduced the profile change. A less expected outcome was that lubrication also reduced the rate of crack propagation; however, the lubricated UIC 1100 grade rail was as sensitive to crack initiation as the unlubricated UIC 1100 grade rail. By comparing the wear depth in the headcheck zone with the crack length, equilibrium between these two damage mechanisms was found for the lubricated UIC 1100 grade rail. Both the crack length and the wear depth showed low values. By using a lubricant with friction modifiers the stresses was low enough to prevent crack propagation; at the same time, the rail was hard enough to reduce the wear rate. This is probably the most favourable state in terms of rail maintenance cost.

A specially designed test system involving boundary lubricated roller bearings was used to study wear at low particle concentration levels. A separate oil system circulated the oil through the test bearings. The effects of self-generated contaminants from the system were studied. Even at very low concentration levels, self-generated contaminants can cause significant wear. The concentration of self-generated particles was very high during the running-in period. It is therefore important that the filtration be very efficient during this period. The experimental results show that filtration during run-in for 1 h with a 3 μm filter can reduce both the mass loss and the number of self-generated particles by a factor of 10. Furthermore, the results also show that while the bearings with standard rollers can have significant wear, those with coated rollers are at the same time almost unaffected by wear. Also, the number of particles generated in the contact was significantly less when using coated rollers. There were twice as many self-generated particles when using a standard bearing as those compared with a coated bearing.

Damage caused by particles within rolling/sliding contacts can severely reduce the operational life of machinery such as roller bearings, gears and pumps. Abrasive wear of spherical roller thrust bearings has been studied using a stylus apparatus and scanning electron microscopy (SEM). Both a standard bearing and a bearing with rollers coated with metal mixed amorphous carbon (Me-C:H) were studied. The SEM measurements were performed systematically across the contact surfaces so that surfaces with gradually different contact situations could be examined. These measurements were compared to the measured wear depth of the components of the roller bearing. Also, the calculated contact conditions in terms of creep, contact size and surface separation have been related to the observed wear pattern at various locations. To attempt to understand the wear behaviour of the bearing with coated rollers, the coating as well as the material content of the surfaces were examined using both SEM and energy dispersive X-ray spectrometry (EDS). This revealed that the coating did not flake off but rather was scratched off. It is possible to link the abrasive wear behaviour to the contact conditions. It is crucial to understand this relationship when building a simulation model of abrasive wear.

Damage caused by lubricant borne particles in rolling/sliding contacts can severely reduce the operational life of machine elements such as cam mechanisms, roller bearings, gears, and pumps. Lubricant supplies frequently contain such contaminating particles, either generated from within the machinery itself or entrained from the surroundings. The particle can be entrained into a lubricated contact and damage the bearing surfaces. Many such individual abrasive actions can lead to significant change in the surface profile of the rolling elements.

In this work, a series of experiments has been carried out to investigate the mechanism of this surface damage and abrasion process when the contaminating particles are small and hard. The tests show, how particles are entrained into the contacts, the form of the scratches they produce, and the resulting surface profile changes. On the basis of these observations, a model of the abrasive wear process has been developed. The prediction of abrasive wear compares qualitatively well with observed form change on the bearing surface.