Producing and Measuring the 3rd Body Layer

Abstract

Decades of measuring friction conditions at the wheel-rail interface have resulted in widely varying answers produced by different measurement methods and devices. This variation is a result of many factors, some of which are difficult to quantify and control, especially under field conditions. One of these variables is the presence of an unknown 3rd body layer material, whose accumulation on the wheel and rail surfaces affect the value of independent tests performed to quantify the nominal state of wheel on rail in actual operating conditions. This paper describes a series of tests undertaken to understand the creation of the 3rd body layer from the starting point of clean wheel and rail surfaces and how that can lead to new standard practices for field testing. A 2 × 2 matrix of wheel and rail conditions is defined to help formalize the discussion of the production and measurement of the 3rd body layer. Case 1 is the nominal virgin state of the two materials. This is actually fairly hard to produce. Almost any process and/or handling will leave some films on both surfaces. Lab tests typically start in this state after cleaning/degreasing the surfaces following machining or sanding. Cases 2 and 3 are potentially the same (clean wheel or rail), however, the more likely case is Case 2 where some existing location on actual track is chosen to determine its current state, thus the use of a Tribometer using a clean wheel. Case 4 represents the nominal state of most track with passing vehicles where the current environmental/ambient conditions encourage a particular set of 3rd body wear products. This formalization helps reveal the prior common practices that have produced assumptions of what is the ‘correct’ value and how that has created a bias in one’s thinking of what really exists in normal revenue service. This issue is particularly consequential in attempting to quantify the benefit of the friction modifiers that are being adopted across the railroad industry.

Challenges and progress in the understanding and modelling of the wheel–rail creep forces

“…This is attributed to the increased surface area on which the oil is spread, reducing the amount of oil on the rail. A much stronger influence of wear debris was found recently by experiments at the Virginia Tech roller rig [87] and by Harrison using his hand-operated tribometer [84], as discussed further in Section 4.2.…”

“…However, that does not do justice to the knowledge and skills of the researchers involved. Further support that low values are true comes from the measurements of Harrison using his hand-operated tribometer [84], showing starting values μ = 0.15 − 0.20 increasing to 'normal' values after a few hundred passes. Voltr and Lata, on the other hand, describe the surfaces cleaned with a solvent as the high friction condition [100].…”

“…A tribometer used in [110] has a driven wheel with a diameter of 100 mm and is attached to the bogie of a vehicle that travels at a speed of up to 80 km/h. Hand-operated tribometers discussed in [84,89,111] employ measurement wheels with diameters 89-254 mm.…”

“…The measured creep force characteristics as a function of pure lateral creep are limited to maximum creep of 1.2% in [102] and 2% in [114]. Bigger values may be achieved on specific experimental devices, such as the VT-FRA test rig that may go up to 10% in principle [101] or the handheld tribometer that goes up to 15% [84]. The lateral creep forces can be measured on real vehicles using force measuring wheelsets.…”