An experimental study into the onset of scuffing using a new contra-rotation test method

Abstract

Scuffing is a surface failure mode that occurs in sliding-rolling contacts subjected to high loads and high sliding speeds, such as those in gears and cam-followers. Scuffing occurs due to excessive adhesion between contacting surfaces and progresses very quickly owing to a strong thermal feedback mechanism. The damage is often catastrophic and occurs without warning, necessitating the replacement of damaged components. This can be costly, not only due to the cost of components but also because of the associated downtime of the affected machine. Therefore, scuffing represents a significant design constraint in many applications and there is a strong incentive for eliminating its occurrence. However, owing to its sudden onset, rapid progression and the influence of both fluid and boundary lubricant films, scuffing is difficult to study in a systematic, repeatable manner. A review of relevant literature identified the need for a reliable scuffing test that can better define the tribological conditions at the onset of scuffing than is possible with the current test methods and that is capable to reliably producing scuffing with modern gear oils which manifest improved scuffing resistance. Consequently, this thesis describes the development and application of a scuffing test method based on contra-rotation which is able to meet these criteria. The method enables the sliding speed to be decoupled from the rolling speed, so that the scuffing properties of a lubricated system can be determined under controlled lubrication conditions over a wide range of sliding speeds. Because both surfaces move relative to the contact, wear is distributed over a large area, minimising changes in contact pressure as test progresses, while the use of step-wise increasing sliding speed rather than increasing load, avoids the risk of sudden scuffing onset which may otherwise occur with increasing load due to introduction of new fresh surfaces into the contact. The introduction of a short ‘rest’ step ensures that adverse effects of frictional heating are minimised. The same rest-stage is used to acquire an interferometry image of any boundary film, which in turn enables the mechanisms of scuffing to be studied more closely. The test method was implemented on a ball-on-disc tribometer, where the two contacting bodies are driven independently to achieve contra-rotation motion. The optimisation of the test method performed here significantly improves the repeatability of the test, with the variability in results shown to be minimal so that three repeats are judged to accurately define the conditions at the onset of scuffing. This was achieved through optimisation of the running-in procedure and a detailed investigation of the conditions leading to potential occurrence of a central dimple under contra-rotation so that test conditions can be set to avoid its adverse effects on test results. The test method was shown to be able to generate scuffing damage that is representative of real applications, with good repeatability and minimum variation in results. The test was shown to be able to successfully differentiate between different oil formulations, materials and surface finishes in terms of their scuffing resistance. The study then applies this method to assess the effects of various lubricant formulations, roughness levels and surface material types on the scuffing behaviour. A wide selection of conditions is investigated in order to not only reproduce the onset of scuffing, but also to study the transition behaviour from non-scuffing to scuffing region as the contact conditions vary. Through this broad range of results, the work provides new insights into scuffing through better understanding of influencing factors and mechanisms of their action. Firstly, the onset of scuffing with simple oils is studied, including pure base oils of Groups I, III and IV, as well as simple model blends of Group III and various concertation of sulphur to act as an extreme pressure additive. Following this, the test was used to carry out a series of scuffing tests with commercially available oils from various real applications, including aviation turbine oils and gear oils. These lubricants are selected for availability of scuffing data from alternative tests. They cover a wide spectrum of base oil types, viscosities and additive packages. The results allow for the effects of lubricant formulation on scuffing to be inferred and for the oil ranking obtained in the present test to be compared with that from alternative tests. Finally, results are presented to illustrate potential effects of surface topography and material on scuffing performance. The results show that the onset of scuffing is characterised by a sharp and unrecoverable increase in friction, and accompanied by the destruction of any boundary films. All tests show that the relationship load×speedn holds at scuffing for each combination of lubricating oil and contact materials. The exact value of the exponent n is dependent on the oil-material combination and in present tests ranges between 1.7 and over 3. Suitable additive packages, particularly those containing sufficient amount of sulphur, are able to significantly improve scuffing resistance owing to their ability to postpone the uncontrollable rise in friction to higher sliding speeds, by arresting any scuffing that initially occurs on micro level. Hard, low-friction DLC coating was also shown to retard the onset of scuffing. When the effect of viscosity is excluded, fully formulated oils from the same series, and hence relatively similar additive packages, were shown to have the same scuffing performance in the present test. The results also suggest that too smooth and too rough surfaces can promote the onset of scuffing so that there may be an optimum roughness level for improved scuffing performance. Finally, the obtained results were used to investigate the applicability of two most widely accepted scuffing criteria, the critical maximum contact temperature and the critical value of frictional power intensity. It was concluded that, although a good agreement exists in some cases, particularly for lightly additivised oils, scuffing failures could not universally be predicted by either of these two criteria. The experimental results suggest that the strength of the tribofilm, which is largely due to additive packages, the fluid retention capability, which is due to the roughness profile, any surface treatment and the rolling speed, all heavily influence overall scuffing performance but are not all accounted for by the conventional scuffing criteria. Consequently, the application of the conventional scuffing criteria should be used with caution and appropriate tests should be performed to reliably establish the scuffing resistance of a given tribological system.