This study investigates the film formation and friction in grease lubricated, rolling-sliding, non-conformal contacts over a range of entrainment speeds, surface roughnesses and contact temperatures. The effects of grease composition are assessed by employing custom made, additive free, lithium and diurea thickened greases, whose composition is systematically varied so that the effects of the thickener and the base oil can be isolated. All film thickness and friction measurements were conducted under fully-flooded conditions. It was found that at low speeds all tested greases are able to form thicker films than the corresponding base oils. The thickness and behaviour of these films is determined by the thickener type and is independent of the base oil viscosity and the test temperature. At higher speeds, the film thickness is governed by the base oil properties alone and can be predicted by the EHD theory. At low speeds, films with diurea greases grow with time under constant speed and residual films persist under load after contact motion ceases. The real lambda ratio, based on measured grease film thickness, was shown to correlate well with contact friction. The transition from the thickener dominated behaviour to that dominated by the base oil occurs at a relatively constant film thickness, regardless of the base oil viscosity and test temperatures, rather than at a given entrainment speed. Based on the presented evidence, it is here proposed that the mechanism of formation of grease films at low speeds, is analogous to that reported to operate in EHL contacts lubricated with colloidal dispersions, namely the mechanical entrapment and deposition of thickener fibres, and that, rather than the widely quoted ???transition speed???, it is the ratio of the thickener fibre size to prevailing film thickness that determines the range of conditions under which the film enhancement due to the action of thickener is present

This article evaluates the frictional performance of different bearing grease formulations in full rolling bearings and a ball-on-disc rig and subsequently assesses whether the ball-on-disc test results can be used to predict the grease performance in actual bearings. A selection of custom-made greases with systematically varied formulations as well as their base oils were tested. Bearing torque was measured in two different cylindrical roller thrust bearings and a thrust ball bearing. The same lubricants were tested with ball-on-disc tribometers, a mini traction machine (MTM) to measure friction and an optical elastohydrodynamic (EHD) rig to measure film thickness. Both lithium complex and diurea greases were observed to produce lower friction than their base oils within the low speed, low nominal lambda ratio region, whereas the greases and oils had the same friction at high nominal lambda ratio values. These relative trends were the same in full bearing and single-contact MTM tests. The reduction in friction was seen to be related to the level of film thickness enhancement provided by greases at lower speeds, which leads to an increase in the effective lambda ratio and hence reduced friction. By extracting the sliding torque component from the overall measured bearing torque, a plot of the friction coefficient against the effective lambda ratio was produced encompassing all bearing and single-contact tests and all lubricants and test conditions. This plot was seen to follow a general shape of a master Stribeck curve, indicating that the numerical values of the friction coefficient from ball-on-disc and full bearing tests overlap and can be related to each other using this approach over the range of conditions employed here. Thus, single-contact ball-on-disc tests can provide a fast and economical way of establishing the frictional performance of bearing greases in full bearings in terms of both relative performance rankings and quantitative values of bearing frictional power losses.