Nachrichten 78 Tribologie + Schmierungstechnik · 66. Jahrgang · 4/ 5/ 2019 GfT-Förderpreis Bachelorarbeit Effect of over rolling frequency on the film formation in grease lubricated EHD contacts under starved conditions* Helko Mues, Dennis Fischer, Georg Jacobs and Andreas Stratmann ** Abstract The service life of rolling bearings is significantly affected by the lubricating film formation in elastohydrodynamic (EHD) contacts. Grease lubricated EHD contacts show a film thickness decay from a characteristic rolling speed, which is referred to as starvation. Thus, the film thickness differs from that of oil lubricated contacts under fully flooded conditions. However, base oil properties under fully flooded conditions are commonly assumed to estimate the service life of grease lubricated bearings, which are usually not fully flooded. However, this assumption results in an overestimation of the film thickness for rotational speeds in the range of starvation, which can lead to uncertainties in the bearing design. The onset of starvation can be related to high rolling speeds, i.e. high over rolling frequencies, when the lubricant does not replenish entirely, so that the contact starves due to insufficient lubricant supply. Therefore, the focus of this contribution is to investigate the effect of over rolling frequency on the lubricating film formation in starved, grease lubricated EHD contacts. The investigations should help to establish an advanced understanding of the mechanisms of grease lubrication, to encourage future work with focus on a method to predict the film formation in grease lubricated EHD contacts. Introduction Rolling bearings are commonly lubricated using oil or grease to separate the contacting surfaces of rolling elements and raceways to reduce friction and wear. Among operational conditions, such as temperature, rotational speed or vibrations, the service life of rolling bearings significantly depends on the lubricating film thickness in elastohydrodynamic (EHD) contacts. The film formation in grease lubricated contacts is determined by the bleed oil, which is released by the grease [1], as well as the grease thickener [2,3]. The thickener forms semi’solid layers on the surfaces or enters the EHD contact as free thickener particles within the bleed oil flow [1,4,5]. These particles are set free by structural degradation of the grease under shear stress during operation [6]. Usually grease lubricated rolling bearings are not fully flooded, so that starvation is likely to occur [7]. Wilson defines the onset of starvation, when the film thickness of the grease lubricated contact drops below the film thickness of the corresponding fully flooded base oil [8]. This definition is used in the following contribution. For an estimation of the service life of grease lubricated rolling bearings, a fully flooded bearing and the corresponding base oil viscosity are commonly assumed [9]. However, this assumption can lead to uncertainties in the bearing design in the range of starvation. If the contact is not fully flooded, the lubricant supply is mainly determined by the bleed oil replenishment near the contact zone [10]. According to the results of [11], the replenishment time, and thus the over rolling frequency, have a significant effect on the regeneration of the oil film in the rolling track. In order to investigate the correlation between the film thickness and the replenishment time in grease lubricated contacts, film thickness measurements on a ball-on-disc tribometer were performed by varying the over rolling frequency. This was achieved by adding a second ball specimen to the tribometer, as shown in Figure 1. The second ball can be adjusted at different angular positions to the measuring ball. Thereby, the over rolling frequency can be increased without changing the rolling speed and thus the centrifugal forces. For the measurements, a polyalphaolefine (PAO) lithium complex grease of consistency class NLGI 2 and its corresponding base oil (η=98 mm²/ s) have been used. A detailed description of the lubricants, the test rig and measurement procedure is given in the full article [12]. Results and Discussion The results of the film thickness measurements at a constant rolling speed of 100 mm/ s using one and two ball specimens are shown in Figure 2. The mean values out of three measurements, including standard deviation er- * This extended abstract is an abbreviated version of the article “Effect of Over Rolling Frequency on the Film Formation in Grease Lubricated EHD Contacts under Starved Conditions”, published in Lubricants 2019, by D. Fischer, H. Mues, G. Jacobs and A. Stratmann. ** Helko Mues, B. Sc. (Bachelorarbeit) Dennis Fischer, M. Sc. Georg Jacobs, Univ.-Prof., Dr.-Ing. Andreas Stratmann, Dr.-Ing. Institute for Machine Elements and Systems Engineering, RWTH Aachen University, 52062 Aachen TuS_4_5_2019.qxp_T+S_2018 23.08.19 13: 15 Seite 78 Nachrichten 79 Tribologie + Schmierungstechnik · 66. Jahrgang · 4/ 5/ 2019 ror bars, are presented. The film thickness of the corresponding base oil under fully flooded lubrication is given as a reference for the lubricating film thickness of the grease measurement. The results illustrate, that during the first 1000 disc revolutions, the film thickness first increases and then decreases, due to the grease distribution on the disc. This initial behaviour is characteristic for grease measurements, and is commonly referred to a churning phase [10,13,14]. For the tests using one ball (0.4 Hz) a constant film thickness is reached after 1000 disc revolutions with a mean value of 110 nm, which is approx. 20 nm higher than the film thickness of the base oil. By applying the second ball, the film thickness reduces in dependency of the angular distance. As presented in Figure 2, at an angular distance of 180° (0.8 Hz) the film thickness reduces to a mean value of 90 nm and at an angular distance of 50° (3.0 Hz) the film thickness reaches a mean value of 50 nm. The second ball displaces bleed oil in the track and thereby reduces the lubricant supply of the measuring ball. Hence, at an angular position of 50°, the high over rolling frequency of 3.0 Hz leads to a starved lubrication, since the film thickness drops below the fully flooded base oil level. However, when the second ball is positioned at 180° angular distance with an over rolling frequency of 0.8 Hz, the replenishment time of the bleed oil is higher, such that the film thickness reaches a value of approx. 90 nm, which corresponds to the fully flooded base oil level. The presented results indicate a dependency of the lubricating film thickness on different angular positions of the second ball specimen, hence on the over rolling frequency and replenishment time. Thus, a stepwise reduction of the resulting film thickness could be shown by a stepwise closer position of the second ball, due to higher over rolling frequencies respectively shorter replenishment times. To investigate the thickener impact on the film formation in grease lubricated contacts, the film thickness was measured at zero speed, after a measurement at 100 mm/ s, using one ball specimen. Thereby, the film thickness could be determined in a non-rotating condition, such that hydrodynamic effects can be excluded. Following this procedure, a mean film thickness of 40 nm results out of six measurements around the track. Due to the absence of hydrodynamic effects, the film thickness of 40 nm can be related to a thickener layer, which forms on the surfaces of the ball and the disc. A detailed description of the measurement procedure and the results are given in the full article [12]. The effect of the thickener layer on the film formation can be seen in Figure 2. When measuring with two balls at an angular position of 50°, the residual film thickness does not reach a zero level, although a lower film thickness would be expected due to the insufficient replenishment at a high over rolling frequency of 3.0 Hz. However, Figure 2 shows, that the film thickness is limited to approx. 40 to 60 nm in the constant range after 1200 disc revolutions, which correlates with the results of the film thickness at zero speed. Thus, it can be concluded, that a thickener layer is formed and supports the separation of the surfaces, although most of the bleed oil in the track is displaced. The assumption of the formation of a thickener layer, which supports the film thickness, is in accordance with the literature [3,15-17]. Figure 1: Ball-on-disc tribometer extended by second ball specimen Figure 2: Film thickness at 100 mm/ s using one and two ball specimens TuS_4_5_2019.qxp_T+S_2018 23.08.19 13: 15 Seite 79 Nachrichten 80 Tribologie + Schmierungstechnik · 66. Jahrgang · 4/ 5/ 2019 Moreover, measurements with increasing rolling speed were performed to investigate the effect of over rolling frequency and replenishment time on the onset of starvation. Figure 3 illustrates the mean values of the film thickness measurement results using one and two ball specimens. The dependency on the rolling speed (left diagram) as well as the over rolling frequency (right diagram) are presented in Figure 3. At low rolling speeds <100 mm/ s, the film thickness with grease using one ball is significantly higher than the base oil film thickness. This result shows a film thickness increase in the range of 15 to 40 mm/ s with increasing rolling speed, indicating a hydrodynamic film formation, which is induced by bleed oil lubrication. However, measuring with two ball specimens, the film thickness is generally lower and the results do not show a hydrodynamic film formation due to the displacement of the bleed oil by the second ball. This leads to an earlier onset of starvation at 90 mm/ s measuring with two balls, in comparison to 180 mm/ s with one ball. The right diagram shows the same measurement results, however, the film thickness is presented as a function of the over rolling frequency. Thus, the film formations for both measurements with one and two ball specimens are in good agreement up to 5.0 Hz as well as the onset of starvation at 0.8 Hz, which corresponds to a replenishment time of 1.25 s. The results presented in Figure 3, confirm the dependency of the film formation on the over rolling frequency, shown in Figure 2. Both measurement results reveal a decreasing film thickness with increasing over rolling frequencies due to lubricant displacement. Thus, the over rolling frequency, respectively the replenishment time, seems to be a significant parameter affecting the lubricating film formation. This correlation is confirmed by the onset of starvation, which occurs at the same over rolling frequency for both measurements, using one and two ball specimens, presented in Figure 3. Moreover, the measurements at zero speed, which indicate a thickener layer on the contacting surfaces, correlate with the results shown in Figure 3. Even at high over rolling frequencies > 9.0 Hz, the film thickness is not reduced to a zero level but reaches a residual level of approx. 60 to 80 nm, due to the formation of a thickener layer. Conclusions In this study, the effect of the over rolling frequency on the lubricating film formation was investigated. Therefore, a ball-on-disc tribometer was extended by a second ball specimen, which was added in front of the measuring ball to increase the over rolling frequency. Using this test setup, the displacement and replenishment of the lubricant can be controlled. The main results are summarized as follows: • The film thickness at a constant rolling speed of 100 mm/ s could be reduced by a stepwise closer position of the second ball to the measuring ball. It could be pointed out, that the film formation in grease lubricated EHD contacts under starved conditions depends on the lubricant supply, which is reduced by a higher over rolling frequency. • Since the film thickness was not reduced towards a zero level using the setup with the second ball, even at high over rolling frequencies >3.0 Hz, it is concluded that the residual film thickness of approx. 40 to 60 nm is formed by a deposited thickener layer on the surfaces. The presence of a semi-solid thickener layer at the rolling track could be verified by measurement results at zero speed, which is in accordance to published studies [3,15-17]. • Using the test setup with the second ball, the effect of replenishment on the film formation with dependency on the over rolling frequency could be emphasized. It has been shown that the lubricating film formation as well as the onset of starvation depends on the over rolling frequency respectively on the replenishment time. Figure 3: Film thicknesses at increasing speed and over rolling frequency TuS_4_5_2019.qxp_T+S_2018 23.08.19 13: 15 Seite 80 Nachrichten 81 Tribologie + Schmierungstechnik · 66. Jahrgang · 4/ 5/ 2019 Acknowledgments: The authors would like to thank the Research Association for Power Transmission Engineering FVA (Forschungsvereinigung Antriebstechnik e. V.) as well as the participating member companies for the support of the IGF project N/ 1 19027, which is funded by the German Federation of Industrial Research Associations AiF (Arbeitsgemeinschaft industrieller Forschungsvereinigungen) within the framework of the program for the promotion of the Industrial Collective Research IGF (Industrielle Gemeinschaftsforschung) by the Federal Ministry for Economic Affairs and Energy BMWi (Bundesministerium für Wirtschaft und Energie) based on a resolution of the German Bundestag. Author Contributions: Helko Mues and Dennis Fischer wrote the article, designed and performed the experiments and analysed the data. Georg Jacobs and Andreas Stratmann supervised the work, discussed the basic design of the experiments and provided suggestions for the final discussion. References [1] Cann P.M., Spikes H.A. Film thickness measurements of lubrication greases under normally starved conditions. NLGI Spokesman 1992; 56(2): 21-7. doi: 10.3390/ lubricants5030034 [2] Morales-Espejel G.E., Lugt P.M., Pasaribu H.R., Cen H. Film thickness in grease lubricated slow rotating rolling bearings. Tribology International 2014; 74: 7-19. doi: 10.1016/ j.triboint.2014.01.023. [3] Cann P.M.E. Understanding grease lubrication. 1996; 31: 573-581. doi: 10.1016/ S0167-8922(08)70819-8. [4] Lugt P.M. A review on grease lubrication in rolling bearings. Tribology Transactions 2009; 52(4): 470- 80. doi: 10.1080/ 10402000802687940. [5] Scarlett N.A. Use of Grease in Rolling Bearings. Proc Instn Mech Engrs 1967; 182(1): 585-624. doi: 10.1243/ PIME_CONF_1967_182_045_02. [6] Kuhn E. Tribological stress of lubricating greases in the light of system entropy. Lubricants 2016; 4(4): 37. doi: 10.3390/ lubricants4040037. [7] Baly H., Poll G., Cann P.M., Lubrecht A.A. Correlation between model test devices and full bearing tests under grease lubricated conditions. In: Snidle RW, Evans HP, editors. IUTAM Symposium on Elastohydrodynamics and Micro-elastohydrodynamics: Proceedings of the IUTAM Symposium held in Cardiff, UK, 1-3 September 2004. Dordrecht: Springer; 2006, p. 229-240. doi: 10.1007/ 1-4020- 4533-6_16. [8] Wilson A.R. The relative thickness of grease and oil films in rolling bearings. Proc Instn Mech Engrs 1979; 193: 185-92. doi: 10.1243/ PIME_PROC_1979_193_019_02. [9] DIN ISO 281: 2010-10. Rolling bearings - Dynamic load ratings and rating life (ISO 281: 2007) 2007. [10]Cann P.M. Starvation and reflow in a grease-lubricated elastohydrodynamic contact. Tribology Transactions 1996; 39(3): 698-704. doi: 10.1080/ 10402009608983585. [11]Kostal D., Sperka P., Svoboda P., Krupka I., Hartl M. Influence of lubricant inlet film thickness on elastohydrodynamically lubricated contact starvation. Journal of Tribology 2017; 139(5): 1-6. doi: 10.1115/ 1.4035777. [12]Fischer D., Mues H., Jacobs G., Stratmann A. Effect of over rolling frequency on the film formation in grease lubricated EHD contacts under starved conditions. Lubricants 2019; 7(2): 19. doi: 10.3390/ lubricants7020019. [13]Lugt, P.M. Grease lubrication in rolling bearings. Chichester: A John Wiley & Sons, Ltd., Publication; 2013. doi: 10.1002/ 9781118483961.ch7. [14]Cousseau T., Björling M., Graça B., Campos A., Seabra J., Larsson R. Film thickness in a ball-ondisc contact lubricated with greases, bleed oils and base oils. Tribology International 2012; 53: 53-60. doi: 10.1016/ j.triboint.2012.04.018. [15]Vengudusamy B., Kuhn M., Rankl M., Spallek R. Film forming behavior of greases under starved and fully flooded EHL conditions. Tribology Transactions 2016; 59(1): 62-71. doi: 10.1080/ 10402004.2015.1071450. [16]Chen J., Tanaka H., Sugimura J. Experimental study of starvation and flow behavior in grease-lubricated EHD contact. Tribology Online 2015; 10(1): 48-55. doi: 10.2474/ trol.10.48. [17]Cann P.M. Grease lubrication of rolling element bearings - role of the grease thickener. Lubrication Science 2007; 19(3): 183-96. doi: 10.1002/ ls.39. TuS_4_5_2019.qxp_T+S_2018 23.08.19 13: 15 Seite 81