eJournals International Colloquium Tribology 23/1

International Colloquium Tribology
ict
expert verlag Tübingen
125
2022
231

Predicting Electric Vehicle Transmission Efficiency Using a Thermally Coupled Lubrication Model

125
2022
Joseph F. Shore
Athanasios I. Christodoulias
Anant S. Kolekar
Frances E. Lockwood
Amir Kadiric
ict2310453
453 23rd International Colloquium Tribology - January 2022 453 Predicting Electric Vehicle Transmission Efficiency Using a Thermally Coupled Lubrication Model Joseph F. Shore Imperial College London, London, UK Corresponding author: joseph.shore15@imperial.ac.uk Athanasios I. Christodoulias Imperial College London, London, UK Anant S. Kolekar Valvoline, Lexington KY, USA Frances E. Lockwood Valvoline, Lexington KY, USA Amir Kadiric Imperial College London, London, UK 1. Introduction Range anxiety remains a primary concern for electric vehicle (EV) consumers, therefore, increasing vehicle efficiency is an important focus for manufacturers. Since electric motors (EMs) have far greater efficiency than internal combustion engines (ICEs), power losses in the transmission account for a greater percentage of the overall losses than in an equivalent ICE vehicle. Thus, reduction in these losses could lead to significant gains in overall vehicle efficiency. Since EV gearboxes are required to operate at high torque, low speed conditions at one extreme and very high speed and low torque at the other extreme, there are competing requirements on lubricant performance. This work presents an efficient numerical model to predict power losses and hence efficiency in a typical EV transmission. The model considers the gear meshing losses, bearing losses and gear churning losses. Importantly, the model accounts for the evolution of lubricant temperature in the gearbox during a drive cycle accounting for heat transfers to the surroundings as well as oil cooling the EM, thereby allowing an entire drive cycle to be analyzed. The model is able to identify inefficiencies within the gearbox design and can distinguish nominally similar lubricants by accounting for differences in lubricant rheology. 2. Model Description 2.1 Gear Meshing Losses Similarly to [1], the path of contact between the gear teeth is discretized into many points and the local gear geometry at each point is used to determine the contact pressure. At each point, the thermal coupling between coefficient of friction (COF), film thickness, oil film and gear bulk temperatures is accounted for by using an iterative procedure. The COF is determined by considering the rheology of the lubricant based on the Eyring stress using an algorithm developed by Olver and Spikes [2]. The value of the Eyring stress is determined by interpolating between experimentally derived values from tribometer tests on the lubricant, further discussed in section 3. By accounting for the lubricant’s rheology, this method allows for nominally similar lubricants to be compared in terms of their impact on EV gearbox efficiency. Since gears frequently operate in the mixed lubrication regime, it is necessary to consider the boundary COF as well as the EHL COF. The overall gear mesh COF is calculated as an average of the fluid COF and the experimentally determined boundary COF, weighted by the lambda ratio, as described in [2,3]. 454 23rd International Colloquium Tribology - January 2022 Predicting Electric Vehicle Transmission Efficiency Using a Thermally Coupled Lubrication Model 2.2 Bearing and Churning Losses The gear loss model is combined with existing numerical model for bearing losses developed by Morales-Espejel [4] at SKF and gear churning losses described by Changenet et al. [5] to create a complete efficiency model of the gearbox. 2.3 Thermal Coupling to the Environment To analyze a gearbox over an entire drive cycle, it is necessary to account for the evolution of lubricant temperature over the course of the cycle. To do this, a thermal model of the transmission has been devised which accounts for heat lost through a heat exchanger and the heating of the lubricant by losses in the transmission and EM. EM losses are estimated from an efficiency map and the heat exchanger is accounted for by using coolant and oil flowrates and coolant temperature through the heat exchanger measured from the real vehicle. 3. Determination of Lubricant Parameters The Eyring stress was found at various contact pressures and temperatures by implementing the procedure developed by Lafountain et al. [6]. Traction tests were performed using a PCS ETM rig to obtain the COF at various slide-roll ratios at several temperatures and pressures representative of EV gear conditions. The results were combined with film thickness measurements to create a plots of shear stress against strain rate from which the Eyring stress could be determined. The boundary COF of the oil was determined via a series of Stribeck curves produced from traction tests performed on a PCS MTM rig. 4. Model Validation The accuracy of the model’s predictions was assessed by comparing predicted evolution of temperatures within the gearbox to experimental temperature measurements on a real-world road test with an EV installed with a similar gearbox. The temperature evolution predicted by the model showed good agreement with the temperature measurements, as shown in Figure 1. Figure 1: Predicted temperatures compared to measurements for a real-world EV road test 5. Breakdown of Power Losses Figure 2 shows the sources of losses within a typical EV gearbox with varying torque and speed at a constant input power of 15 kW. Figure 2: Breakdown of losses with varying speed at constant 15kW power This figure allows us to identify the major contributors to power losses within the transmission and thereby identify inefficiencies in the transmission design. Bearing losses are the greatest contributor to power losses, especially at higher speeds. Churning losses are negligible at low speed but account for significant losses as speed increases, emphasizing the importance of lubricant selection to minimize this loss type. Gear losses make up a substantial proportion of the total power losses under high torque, low speed conditions but reduce as the speed increases in this example, primarily due to the increase in specific film thickness. 23rd International Colloquium Tribology - January 2022 455 Predicting Electric Vehicle Transmission Efficiency Using a Thermally Coupled Lubrication Model Considering the total powertrain losses, with an EM efficiency of around 90%, the total transmission losses account for between 15 to 25% of the total powertrain losses under these conditions, underlining the potential for improving vehicle efficiency and range by optimizing transmission design and lubricant selection. 6. Conclusion • A thermally coupled model of a full EV transmission has been presented. The model accounts for lubricant rheology and is computationally efficient enough to allow an entire drive cycle to be analyzed • Predictions using the model show good agreement with results from real world drive cycle tests on a real EV • The model allows for the sources of losses to be analyzed, allowing inefficient components to be identified, aiding gearbox design • The model is able to distinguish between two nominally identical lubricants in terms of their impact on transmission efficiency and hence can aid in early lubricant selection. References [1] A. Christodoulias, “Prediction of Power Losses in an Automotive Gearbox,” PhD Thesis, Imperial College London, 2017. [2] A. V. Olver and H. A. Spikes, “Prediction of traction in elastohydrodynamic lubrication,” Proc. Inst. Mech. Eng. Part J J. Eng. Tribol., vol. 212, no. 5, pp. 321-332, 1998, doi: 10.1243/ 1350650981542137. [3] M. Smeeth and H. A. Spikes, “The influence of slide roll ratio on the film thickness of an EHD contact operating within the mixed lubrication regime.” Presented at the Twenty-second Leeds-Lyon Symposium on Tribology, The Third Body Concept, Lyon, France, 5-8 September 1995 [4] G. Morales-Espejel, “Using a friction model as an engineering tool,” Evolution SKF, vol. 2, pp. 27- 30, 2006. [5] C. Changenet and P. Velex, “A model for the prediction of churning losses in geared transmissions - Preliminary results,” J. Mech. Des. Trans. ASME, vol. 129, no. 1, pp. 128-133, 2007, doi: 10.1115/ 1.2403727. [6] A. R. Lafountain, G. J. Johnston, and H. A. Spikes, “The elastohydrodynamic traction of synthetic base oil blends,” Tribol. Trans., vol. 44, no. 4, pp. 648- 656, 2001, doi: 10.1080/ 10402000108982506.