Development and Comparison of Benchmarking Methods for Electric Vehicle Drive Units

At the U.S. Environmental Protection Agency’s National Vehicle and Fuel Emissions Laboratory, a development project was implemented to compare various test methods for benchmarking the operation of vehicle electric drive units (EDUs). In earlier research, several test methods were identified, of which two were used to test a Chevrolet Bolt EDU: (a) in-vehicle testing of the complete EDU on a chassis hub dynamometer and (b) stand-alone testing of the EDU’s electric motor and inverter in a dedicated test cell after removal from the vehicle. The resulting data sets were compared with each other and with similar data previously published by GM. In this paper, additional EDU test methods are explored. First, the stand-alone testing of the EDU and its subcomponents is expanded to include testing both with and without the EDU gearing. This testing allows the electric motor, inverter, and gearbox to be characterized separately and the EDU to be characterized as a complete unit. Second, in-vehicle testing of the complete EDU is performed on a roll dynamometer. EDU efficiency and losses are determined using data obtained from wheel hub torque sensors; these results are compared to estimated values based on the force and speed data collected by the roll dynamometer. Finally, the relative difficulty of implementation and the robustness of the resulting data for each of the benchmarking methods developed is analyzed and compared to methods developed and described in earlier research. Metrics used for comparison include the cost and level of effort of the implementation, complexity of the setup, repeatability of the test conditions, safety considerations, ability to separate EDU components, completeness of the data set produced, and the variability, quality and accuracy of the resulting data.