Sound pressure prediction of an electric powertrain performed in an engine dyno test cell using 3D sound intensity
ABSTRACT:
Increasing pressure from the consumer market challenges automotive OEMs to meet customer expectations for NVH performance in ever more efficient ways. The rapid development pace of new vehicles results in short development cycles, where vehicle testing is kept to a minimum and preferably focused on individual sub-components. Complex elements, such as electric powertrains, are often designed to be compatible with multiple vehicle models, so as to maximize their usage while effectively reducing their development cost.
As a result, a thorough characterization and understanding of these primary components are key to preventing potential issues. From an acoustic perspective, this scenario can be simplified by characterizing the vibro-acoustic emission of the source and the propagation paths toward the car passengers separately. A detailed characterization of the sound field around the engine will not only help predict the sound pressure perceived inside the cabin but also understand the impact of individual elements on the noise emitted. In this paper, we present a novel technique using multiple 3D sound intensity probes and a three-dimensional tracking system to measure the sound radiated by an electric powertrain during Wide-Open Throttle (WOT) on a dyno test cell. The 3D radiation data is then combined with acoustic transfer functions measured reciprocally on a full vehicle. This approach enables the synthesis of the individual contributions of the Electric Drive Unit (EDU), junction box, and inverter for a particular vehicle model but it can also be extended to other vehicle implementations just by measuring a new set of acoustic transfer functions.
Experimental results are compared with full-vehicle measurements performed on a roller test bench. Results demonstrate the effectiveness of the proposed approach, enabling the prediction of key acoustic features induced by the powertrain inside the cabin while identifying the main areas responsible for the noise emission on critical excitation bands. Furthermore, the usage of order and frequency filtering on the 3D sound visualization maps was proven to be very useful for troubleshooting purposes. In conclusion, the proposed methodology can be used to improve the refinement process of an electric powertrain in an easy and intuitive way, enabling it to identify areas of high noise radiation and predict potential problems of an electric powertrain mounted into different vehicle implementations.
CITATION
Güven, E., D. Fernandez-Comesaña, D., Marini Storani T., (2022).Sound pressure prediction of an electric powertrain performed in an engine dyno test cell using 3D sound intensity In Proceedings of Aachen Acoustics Colloquium.
USP regular | 3D Acoustic Vector Sensor
Acoustic Shape