A Novel Calculation Method to Quantify the Torque Dependency of the Masking Threshold of Tonal Powertrain Noise in Electric Vehicles

Tonal powertrain noise can have a strong negative impact on passengers’ quality and comfort perception in the interior of electric vehicles. Therefore, in the vehicle development process, the assessment of the perceptibility of tonal powertrain noise is essential. As wind and tire noise can possibly mask tonal noises, engineers use modern masking models to determine the masking threshold of tonal powertrain noise from vehicle interior measurements. In the presently used method, the masking threshold is mostly generated with torque-free deceleration measurements. However, the influence of torque on masking tire noise must be considered. As this requires time-consuming and costly road measurements, an extension of the method is being developed, which will also enable the use of roller dynamometer measurements for the assessment. For the extension of the method, however, the influence of the torque must also be considered. This paper presents a novel calculation method that quantifies the influence of torque on the masking threshold and converts masking thresholds from an arbitrary torque level to another. By identifying the frequency and speed range that is mainly affected by the torque-dependent tire noise, a regression model with respect to the tractive force on the tires can be used to calculate a torque-dependent correction factor. The developed method can significantly improve the validity of masking thresholds and quantitatively, the method generalizes well across different vehicle segments. The error can be reduced to below 2 dB above 2000 rpm and to below 1 dB above 4000 rpm. By using this method, more valid target level settings for tonal powertrain noise can be derived.