Vehicle accessory tonal noise: experimental determination and subjective assessment

Introduction: There is a general trend in the motor industry towards reduced noise levels inside vehicles. Low frequency vehicle interior noise, below 500 Hz, is mostly generated through structure-borne vibration emanating from the powertrain and transmitted through the various connection points to the vehicle body and into the passenger compartment. This noise contribution has been reduced dramatically over the past few years by control of the vibration levels generated by the powertrain, engine mount isolation and improved body structures. The consequence of this is that high frequency noise has become much more intrusive because of the reduction in the masking effect of low frequency noise. High frequency noises come from a variety of sources, and are generally airborne rather than structure-borne. Possibly the most annoying to the driver are narrow band high frequency noises, classified as ‘whine’ noises, the majority of which are attributable to noise radiated from the powertrain and the various auxiliary components attached to it. Engine radiated noise has also been reduced significantly over recent years, which means that the contribution of auxiliary components has become more important. A component such as the alternator can produce more overall radiated noise than the rest of the powertrain, if badly designed. So that an accessory will not give rise to unacceptable noise inside a vehicle the component suppliers must develop their parts such that radiated noise levels are sufficiently low. This can only be achieved if the suppliers know what levels of radiated noise are acceptable. This paper presents the results of an investigation into the influence of accessory noise on the overall interior noise of a vehicle, and the setting of tonal noise targets for vehicle accessories, concentrating on an automotive alternator. The investigations are based on detailed transfer function measurements carried out using a wideband sound source and the principle of acoustic reciprocity. A subjective target setting experiment is then described which allowed rig based target values to be derived