Search Results

A method of mapping a sound field using sound quality metrics has been investigated with an aim of identifying noise sources based on their sound characters rather than traditional measures such as sound pressure level (SPL) and intensity. The method having 11 metrics was implemented in four different array applications, namely near-field acoustical holography, planar beamforming, spherical beamforming, and patch acoustical holography. The sound quality metrics (SQ) mapping was applied to diesel engine measurements as well as vehicle interior measurements. In both applications, there have been a number of attempts to identify impulsive noise sources and therefore in this investigation an impulsiveness metric was developed. The proposed metric was validated against ideal impulses as a function of impulse repetition frequency, amplitude and duration of impulses.

Beamforming based on measurements with a spherical microphone array is a recognized technique for localization of noise sources. The method is very efficient for getting a quick overview of the positions of the most dominant sources. However, when more detailed information about the radiation from a source is needed, acoustical holography is normally more suitable, because it provides the actual acoustical quantities (pressure, velocity and/or intensity) at positions on or near the source surface, whereas beamforming only gives a directional map of contribution at the array position. Even though spherical arrays are most often used for beamforming they can also be applied for holography purposes. This means that with the same hardware beamforming can be used to identify a noise source, and then the array can be moved closer to the source to get a clearer picture of the radiation by using holography.

For many years engineers in the automotive market have struggled to find ways to accurately and efficiently map the noise sources found inside a vehicle. Many techniques, both theoretical and measurement based, have been proposed and used, but there has always been a trade off between accuracy and efficiency. Techniques like sound intensity mapping and Statistical Energy Analysis have proven to be accurate when mapping noise sources in vehicle, but require a large investment in time and money to create a simple, easy to interpret picture showing where dominant noise sources come from. In this paper the authors will introduce and demonstrate a novel technique, spherical beamforming, which can overcome the issue of test time and produce fast, accurate noise maps from the interior of a vehicle.