NOISECAM-USING 3D BEAMFORMING TO BETTER LOCALISE NOISE SOURCES ON HOVERCRAFT

Light Hovercraft are fun but with high revving engines and fans they can be noisy. NoiseCam, a homebrew 2D beamforming system, has been used on over 130 amateur craft worldwide to identify the loudest noise sources, resulting in several craft being reduced by up to 10dba. It localises and quantifies individual tonal sources such as engine exhausts, air intakes and fan turbulence in the range 40hz to 800hz, typically within about 100mm and about 5hz. A number of improvements have been made to the system. It now incorporates a ring of far-field dba meters to provide direct traceability from regulatory dba measurements, through frequency analysis and then beamforming to locate the key sources. The problem of large spot size and positional displacement of low frequency sources has been addressed by a technique for adding spikes to the beamformer waveforms. A 3D microphone array with 24 mics is being developed to reduce astigmatism and better identify fan turbulence sources. Turbulent areas identified by the system have been confirmed by probing with pressure sensors. An air horn audio marker beacon now provides better alignment between the noisepics and conventional pictures. 3 rd Berlin Beamforming Conference 2