Beamforming with modified steering vectors for jet noise source location

Abstract

Prof. Krish Ahuja has a longstanding interest in jet noise source location. His work in this area is grounded in the idea that if the assumed source location is correct, then the sound should obey the inverse square law relative to that point and the phase should be constant along lines originating at that point. He applied this with, conceptually, one microphone in 1985 and two microphones in 1998. In 2006 he commissioned a beamforming system, Array 48, from OptiNav, Inc. His student, Nick Breen, used this to measure subsonic jet noise source location in detail. The NASA-Glenn Research Center also purchased an Array 48. In the current work, a jet noise data set measured by Gary Podboy using Glenn’s array in 2008 is revisited with a new beamforming algorithm, Robust Functional Beamforming, to further support Tam’s two-source model and Breen’s source location. Beamforming with modified steering vectors is performed to measure the parameters of the wavepacket source model from the far field. This process suggested replacing the wavepacket spatial length parameter with a temporal lifetime parameter. Another steering vector modification aimed to measure modes with odd spinning order. It seems to have found them at an apparent location 10 jet diameters removed from the jet, laterally. This is tentatively interpreted as a Mach radius phenomenon like one observed by Csaba Horvath at NASA-Glenn, also using Array 48, to study counter-rotating propeller noise. In an observation unrelated to beamforming, the excess noise measured at 40° from the jet axis as compared with the 90° angle, is fully contained in the first few cross spectral matrix eigenvalues, or Spectral Proper Orthogonal Decomposition modes, in some cases.