Comparison of Noise Sources in High and Low Reynolds Number High Speed Jets

Abstract

Results from low and high Reynolds number jets are compared to investigate how the Reynolds number (Re) of a jet influences the mechanisms generating jet noise. Direct numerical simulations (DNS) and experimental results of the authors were used for this purpose. The DNS results are for a Mach 0.9, low Re (ReD~3600) axisymmetric jet (Freund, 2001) and the experimental results are for an ideally expanded, Mach 1.3, high Re (ReD~1.06 x 10 6 ) axisymmetric jet (Hileman et al. 2004b). These two cases will be referred to as LReJ and HReJ, respectively, hereafter. Previous experimental work on the HReJ using a three- dimensional microphone array, located at 30° with respect to the jet axis, with a novel beamforming technique estimated the source location and the time of generation of each large amplitude sound wave that reached the far-field. Simultaneous with the pressure measurements, the flowfield, the source region, was visualized using a MHz rate imaging system, and was analyzed using reconstructions via Proper Orthogonal Decomposition (POD). These results were compared to periods involving no large amplitude far-field noise events and revealed that the growth and decay of large structures in the mixing layer (a wave-like series) are the dominant flow feature during the emission of sound waves that travel in at the aft quadrant (Hileman et al., 2004b). In the current work, this technique is used with the LReJ and the results are compared. There are many similarities between the two including the far-field acoustic spectrum, coherence, average waveform and mean noise source location. The few differences observed are believed to be due to the limited extent of turbulence scales and/or because the initial shear layer was laminar in LReJ, both related to the Re. The main conclusion is that the rapid breakdown of the large-scale structure appears to be an important, and perhaps the main mechanism of jet noise. Right before the breakdown, the structures seem to contract in size, tilt and eventually disintegrate. To offer a possible explanation to the observed noise mechanism, a simple one-dimensional instability wave model is shown to create more noise in the far-field when the wave is subject to truncation simulating a breakdown.