An experimental investigation of techniques for reducing flow-induced cavity resonance

Abstract

Excitation of cavity resonance by flow over an aperture is often a source of unwanted noise in aerospace, automotive, and marine applications. An experimental investigation of three resonance reduction techniques was conducted. These were: a fence at the upstream edge, fluid injection, and a new technique developed by the author, in which fluid from the boundary layer is diverted into the cavity. Spectra of the pressure in the cavity were obtained for various flow speeds. Results show that the fence and the boundary-layer diversion technique have the effect of reducing the Strouhal number of the flow-excited sheartones, delaying resonance to higher speeds. Reduction at the resonant speed is also observed. The boundary layer diversion technique was much more effective than the fence. The Strouhal number reduction effect is not observed with fluid injection. The effect on resonance reduction by fluid injection as a function of several parameters, including the rate of fluid injection, is discussed.

Effect of opening obstructions on the flow-excited response of a Helmholtz resonator

“…The convection velocity is normalized by the free-stream velocity, and x is normalized by the overall length of the opening. The unobstructed condition is essentially a baseline, and is similar to the results reported in Zoccola (2000), who showed that for flow over an unaltered opening, the vortex decelerates as it approaches the downstream edge. For the 17 m/s condition, the convection velocity in the upstream half of the opening shows essentially the same behavior as the baseline condition.…”

“…The results show a region of positive energy production (source) near the downstream edge and a region of negative energy production (sink) in the upstream half of the opening. Reasonably good agreement with this result was found by Zoccola (2000), who reported measurements of the energy production term distribution for conventional cavity opening which showed regions of positive and negative energy production in the opening, with the positive being of greater magnitude and positioned above and somewhat downstream from the negative region. Essentially the same distribution was observed for measurements on configuration S0a at 14 m/s.…”

“…But it remains the case that here, as in the upstream half and the baseline condition, and indeed, for all conditions reported in Zoccola (2000), the vortex decelerates as it approaches the downstream edge, or, in the case of the 17 m/s condition, the obstruction. However, for the 30 m/s condition, the convection velocity, like the energy distribution, the behavior of U c over the 269 mm length scale is not the same as the scaled baseline condition.…”

“…Fig. 19(a) shows for the unobstructed case the regions of positive and negative energy productions seen in the data from Zoccola (2000) and predicted by Nelson et al (1983). The coarse and fine grids appear to have the effect of ''smearing'' these regions in the streamwise direction.…”