Search Results

Grazing flows over open windows or sunroofs may result in “flow buffeting,” i.e. self-sustained flow oscillations at the Helmholtz acoustic resonance frequency of the vehicle. The associated pressure fluctuations may cause passenger fatigue and discomfort. Many solutions have been proposed to solve this problem, including for example leading edge spoilers, trailing edge deflectors, and leading edge flow diffusers. Most of these control devices are “passive” i.e. they do not involve dynamic control systems. Active control methods, which do require dynamic controls, have been implemented with success for different cases of flow instabilities. Previous investigations of the control of flow-excited cavity resonance have used mainly one or more loudspeakers located within the cavity wall. In this study, oscillated spoilers hinged near the leading edge of the cavity orifice were used. Experiments were performed using a cavity installed within the test section wall of a wind tunnel.

The present work was directed towards the design of a sideline microphone array specifically adapted to the visualization of automotive noise sources in the 500 Hz to 2000 Hz range. The particular design philosophy followed here involved the minimization of the array redundancy: i.e., the minimization of the number of pairs of microphones that are separated by the same distance in the same directions. The performance of sixty-four element microphone arrays designed according to this principle will be illustrated through the use of simulated motor vehicle passbys. In addition, their performance will be compared with more conventional array designs: e.g., elliptical, and spiral arrays.

Spherical beamforming was used to visualize sound radiation during a vehicle passby test. Forward and backward propagation procedures are compared in terms of computational expense. A spherical spreading correction factor is described, along with a maximum liklihood procedure for obtaining an optimal array weighting dependent on the relative distance between the microphones and the focus point. The de-Dopplerized microphone outputs are multiplied by the weighting factors and summed to yield the source strengths over a reconstruction plane “attached” to the vehicle. Results obtained using a 16 element sparse array during an actual passby are used to demonstrate the present approach.

An analytical model based on “vortex sound” theory was investigated for predicting the frequency, the relative magnitude, the onset, and the offset of self-sustained interior pressure fluctuations inside a vehicle with an open sunroof. The “buffeting” phenomenon was found to be caused by the flow-excited resonance of the cavity. The model was applied to investigate the optimal sunroof length and width for a mid-size sedan. The input parameters are the cavity volume, the orifice dimensions, the flow velocity, and one coefficient characterizing vortex diffusion. The analytical predictions were compared with experimental results obtained for a system which geometry approximated the one-fifth scale model of a typical vehicle passenger compartment with a rectangular, open sunroof. Predicted and observed frequencies and relative interior pressure levels were in good agreement around the “critical” velocity, at which the cavity response is near resonance.