Sunroof buffeting noise is annoying to drivers and passengers. The conventional method for suppressing sunroof buffeting noise is to use passive deflectors. A recent trend has been large sunroof openings, such as panoramic sunroofs, in accordance with customer preferences for a feeling of openness. Since sunroof buffeting noise tends to become louder as the sunroof opening area becomes larger, a conventional passive deflector may not be a solution in this case, and a new effective method for reducing the sunroof buffeting noise is required. Previous work showed that a strong, self-sustained tonal noise, generated from a Helmholtz resonator exposed to a grazing flow, could be significantly reduced by closedloop control of an active deflector installed near the upstream edge of the resonator opening. The active deflector system is a cascade of a microphone sensor mounted inside the cavity, controller, power amplifier, and deflector mechanism vibrated by a voice coil actuator. Since the acoustic pressure inside the cavity is influenced by the shear layer modified by the active deflector, the active deflector and acoustic response of the cavity form a closed-loop control system. The main objective of the present study is to implement this technology on a real vehicle and evaluate whether the technology can be utilized to suppress sunroof buffeting noise. A simple active deflector system was assembled and installed in a compact-sized hatchback car with a sunroof opening length of 460 mm. The active deflector system was tested both in a wind tunnel and on a proving ground. The test results showed that the active deflector reduced the sunroof buffeting noise by as much as 25 dB. The active deflector was shown to be stable and robust regardless of changes in the wind speed and wind yaw angle.

No abstract

The large-scale shear flows over the sunroof opening of a mid-sized SUV measured using a PIV system were investigated. The shear flows were measured for five different cases of deflector protrusion (one case was the baseline test without deflector) at two different free stream flow velocities below the critical velocity where the buffeting noise level reached a maximum. The structures of the shear flows were observed to differ, apparently depending on whether the radiated buffeting noise is relatively strong or not. For strongly buffeting experimental cases, the momentum thicknesses of the shear layers were observed to grow rapidly and saturated at a station near the downstream edge of the sunroof opening, where the saturation of the transverse velocity fluctuations was also observed, and where the vortex coalescence process was presumably completed. On the other hand, no discrete large-scale vortex structures were observed for none-buffeting or weakly buffeting cases. Streamwise growth of the velocity fluctuations was found to be well predicted by a linear hydrodynamic instability analysis for the strongly buffeting cases. Numerical results obtained from a linear inviscid instability analysis using a hyperbolic tangent mean velocity profile were used to calculate the amplification factors with the initial momentum thickness and the streamwise fluctuation wavenumber. The shear flows were found to form large-scale discrete vortices when the linear inviscid amplification factors exceeded a threshold amplification factor.

No abstract