Aeroacoustic Wind Tunnel Tests of Wind Turbine Airfoils

Abstract

Aeroacoustic wind tunnel tests were performed for six airfoils that are candidates for use on small wind turbines. One additional airfoil (NACA 0012) was tested for comparison to benchmark data. The acoustic measurements were performed in NLR's Small Anechoic Wind Tunnel, for a range of wind speeds ( U) and angles of attack, with and without boundary layer tripping. Besides the airfoil self-noise measurements in a clean tunnel flow, the models were also tested with a turbulence grid in the nozzle, to investigate airfoil noise associated with inflow turbulence. A 48-microphone out-of-flow acoustic array was used to locate noise sources and to separate airfoil noise from extraneous wind tunnel noise. Special techniques were applied to translate acoustic source plots to absolute airfoil noise spectra. The acoustic results indicate that in a clean tunnel flow trailing edge noise is dominant for all airfoils. In the untripped condition a number of airfoils exhibit intense tones, that disappear after proper tripping is applied. Broadband trailing edge noise levels are found to scale with U 4.5 . The agreement with the benchmark data was generally good. In case of inflow turbulence, leading edge noise is dominant for all airfoils, and no difference is observed between results with and without tripping. The inflow turbulence noise levels are found to scale with U 6 . Comparison of the acoustic results for different airfoils indicates that inflow turbulence noise levels increase with increasing sharpness of the model leading edge. The directivity of both trailing edge and inflow turbulence noise is found to be symmetrical with respect to the chord. With regard to the test set-up, it was found that a treatment of porous material at the model-endplate junctions yields a reduction of broadband extraneous noise up to 10 dB. As a result, the array can look much 'deeper', which enables the detection of very low trailing edge noise levels.