Effect of blade flutter and electrical loading on small wind turbine noise

Abstract

The effect of blade flutter and electrical loading on the sound level of two different size wind turbines was investigated at the Conservation and Production Research Laboratory near Bushland, TX. Data were collected on two blade designs tested on a wind turbine rated at 1 kW, and there also was a third blade tested on an improved version of the same wind turbine. The 1 kW wind turbines were used for off-grid water pumping. If the 1 kW wind turbines were electrically loaded, the start of blade flutter was delayed from 700 to 900 rpm when the blade length was reduced 9.2%, and no blade flutter occurred when the shorter fiberglass blades were stiffened. However, if there was no electrical load (e.g. offline), the shorter stiffer blades would start to flutter at 1050 rpm. For the 10 kW wind turbine, two blade designs were tested for off-grid water pumping, and a third blade design was tested on a utility grid-tie system. The original blades fluttered when the 10 kW off-grid wind turbine was offline at a rotor speed of 260 rpm, but the second blade design (different airfoil) appeared to never flutter at any rotor speed when offline. For the 10 kW wind turbine, the sound emission of the second blade design was approximately the same as that of the third blade design – both had lower sound emissions than the original blade design. Blade flutter increased the average sound power level on the 1 kW wind turbine 15–20 dB and on the 10 kW wind turbine 5–7 dB. A procedure was developed which allowed the best blade design for sound emission to be identified even though the microphone measurement position varied (due to different tower heights) and electrical loading was different (on-grid, off-grid, or offline). This paper should be of help to design engineers interested in designing wind turbines with a lower sound emission, and test engineers collecting and analyzing sound level data of existing wind turbines. This paper may also help in the validation of computer codes that are used to predict whether blade flutter will occur on large MW size wind turbines.