Airfoil Trailing Edge Noise Measurements in an Open-jet, Hard-wall and Kevlar-wall Test Section: a Benchmark Exercise

We experimentally investigate the comparability of airfoil trailing-edge noise measurements in an open-jet, hard-wall and hybrid test section wind tunnel in a set of benchmark exercises. We used three different airfoils, namely a NACA-0012, NACA-0018 and NACA-63018 with 0.2 m chord and 0.7 m span. We focus on the results for zero angle of attack and a free-stream velocity of 25 and 30 m/s giving a chord-based Reynolds number of 330,000 and 400,000, respectively. All measurements were conducted in the aeroacoustic wind tunnel facility of the University of Twente. This allows us to vary the test section type while keeping other parameters constant. We measured the wall pressure fluctuations near the trailing-edge on all airfoils which shows identical source terms of the trailing-edge noise mechanism when performing measurements using the same conditions in all test section types. Both the wall pressure spectrum and spanwise coherence remain constant regardless of test section type. Far-field noise is measured by a microphone phased array, and spectra are determined by using the source power integration technique of conventional beamforming maps. In the closed test section, wall-mounted microphones are recessed behind a stretched Kevlar cloth to reduce boundary layer self-noise and to improve the signal-to-noise ratio. A far-field microphone phased array is used in the open-jet and hybrid test section. Absolute noise levels of the baseline airfoils are compared and show good overall agreement of the noise levels and spectral shape within approximately ± 2 dB. We found an exception at higher frequencies in the closed test section where the noise level was considerably lower than the noise measured in the open-jet and hybrid test sections. We also compared the relative noise levels by evaluating the noise reduction caused by add-on trailing-edge serrations on each airfoil. The noise reduction measured in each test section was identical, within approximately ± 2 dB in the entire frequency range. The measured differences are typical for the amount of uncertainty that can be expected in microphone phased array measurements. The results presented provide a valuable methodology and dataset for benchmark exercises of hybrid wind tunnels and airfoil trailing-edge noise measurement and prediction.