Wall Pressure Spectra on a DU96-W-180 Profile From Low to Pre-Stall Angles of Attack

Abstract

In an effort to characterize noise induced by separated turbulent boundary layers, surface pressure fluctuations on a DU96-W-180 airfoil were measured using miniature pressure sensors. Because of limitation in amplifier channels and available sensors, a rearrangeable configuration of sensors was applied. Chordwise distributions of the surface pressure were obtained for aerodynamic angles of attack of −0.8 • ≤ α ≤ 10.3 • and at three Reynolds numbers (0.8, 1.0, and 1.2) ×10 6 . The boundary layer profile at 1%c behind the trailing edge was measured using constant temperature anemometry. The boundary layer thickness compares well with that simulated using XFOIL for α ≤ 7.8 • . Within the limits of the simulation, other relevant boundary layer properties from XFOIL were used to calculate the surface pressure spectrum predicted from published empirical models for zero and non-zero pressure gradient turbulent boundary layers. Finally, a modified Blake-TNO semiempirical model was used to predict the surface pressure spectrum near the trailing edge for separated flow. The modification is introduced to the so called 'moving axis spectrum' and the chord-normal correlation length scale. It is found that in the low frequency range, the modified semi-empirical model fits well with the measured surface pressure spectrum of a separated turbulent boundary layer. Nomenclaturea Speed of sound, m·s −1 c Chord length, m c l Coefficient of lift, -c 1 , c 2 Modification parameters to Φ m , -k e Eddy containing wavenumber, 0.7468/L y , m −1 k x Chordwise wavenumber, 2π/L x , rad·m −1 k z Spanwise wavenumber, 2π/L z , rad·m −1 L Wetted span, m mix Mixing length, 0.085δ tanh(κy/0.085δ), m L x Chordwise correlation length scale, m L y Chord-normal correlation length scale, m L z Spanwise correlation length scale, m R Radial distance from a sound source, m Re Reynolds number, -R T Ratio of the outer and inner boundary layer time scales, u τ δ/ν C f /2 U (y) Time averaged velocity component, m·s −1 U ∞ Freestream velocity, m·s −1 U c (y) Convective velocity, m·s −1 U e Turbulent boundary layer edge velocity, m·s −1 u, v, w Instantaneous velocity components in the order x, y, z, m·s −1

A semi-empirical airfoil stall noise model based on surface pressure measurements

“…It may therefore indicate that the present model is not valid in early stall conditions. Note that Suryadi and Herr [14] also interpret this part of the spectrum as a contribution from the suction side (see discussion at the end of Section 8.2).…”

“…The measured SP spectrum at x/C = 98% and far-field noise spectrum at 2 m above the mid-span of the airfoil are displayed in Fig. 27 together with In the first case, the airfoil was equipped with SP microphones [14], and in the second case far-field noise was measured [13].…”

“…It is more specifically referred to as airfoil stall noise, but also sometimes as separation noise when the flow remains attached on a significant part of the airfoil chord (as opposed to 'deep-stall' when the flow separation stretches over the entire chord). Airfoil stall noise has been investigated experimentally and acoustic measurements are usually performed in dedicated anechoic 45 wind tunnels [8,9,10,11,12,13,14]. A few models have been devised, the most popular one probably being the so-called BPM model [8].…”

“…They derived a model based on classical trailing edge noise theory [15,16] and tuned for the stalled conditions. Subsequently and in order to study lower frequencies that could not be measured acoustically, this experiment was conducted again by Suryadi and Herr [14] some cases, the airfoils were also tested with zig-zag (ZZ) tapes in order to trigger transition. The tapes thickness is equal to 0.4 mm and their width between extremities 5 in the flow direction is equal to 1 cm.…”

Mechanisms of airfoil noise near stall conditions

“…AerofoilAirfoil noise at near-stall conditions has been measured in a number of studies ([1, 5,[8][9][10]). These measurements have been used to develop models of the radiated noise.…”