Session LJ: Acoustics II

Acoustic-structure interaction of a plate in a duct with flow

The interaction of sound in the presence of flow with a cavity-backed, clamped elastic plate in a duct is analyzed. The problem consists of an incoming plane wave at fixed frequency in a square duct with an elastic plate mounted flush on one wall of the duct. The fluid is inviscid and compressible. A loosely-coupled numerical simulation is carried out using a high-fidelity explicit finite difference fluid solver and a mixed-enhanced implicit finite element structural solver. Also a fully coupled, single frequency, theoretical model for the same system has been derived and is compared to the simulation results for a low-amplitude linear case. The conditions chosen correspond approximately to a 3mm thick, 1ft x 1ft Aluminum plate subjected to a 100 dB pressure load at the in vacuo fundamental frequency of the plate. The temporal displacement of the plate and duct pressure are calculated using the two methods. Comparison of the results suggests that the assumption of single frequency response made in the analytical solution is valid. Significant frequency-dependent changes in the plate response are observed for some of the peaks when a non-zero mean flow is present.   

Direct Numerical Simulation of Three Dimensional Honeycomb Liner with Circular Apertures

Acoustic liners are effective methods to decrease aircraft engine noise. Early designs and understanding of them were mostly made through theoretical analysis and experiments. More recently, numerical simulations of the detailed fluid mechanics of liners have been used to provide A better understanding of the liner working mechanisms but are typically limited to 2-D geometries. In this work, a 3-D model of a NASA Langley honeycomb liner with circular holes is studied by direct numerical simulation using a high-fidelity compressible Navier-Stokes code with overlapping mesh capability. The simulations are first validated by predicting the liner impedance as a function of frequency (1--3kHz) for normally incident sound waves at 130 dB amplitude and comparing to Langley experimental data. The flow through the circular opening, or aperture, was examined and found to be primarily laminar with the vorticity attached to the aperture walls. The effect the sound amplitude was then investigated, up to 160 dB. Around an amplitude of 140 dB the attached aperture wall vorticity begins to separate with an oscillatory, axisymmetric, laminar jet, dominated by vortex rings, appearing at 150 dB. At 160 dB the oscillatory jet becomes turbulent. Connections between the liner impedance and the jet structure are made, as are detailed phase-averaged statistics. The jet production process was also examined in detail.   

Sound generated by a vortex convected past an elastic sheet

We study the motion and sound generated when a line vortex is convected in a uniform low-Mach stream parallel to a thin elastic sheet. The linearized sheet motion is analyzed under conditions where the unforced sheet (in absence of the line vortex) is stationary. It is found that the vortex-sheet interaction excites a resonance response in the sheet, where the sheet oscillates at its least stable eigenfrequency. The sources of sound in the acoustic problem include the sheet velocity and fluid vorticity. It is shown that the release of trailing-edge vortices, resulting from the satisfaction of the Kutta condition, has two opposite effects on sound radiation: while trailing-edge vortices act to reduce the pressure fluctuations occurring owing to the direct interaction of the line vortex with the unperturbed sheet, they amplify and extend the acoustic signal produced by the motion of the sheet. The sheet motion becomes an increasingly important source of sound as the system approaches its critical conditions for instability, where the effect of resonance becomes more pronounced.   

Implementation of Active Noise Control in a Closed-Circuit Wind Tunnel

Closed return wind tunnels, such as the Klebanoff--Saric Wind Tunnel (KSWT) at Texas A\&M University, can provide relatively low freestream turbulence levels but include noise sources that do not exist in flight. This background noise, such as fan and motor noise, can adversely affect boundary-layer transition experiments if the frequencies are in the range of unstable Tollmien-- Schlicting waves. Passive acoustic treatments eliminate most noise propagating downstream from the fan to test section in the KSWT, but measurements showed upstream-traveling tonal noise propagating from the fan into the test section. To eliminate this, an active noise control system utilizing an adaptive filter algorithm was implemented targeting frequencies in the TS band below the planar duct mode cut off. Multiple microphones are used to detect and cancel upstream traveling sound without affecting downstream traveling sound. Microphone measurements are used to document the noise reduction at multiple locations in the test section.   

Aero-acoustic performance of Fractal Spoilers

One of the major environmental problems facing the aviation industry is that of aircraft noise. The work presented in this paper, done as part of the OPENAIR Project, looks at reducing spoiler noise through means of large-scale fractal porosity. It is hypothesised that the highly turbulent flow generated by these grids, which have multi-length-scales, would remove the re-circulation region and with it, the low frequency noise it generates. In its place, a higher frequency noise is introduced which is susceptible to atmospheric attenuation, and would be deemed less offensive to the human ear. A total of nine laboratory scaled spoilers were looked at, seven of which had a fractal design, one conventionally porous and one solid for reference. All of the spoilers were mounted on a flat plate and inclined at $30^{\circ}$ to the horizontal. Far-field, microphone array and PIV measurements were taken in an anechoic chamber to determine the acoustic performance and to study the flow coming through the spoilers. A significant reduction in sound pressure level is recorded and is found to be very sensitive to small changes in fractal grid parameters. Wake and drag force measurements indicated that the spoilers increase the drag whilst having minimal effect on the lift.   

Investigation and Control of Flow-Induced Helmholtz Resonance

Grazing flow over the orifice of a Helmholtz resonator can result in a self excited resonance inside the cavity. A practical example is automotive sunroof buffeting. The resonance is generated by the vortical-acoustic coupling between the instability of the shear layer cross the orifice and the Helmholtz mode of the resonator. A simplified model was developed based on a control volume momentum analysis, which allows accurate predictions of both the frequency and amplitude of the cavity pressure fluctuations. The present study has specially focused on further understanding of the interaction between the forcing and the acoustic characteristics of the Helmholtz resonator to improve the model, as well as modification of the orifice geometry and exterior grazing flow to minimize the cavity pressure fluctuation.   

Diffraction of an acoustic wave by a cavitating hydrofoil

Diffraction of a plane acoustic wave from a curvilinear foil placed in an inviscid fluid is considered. The flow is irrotational and steady-state. The upper and lower boundaries of the foil, $h_+(x)$ and $h_-(x)$ satisfy the condition $h'_\pm(x)<<1$. Because of the hydrofoil profile, it may be partly cavitating or fully cavitating. The problem is linearized, and the boundaries of the foil are replaced by their projections on the real axis. It is found that at the fore point of a foil whose upper and lower boundaries are wet or at the rear point of the supercavity, the complex velocity $\overline{w'(z)}$ has a $-1/2$-singularity. At the fore point of a foil whose one side is wet and another one is cavitated, the function $\overline{w'(z)}$ has a $-1/4$-singularity. At the aft points of a wet foil, the function $\overline{w'(z)}$ is bounded. The problem is solved in a closed form in terms of singular integrals. The unknown boundary of the cavity is recovered. Since the boundary of the cavity is known, and the fluid mechanics problem is linearized, the diffraction problem reduces to a system of singular integral equations which are solved by the method of orthogonal polynomials. Generalizations to the case of the nonlinear Tulin single-spiral-vortex model and a system of hydrofoils are discussed.   

Modeling of the Flow Field from Turbofan Nozzles with Porous Fan Flow Deflectors

Wedge-shaped fan flow deflectors have shown promise in reducing noise from turbofan-type nozzles. Porous deflectors have a particular advantage as they allow some flow through the wedge, thus preventing strong velocity gradients that can cause excess noise near the wedge base. Computational modeling of the resulting flow field is challenging because it is not feasible to grid the perforations of the deflector. Instead, we use a body-force term in the momentum equation that is applied locally in the vicinity of the deflector porous surface. The body-force term is calibrated based on experimental velocity measurements inside and outside the flaps forming the wedge, as well as computations on simplified two-dimensional models of the wedge. For a given wedge angle, the flow field is governed by the porosity of the wedge surface as well as the ``illumination angle'' of the perforation holes. The latter parameter is crucial for obtaining a uniform velocity distribution at the wedge base, and can be adjusted by varying the surface thickness or the aspect ratio of the perforation holes.   

Application of Conformal Transformations to Velocity Sources in Cavity Aeroacoustics

The surface pressure fluctuations observed in open cavity flows are related to the velocity sources present in the shear layer spanning the cavity opening. The relationship between these velocity sources and the pressure fluctuations can be expressed by Poisson's equation giving a functional description of $\nabla^2 p$. This relationship can be cast in an incompressible form, $\frac{1}{\rho}\nabla^2 p = f(\vec{x},\vec{u})$, or in a compressible form, $\nabla^2 p = f(\vec{x},\vec{u},\rho)$. In either case, the source terms can be integrated to yield the resulting pressure at a point. In order to accomplish this integration for cavity flows, a conformal transformation is needed to project the cavity flow domain into a flat plate. Here, the required conformal transformation is defined and its effect on the integration domain is examined for incompressible, spanwise homogeneous, cavity flow. The results provide a look at the regions of the shear layer which directly affect the pressure at various locations along the streamwise extent of the cavity. These regions are then compared to dominant POD modes of the shear layer velocity.   

Modification of Instability Waves and Radiated Sound due to Heating a Compressible Mixing Layer

It is well known that heating a turbulent jet at constant velocity alters its sound field and, to a lesser extent, the turbulent statistics. For low speed jets ($U_j/a_\infty < 0.7$) heating increases the radiated sound while at higher speeds heating decreases the radiated sound. In both cases the turbulence root-mean-square levels increase by 10\% relative to the unheated jet. The cause of the sound field change is not known. This work examines the early jet development by considering the modification of instability waves on a compressible mixing layer due to heating using calculations of the linearized Euler equations coupled to a multiple scale expansion analysis of the governing problem. The near-field instability wave solution is matched asymptotically to a globally valid acoustic field for a uniformly valid solution. It is found that for high-speed mixing layers the growth-and-decay cycle of the instability waves is altered by heating, leading to increased entropy fluctuations which are less efficient sound radiators relative to vortical fluctuations.