Christophe Schram

Identification of the aeroacoustic response of a low Mach number flow through a T-joint

A methodology to study numerically the aeroacoustic response of low Mach number confined flows to acoustic excitations is presented. The approach combines incompressible flow computations, vortex sound theory, and system identification techniques, and is applied here to study the behavior of a two-dimensional laminar flow through a T-joint. Comparison with experimental results available in literature shows that the computed source models capture the main physical mechanisms of the sound production in the shear layer of the T-joint.

Reduction of Airfoil Turbulence-Impingement Noise by Means of Leading-Edge Serrations and/or Porous Material

The paper is about the sound produced as turbulence impinges on the leading edge of an airfoil and its reduction by means of either leading-edge serrations (tubercles) or the use of porous materials. The first part describes a series of experiments performed on a NACA-12 airfoil in a low-speed open-jet anechoic wind tunnel. The airfoil is held between end-plates and the sound is measured in the far field in the mid-span plane The chord-based Reynolds number ranges from 1.3 105 to 2 105. Various versions of the airfoil are tested and compared to the baseline. Sound reduction is achieved by both serrations and porosity in a wide frequency range. The second part is devoted to dedicated prediction techniques. A new analytical model of the response of a serrated leading-edge is proposed, extending Amiets theory, in the limit of arbitrary large chord. Preliminary numerical modeling is also discussed for the response of a porous aifoil to incident disturbances, based on a panel method combined with a locally-reacting impedance model.

Application of vortex sound theory to vortex-pairing noise:sensitivity to errors in flow data

Aeroacoustical analogies allow one to extract acoustical information from limited information about the flow. In the particular case of low Mach number compact flows, vortex sound theory has been quite successful. In the present paper, different formulations of the vortex sound theory are compared on the basis of their ability to provide realistic results for vortex-pairing sound when approximate flow models are used. In particular, these theories do not perform equally well when applied to a flow model in which the effective conservation of momentum and kinetic energy is not respected, as it should be in the absence of external forces and neglecting viscous dissipation and compressibility effects. A conservative form of the vortex sound theory is obtained by reiterating the assumptions of conservation of these flow invariants. This alternative form of the analogy allows one to obtain more robust results when applied to perturbed analytical flow models and experimental data.

Correction techniques for the truncation of the source field in acoustic analogies

The truncation of the source field may induce large overpredictions in the acoustic field computed through acoustic analogies. A comparative study of different correction approaches proposed in the literature is carried out, considering three different techniques: correction terms based on a convection assumption, use of model extensions, and windowing techniques. It is shown that convection-based correction terms need to take into account noncompactness effects of the source field in order to yield accurate results. A modified correction term that includes these effects is derived, and its equivalence to the method of model extensions in the case of purely convected flows is highlighted. Moreover, the performance of different windowing techniques is investigated.

Broadband Scattering of the Turbulence-Interaction Noise of a Stationary Airfoil: Experimental Validation of a Semi-Analytical Model

A novel semi-analytical model based on Amiets theory is proposed to predict the scattered acoustic field related to turbulence-interaction noise. The simulated scattered acoustic field is compared to the corresponding measurements of a stationary airfoil in a turbulent stream for validation purposes. The paper is composed of three main parts, where improvements and extensions of this semi-analytical model are presented. In the first part of the paper, a new formulation taking an intermediate level of geometrical near-field correction into account is derived. The second part provides an implementation of a strip method accounting for spanwise varying incoming flow conditions. The acoustic free-field response of the airfoil is computed using the geometrical near-field formulation with the strip method. In the third part, an innovative Boundary Element Method (BEM) approach is proposed in order to compute the scattered acoustic field of the airfoil by an obstacle. Finally, the scattered acoustic field resulting from the presence of a flat screen is computed by the semi-analytical method combined with the new BEM approach. A good agreement is obtained comparing with the measurements performed in an anechoic room.

Sound Produced by Vortex Pairing: Prediction Based on Particle Image Velocimetry

In some cases, the prediction of the sound generated by flows is impaired by inaccuracies in the available flow data. These inaccuracies can be due to the degree of approximation of a flow model, or to the uncertainty of a measurement technique. A particular crucial point for sound prediction is the conservation of the flow invariants. The sound radiated by vortex pairing in a subsonic excited jet is deduced from experimental data obtained by particle image velocimetry. The performance is compared of different formulations of vortex sound theory to predict the corresponding sound production. It is shown that assuming several times the conservation of the momentum and kinetic energy in the implementation of the vortex sound theory improves considerably the robustness of the prediction. In fact, a prediction in good agreement with both a theoretical model of leapfrogging and numerical simulations of merging is obtained, although the basic data do not respect precisely these conservation laws.

Prediction of Incoming Turbulent Noise Using a Combined Numerical / Semi-Empirical Method and Experimental Validation

The present paper investigates the case of a NACA0012 airfoil placed in a turbulent jet. The nozzle outlet diameter is equal to D = 0.041 m. The airfoil is placed at zero angle of incidence and with its leading edge located at 6D from the jet outlet. The chord of the airfoil is equal to D, and has a constant section over its span of 10D. The outlet velocity magnitude U0 is fixed to 13.2 m/s resulting in a Reynolds number based on the chord length of 36,000 and a Mach number of 0.04. The unsteady, three-dimensional incompressible flow around the airfoil is first computed with the LES module of the commercial solver FLUENT Rev. 6.2. The numerical flow results are compared with statistics on the velocity field (mean, RMS and spectra) obtained experimentally with hot wire anemometry on the same geometry and for the same operating conditions. This comparison reveals, as expected, that the mesh refinement influences the cut-off frequency resolution. Besides, an innovative procedure is proposed to compute from the same CFD computation, the noise radiated for the all frequency spectrum. The SYSNOISE Rev.5.6 solver, integrating Curle’s analogy, is used to predict the low frequency part of the noise spectrum, while Amiet’s theory is used to predict the high frequency range. The first one, limited to the computation of sound radiated by compact sources and then to the low frequency range, uses the unsteady pressure fluctuations on the airfoil stored during the CFD flow computation. The second one is a theory specially developed for airfoil sound radiation at high frequency and taking then into account, in an explicit way, non-compactness effects. The statistical flow data, needed by Amiet’s model, are fitted on the CFD data.

Calculation of sound scattering using curle's analogy for non-compact bodies

Aeroacoustics is a growing fleld in aeronautical and automotive industry, mainly through the application of the aeroacoustical analogy. Curle’s analogy is particularly popular in automotive industry. This paper presents some preliminary results of an aeroacoustical study of the sound produced by the leapfrogging of two rectilinear fllament vortices within an inflnite two-dimensional straight duct. The ∞ow fleld is obtained from an inviscid kinematic description of the vortical motion. Such incompressible ∞ow description does not contain any acoustical information over the source region. The sound is predicted in frequency domain following two approaches: using a tailored Green’s function based on the non-evanescent duct modes, and applying Curle’s analogy to the incompressible ∞ow model. The results show that the pressure fleld obtained applying Curle’s analogy difier signiflcantly from the pressure fleld obtained using the tailored Green’s function. This is particularly severe at frequencies for which non-planar duct acoustical modes are excited by the vortex motion. The discrepancies are attributed to the lacking acoustical information in the source description. A hybrid method combining Curle’s analogy with a boundary element method to determine a numerical tailored Green’s function, is likely to yield a better sound prediction.

A Primitive Approach to Aeroacoustics

A primitive approach to aeroacoustics is presented using the aeroacoustic analogies. This aproach is illustrated by a few examples like musical instruments, the Rijke tube, speech production etc.

Free and scattered acoustic field predictions of the broadband noise generated by a low-speed axial fan

Broadband noise generated by a low-speed industrial axial fan and its scattered field by a benchmark obstacle have been addressed. Amiets theory on turbulence-interaction noise has been extended in order to predict the acoustic response of a fan in its geometrical near-field. A segmentation technique has been applied for spanwise varying flow conditions. The improved model has been combined with boundary element method (BEM) for acoustic scattering. The validation of the broadband scattering technique has been performed through comparisons with an analytical model considering acoustic scattering from an infinite plate and with measurements of a low-speed axial fan operating nearby a flat scattering screen.