Search

The spectral decomposition algorithm presented in the paper may be applied to selected parts of the SPL spectrum, i.e. to specific noise generating mechanisms. It yields the propagation and the generation functions, and indeed the Mach number scaling exponent associated with each mechanism as a function of the Strouhal number. The input data are SPL spectra obtained from measurements taken during speed ramps. Firstly, the basic theory and the implemented algorithm are described. Then, the behaviour of the new method is analysed with reference to numerically generated spectral data and the results are compared with the ones of an existing method based on the assumption that the scaling exponent is constant. Guidelines for the employment of both methods are provided. Finally, the method is applied to measurements taken on a cooling fan mounted on a test plenum designed following the ISO 10302 standards. The most common noise generating mechanisms are present and attention is focused on the low-frequency part of the spectrum, where the mechanisms are superposed. Generally, both propagation and generation functions are determined with better accuracy than the scaling exponent, whose values are usually consistent with expectations based on coherence and compactness of the acoustic sources. For periodic noise, the computed exponent is less accurate, as the related SPL data set has usually a limited size. The scaling exponent is very sensitive to the details of the experimental data, e.g. to slight inconsistencies or random errors. [ABSTRACT FROM AUTHOR]

This paper presents the acoustic performance of a novel approach for the passive adaptive control of tonal noise radiated from subsonic fans. Tonal noise originates from non-uniform flow that causes circumferentially varying blade forces and gives rise to a considerably larger radiated dipolar sound at the blade passage frequency (BPF) and its harmonics compared to the tonal noise generated by a uniform flow. The approach presented in this paper uses obstructions in the flow to destructively interfere with the primary tonal noise arising from various flow conditions. The acoustic radiation of the obstructions is first demonstrated experimentally. Indirect on-axis acoustic measurements are used to validate the analytical prediction of the circumferential spectrum of the blade unsteady lift and related indicators generated by the trapezoidal and sinusoidal obstructions presented in Ref. [A. Gérard, A. Berry, P. Masson, Y. Gervais, Modelling of tonal noise control from subsonic axial fans using flow control obstructions, Journal of Sound and Vibration (2008), this issue, doi:10.1016/j.jsv.2008.09.027.] and also by cylindrical obstructions used in the literature. The directivity and sound power attenuation are then given in free field for the control of the BPF tone generated by rotor/outlet guide vane (OGV) interaction and the control of an amplified BPF tone generated by the rotor/OGV interaction with an added triangular obstruction between two outlet guide vanes to enhance the primary non-uniform flow. Global control was demonstrated in free field, attenuation up to 8.4dB of the acoustic power at BPF has been measured. Finally, the aerodynamic performances of the automotive fan used in this study are almost not affected by the presence of the control obstruction. [Copyright &y& Elsevier]

This paper investigates the analytical calculation of blade unsteady lift spectrum when interacting with a neighboring obstruction, designed to control tonal noise. The approach used in this paper is to add a secondary unsteady lift mode, of equal intensity but opposite in phase with the primary unsteady lift mode which radiates most of tonal noise, so that the resultant of both primary and secondary modes is null. To control one unsteady lift mode (consequently an acoustic tone) without affecting the harmonics of the controlled mode (consequently the harmonics of the acoustic tone to be controlled), it is important for the secondary unsteady lift to be harmonically selective. We have therefore evaluated the harmonic content of the blade unsteady lift generated by the proposed control obstructions. To this purpose, an original equation is derived using the infinitesimal radial strips theory coupled with the one-dimensional Sears gust analysis. The spectrum of the blade unsteady lift is then analyzed for three types of obstructions: a series of -trapezoidal obstructions, a -periodic sinusoidal obstruction and a series of -rectangular obstructions (where is the number of blades). The use of salient obstructions leads to a large unsteady lift harmonic content. An optimized wake width of the trapezoidal obstruction leads to a low harmonic content rate of . A Gaussian approximation of the measured inflow velocity profile generated by a sinusoidal obstruction leads to a relatively low harmonic content rate of , which indicates that most of the energy is contained in the fundamental mode of the blade unsteady lift. Finally, a rotor/rectangular interaction shows that the use of small-width rectangular obstructions leads to a higher harmonic content rate of . [Copyright &y& Elsevier]

Abstract: If the inlet flow to a fan is non-uniform, as is often the case for aircraft engines, then undesirable tonal noise can be generated. A number of authors have suggested using active cancellation to reduce the noise. The secondary field can either be generated by loudspeakers or by the fan itself if secondary non-uniformities are deliberately introduced into the flow. In the research reported here rods inserted radially into the duct were used to generate the secondary field. The distance by which each rod protrudes into the duct was adaptively adjusted in response to an array of in-duct microphones so as to minimise the radiated sound power, whereas previously only fixed rods have been considered. The ability of the steepest-descent algorithm to minimise in-duct sound power under suitable conditions, and hence reduce radiated sound power is demonstrated in both simulations and low Mach number experiments. It is shown how the ability of such a system to control noise depends on the number and position of the controller rods, and the number of acoustic duct modes to be controlled. Thus at low fan speed, when only one mode was present just two controllers achieved an in-duct noise reduction of 25dB at the blade passing frequency, whereas at a higher fan speed with three modes present six controllers only achieved 2dB. To implement such a scheme in practice, where large numbers of modes are typically present, it would be necessary to develop controller arrays with many actuators, but with low aerodynamic penalty. Such a system might also be useful in HVAC applications, or in wind-tunnel testing. [Copyright &y& Elsevier]

Abstract: Fan noise is one of the principal noise sources in turbofan aero-engines. At supersonic fan speeds, fan tones are generated by the “rotor-alone” pressure field. In general, these tones can be well absorbed by an inlet duct acoustic liner, except at high supersonic fan speeds when the rotor-alone pressure field is well cut-on. In this article an axially segmented liner is proposed, which is predicted to improve the attenuation of tones at high supersonic fan speeds. The analysis is based on locally reacting cavity liners. The axially segmented liner is axisymmetric and consists of two circular sections of different linings joined together. The optimum design consists of two linings with the same face-sheet resistance, but with different cavity depths. The depth of the liner adjacent to the fan is very thin. This means that where the two liners are joined there is a wall impedance discontinuity that can cause acoustic scattering. Fan tones can be modelled in terms of spinning modes in a uniform circular-section duct. The liner is axisymmetric, so modal scattering will be only between different radial modes. The optimum design minimizes the acoustic energy scattered into the first radial mode. This improves the attenuation of fan tones at high supersonic fan speeds, because acoustic energy is scattered into high radial mode orders, which are better absorbed by the lining. [Copyright &y& Elsevier]

Abstract: Traction noise is one of the noise sources of powered railway vehicles such as locomotives, electric- and diesel-powered multiple unit trains and high-speed trains. Especially at speeds below 60km/h and at idling, but also at acceleration conditions for a wide range of speeds, traction noise can be dominant. This is relevant for noise in residential areas near stations and shunting yards, but in some cases also along the line. The other relevant sources are rolling noise, often dominant between 100 and 250km/h, aerodynamic noise, which can be dominant above 300km/h, braking noise, curve squeal and impact noise. The braking system can often technically be considered part of the overall traction system, although acoustically it will often have separate noise sources. In the Harmonoise and IMAGINE EU projects, a generalised prediction model for railway traction noise has been proposed to cover a broad range of powered railway vehicles. The model is one of the prediction modules for overall rail traffic noise, which also covers the other main sources. The traction noise model includes the main operational parameters such as driveshaft speed and power settings, and also takes individual auxiliary components and their duty cycles into account, such as compressors, valves and fans. Source height is included in the model. The level of modelling detail in the many potential traction noise sources has been kept to a minimum, as for the purpose of rail traffic noise prediction it often suffices to model only the dominant sources. Measurement methods are outlined to determine the noise emission spectra, from which extrapolations are made to obtain estimates for different operating conditions. [Copyright &y& Elsevier]

Abstract: Fan noise is one of the principal noise sources generated by a turbofan aero-engine. At supersonic fan speeds, fan tones are generated by the “rotor-alone” pressure field. In general, these tones can be well absorbed by an inlet duct acoustic liner, apart from at high supersonic fan speeds. However, in practice inlet duct liners contain acoustically hard longitudinal splices which cause scattering. This leads to acoustic energy being scattered into a range of different azimuthal mode orders, similar to the modal content resulting from rotor–stator interactions. The effectiveness of an inlet duct lining is reduced because acoustic energy is scattered into modes that are poorly absorbed by the liner. In this article, the effect of this acoustic scattering is examined by three-dimensional finite-element simulations of sound transmission in a turbofan inlet duct. Results include predictions of the sound power transmission loss with different splice widths, and at different supersonic fan speeds. These results demonstrate how acoustic scattering by liner splices can adversely affect fan tone noise levels at low supersonic fan speeds, but have little adverse affect on noise levels at high supersonic fan speeds. The potential noise benefit that could be achieved by manufacturing thinner splices is also examined. [Copyright &y& Elsevier]

Abstract: Multidisciplinary Design Optimization (MDO) is an essential part for low noise axial fan design since various parameters, such as flow rate, efficiency, noise etc., should be considered. For this reason, Response Surface Method (RSM) design technique is adopted as an axial fan design method. RSM has an advantage of choosing objective functions and constraint conditions unrestrictedly on a design space. However, RSM needs a lot of independent variables to construct a proper response surface. Thus an efficient and accurate flow analysis tool is indispensable for optimization. In an axial fan, the discrete (commonly called Blade-Passage-Frequency) components are usually dominant in the noise spectrum. Especially the blade–guide vane interaction is one of most important noise sources. In order to predict this noise component efficiently at the design stage, a new free wake model named Finite Vortex Element (FVE) is devised to simulate this blade–guide vane interaction, which is very difficult to analyze numerically in a conventional free wake model. In this new free wake model, the blade–wake–guide vane interaction is described by cutting a vortex filament when the filament collides with a guide vane. This FVE model is compared with a conventional curved vortex methodology and verified by a comparison with measured data to show its effectiveness and validity. Then FVE model is coupled with RSM to implement a low noise axial fan blade optimization. Using this method, a reduction of 8 dB(A) at 2m from fan hub in the overall noise level is achieved while the flow rate and the efficiency are maintained as the values of the baseline blade, which implies that FVE wake model coupled with RSM is very effective methodology for MDO problems such as a low noise axial fan design. [Copyright &y& Elsevier]

This paper describes a frequency-domain numerical method for predicting noise radiation from ducted fans, including acoustic treatment and non-uniform background flow effects. The method solves the Euler equations linearized about a mean flow in the frequency domain. A pseudo-time derivative term is added to the frequency-domain equations so that a time marching technique can be employed to drive the acoustic field to steady state explicitly. This approach makes distributed parallel computing more viable for equations of this type and will allow for future use of well-known convergence acceleration techniques, such as multigrid, to obtain the solutions efficiently. Simulations of the JT15D static test inlet are performed including the effects of liners, and the results are compared with experimental data. A generic engine geometry is used for demonstrating further the prediction capability of the code, calculating the attenuation effects of different liner impedances and liner installation locations on the radiated sound fields. [Copyright &y& Elsevier]

Discrete-frequency tones generated by rotor–stator interactions are of particular concern in the design of fans and compressors. Classical theory considers an isolated flat-plate cascade of identical uniformly spaced airfoils. The current analysis extends this tuned isolated cascade theory to consider coupled aerodynamically detuned cascades where aerodynamic detuning is accomplished by changing the chord of alternate rotor blades and stator vanes. In a coupled cascade analysis, the configuration of the rotor influences the downstream acoustic response of the stator, and the stator configuration influences the upstream acoustic response of the rotor. This coupled detuned cascade unsteady aerodynamic model is first applied to a baseline tuned stage. This baseline stage is then aerodynamically detuned by replacing alternate rotor blades and stator vanes with decreased chord airfoils. The nominal aerodynamically detuned stage configuration is then optimized, with the stage acoustic response decreased 13 dB upstream and 1 dB downstream at the design operating condition. A reduction in the acoustic response of the optimized aerodynamically detuned stage is then demonstrated over a range of operating conditions. [Copyright &y& Elsevier]