Search

International audience; A modal model, hitherto used to predict electromagneticand ultrasonic surface waves, is employed in deriving aneffective impedance for a grooved acoustically hard sur-face. This is inserted in the classical theory for propagationfrom a point source above an impedance plane to predictpoint-to-point propagation above regularly spaced grooveswith rectangular cross sections. Predictions of excess at-tenuation (EA) spectra obtained in this manner compareclosely with those obtained by a numerical method (BEM)but take a fraction of the time. The modal method is ex-tended to predict EA spectra above metasurfaces consist-ing of repeated arrays of variable depth grooves with dif-ferent depths of a low flow resistivity porous material ineach groove. The porous infill is predicted to reduce thenegative EA peaks associated with quarter wavelength res-onances in the grooves. By applying the modal model ef-fective impedance theory, porous-infill groove arrays havebeen designed to yield wideband excess attenuation spectrawith potential for reducing noise from surface transport.

Proceedings of the ICA 2019 and EAA Euroregio : 23rd International Congress on Acoustics, integrating 4th EAA Euroregio 2019 : 9-13 September 2019, Aachen, Germany / proceedings editor: Martin Ochmann, Michael Vorländer, Janina Fels 23rd International Congress on Acoustics, integrating 4th EAA Euroregio 2019, ICA 2019, Aachen, Germany, 9 Sep 2019 - 13 Sep 2019; Aachen (2019)., Published by Aachen

A surface composed of regularly-spaced rectangular strips is known to produce airborne acoustic surface waves that travel slower than the speed of sound and attenuate in the propagation direction with the square root of distance. Further laboratory measurements of propagation over a regular array are reported which seem to indicate the possibility of more than one surface wave. A code based on the Boundary Element Method has been extended to the time domain. It has been used to compare the data and to explore the surface wave effects. Varying the spacing, thickness and height of the strips is shown to affect the magnitude and frequency content of the air-borne surface waves generated by the periodic strip surface.

The potential use of forests or narrow belts of trees alongside surface transport corridors to reduce noise is often dismissed. This may be a consequence of conflicting experimental evidence and incomplete understanding of the various attenuation mechanisms involved. Important mechanisms include (a) destructive interference between sound travelling directly between source and receiver and reflected from the ‘acoustically-soft’ ground formed by decaying leaf litter, (b) the influence on this interference of loss of coherence due to the reverberant scattering by trunks and branches and (c) visco-thermal scattering by foliage. \ud \ud First the paper lists experimental evidence of significant attenuation due to forests and tree belts. Subsequently models for predicting the various contributions to overall attenuation are outlined. Predictions of the ‘soft’ ground effect are made using physically admissible ground impedance models. Incoherence due to trunk and branch scattering is modelled as enhanced turbulence. An empirical formula involving leaf area density and mean leaf size is used to predict foliage attenuation. Predictions that sum these contributions are compared with data. Regular or near-regular tree planting can cause ‘sonic crystal’ effects but the relatively sparse distributions of scatterers in realistic tree planting schemes means that the first band gap due to the periodic structure is weak. On the other hand the first pass band may be reduced and the second band gap can be enhanced by perturbing the tree locations with respect to periodic spacing. Finally results of numerical simulations showing the potential for traffic noise reduction by narrow tree belts are outlined.

Transport noise is an ever present concern in urban areas affecting the quality of life for millions of people. The traditional noise barrier is not always a convenient method of noise control and can divide communities. Deliberate introduction of small scale (0.3 m high or less) periodic roughness on otherwise acoustically-hard ground has been investigated as a way of reducing noise near to a surface transport corridor. The roughness alters the effective surface impedance of the ground and thereby creates a 'soft' ground effect. Moreover the effectiveness if the rough surface is not reduced significantly if there are pathways through it. However the rough ground also creates surface waves that must be absorbed for the noise reduction to be effective. An alternative way of reducing surface waves may be to alter the periodicity. The effects of altering the periodicity of circular rods placed on a hard surface in the laboratory have been investigated. Predictions of multiple scattering theory and a boundary element code have been compared with the experimental data.

Although plants and cultivated areas could be a visually acceptable means of noise abatement in urban areas, their effectiveness for reducing sound levels is still an open question. There have been several studies of the attenuation by trees bushes and crops but few theories predict all the observed attenuation spectra fully. It is known that the attenuation due to ground effect alone can account for low frequency (below about 1 kHz) insertion loss spectra while the high frequency behaviour is determined by multiple scattering from the stems/trunks and more importantly surface viscous and thermal absorption by leaves. A simple engineering model for scattering, viscous and thermal losses is proposed based on data in the literature. This model is shown to give reasonable agreement with recent insertion loss data obtained with a winter wheat crop and a laurel hedge.

Synthetic partially reticulated polyurethane foams are deployed for many situations for noise and vibration control. So understanding the behaviour of these foams when exposed to acoustic and/or mechanical waves is essential. Much progress has been made through identification and non-acoustic measurements of parameters introduced in the seminal works by Biot and, subsequently, by Allard, that describe the acoustical and elastic properties. Conversely, these parameters can be estimated from measurements of vibroacoustical responses. In principle measurements of acoustic-to-frame coupling should be more effective than measurements that focus on either acoustic or frame behaviour separately. This idea has been pursued in many papers co-authored by Walter Lauriks and Jean- Francois Allard. This paper describes recent demonstrations that pore-related and elastic parameters of air-saturated plastic foam can be estimated from data obtained by conjunctive use of microphones and a laser Doppler vibrometer.

Sound propagation over a smooth hard surface can be significantly altered by introducing roughness. For near grazing sound propagation over a smooth hard surface, the first destructive interference occurs at relatively high frequencies. However, the ground effect dips, corresponding to the first destructive interference, observed in Excess Attenuation (EA) spectra measured over rough surfaces, are at significantly lower frequencies. Sound propagation over periodically spaced roughness elements and randomly spaced roughness elements with different cross-sectional profiles (semi-cylindrical, rectangular and wedge-shaped strips) is investigated. Periodic spacing results in multiple distinct maxima, compared with the single broad EA maxima observed for random spacing. Roughness also causes surface waves. These surface waves are strongest near grazing incidence and their amplitudes and the frequencies at which they occur depend on the roughness height and mean centre-to-centre spacing. Numerical models such as Multiple Scattering Theory (MST) and the Boundary Element Method (BEM) are used to make predictions of the EA spectra which are compared with measurements. [Work supported by European Community FP7 project number 234306].

In horticulture, the soil structure, moisture content and strength have profound effects on plant growth. When soils are regarded as porous media, sub-surface wave propagation can be indicative of the soil status. Such propagation can be initiated by sources of airborne sound through acoustic-to-seismic coupling. Measurements of the ratio of near-surface sound pressure to acoustically induced solid particle motion can be exploited to estimate the acoustic and elastic properties of soils. Traditional methods of monitoring seismic signals by use of buried geophones are invasive and may affect the soil samples of interest. This paper describes a non-invasive acoustic-seismic technique. A loudspeaker was used to generate airborne sound, and the reflected sound and the subsequent vibration of the soil surface were recorded by microphones and a laser Doppler vibrometer respectively. These data were used to estimate acoustic and elastic soil parameters through an optimisation process minimising the differences between the data and model predictions based on incorporation of a modified Biot theory in a Fast Field Program. Example results of laboratory and field measurements are reported.

This paper reports work that is part of a larger project concerned with correlating differences in the perceived playing characteristics of musical wind instruments with physical differences between them. Here we focus on five different student model clarinets. Some of the practical difficulties of (i) directly measuring the bore profiles of the clarinets and (ii) measuring their input impedances are discussed. Results are presented which show significant differences in bore profile between the five instruments, leading to clear differences in the frequencies and magnitudes of their resonance peaks. In addition, some initial thoughts regarding playing tests designed to establish clarinetists’ perceptions of the instruments are considered.

Sonic crystal structures consisting of periodically-arranged solid vertical cylinders can act as sound barriers at certain frequencies. Their performance depends on the filling fraction which is determined by the spacing and cylinder radius. To be effective the filling fraction must be high i.e. the diameter of the scattering elements must be comparable to the centre-to-centre spacing. This means that periodic arrays with relatively low filling fractions such as in trees belts are not effective as traffic noise barriers. The effects of partially perturbing the positions of sonic crystal elements have been investigated by modelling and laboratory measurements and have been shown to improve the insertion loss of the periodic structure. It is argued that partial perturbation of regular tree planting near highways will improve their noise attenuation.

Fluid-saturated porous elastic solids support two compressional wave types, often called 'fast' and 'slow', and a shear wave. Walter Lauriks recognised the commonality of the physical mechanisms which govern the acoustical properties of such media and initiated a great many studies ranging from the acoustical monitoring of marrow filled bone to properties of air-filled bread, soils and snow. The Biot parameter values that create distinct slow waves in transmission experiments are explored. There are two types of surface waves at a poroelastic boundary: a fluid-coupled Rayleigh wave and a wave that depends on the surface impedance. A rough hard surface is also able to generate an acoustic surface wave. Walter Lauriks studied pore-related surface waves and those over a periodically-rough hard surfaces. It is shown that propagation over hard-backed square lattice grating layers modelled previously using modal analysis can be described as that over a square pore layer. Similarly propagation over rectangular grooves or bosses also modelled previously using modal analysis can be modelled as that over a slit pore layer.

To predict the acoustical effects of green roof structures it is necessary to be able to model the acoustical properties of their materials including gravel. For time domain calculations it is convenient to use the phenomenological model due to Zwikker and Kosten. However this phenomenological model is related to a low frequency/high flow resistivity approximation of more ‘exact’ identical pore models. The results of fitting predictions to short range level difference data and to impedance data for gravel indicate that the Zwikker and Kosten model does not perform as well as other models for low flow resistivity surfaces. For frequency domain calculations of sound propagation over extensive green roofs other models should be more accurate. For time domain calculations use of a two-parameter version of a relaxation model is recommended.

Surface roughness introduced on to horizontal ground between a noise source and receiver has previously been demonstrated to provide additional attenuation for sound near grazing incidence. Roughness-enhanced ground effect has been shown to depend on the relative source-receiver geometry. In most studies of the effects of surface roughness there has been line of site between source and receiver [see for example “Diffraction assisted ground effect” Bashir, Attenborough, Hill and Taherzadeh, Proc. 2010 ASA, Cancun] whereas for potential applications this may not be the case. Numerical calculations and data indicate the benefits of applying specific surface roughness profiles to large-scale acoustic obstructions such as an earth berm. Boundary element calculations have been used to investigate the attenuation of sound by the scattering of surface/creeping waves from surface roughness. The influence of different roughness cross-sections and distributions on the attenuation of noise from standard road traffic sources are discussed [work supported by European Community FP7 project number 234306].