B.M. Gibbs

Acoustic absorption behaviour of an open-celled aluminium foam

Metal foams, especially close-celled foams, are generally regarded as poor sound absorbers. This paper studies the sound absorption behaviour of the open-celled Al foams manufactured by the infiltration process, and the mechanisms involved. The foams show a significant improvement in sound absorption compared with close-celled Al foams, because of their high flow resistance. The absorption performance can be further enhanced, especially at low frequencies, if the foam panel is backed by an appropriate air gap. Increasing the air-gap depth usually increases both the height and the width of the absorption peak and shifts the peak towards lower frequencies. The foam samples with the smallest pore size exhibit the best absorption capacities when there is no air gap, whereas those with medium pore sizes have the best overall performance when there is an air gap. The typical maximum absorption coefficient, noise reduction coefficient and half-width of the absorption peak are 0.96–0.99, 0.44–0.62 and 1500–3500 Hz, respectively. The sound dissipation mechanisms in the open-celled foams are principally viscous and thermal losses when there is no air-gap backing and predominantly Helmholtz resonant absorption when there is an air-gap backing.

Application of a finite-element model to low-frequency sound insulation in dwellings

The sound transmission between adjacent rooms has been modeled using a finite-element method. Predicted sound-level difference gave good agreement with experimental data using a full-scale and a quarter-scale model. Results show that the sound insulation characteristics of a party wall at low frequencies strongly depend on the modal characteristics of the sound field of both rooms and of the partition. The effect of three edge conditions of the separating wall on the sound-level difference at low frequencies was examined: simply supported, clamped, and a combination of clamped and simply supported. It is demonstrated that a clamped partition provides greater sound-level difference at low frequencies than a simply supported. It also is confirmed that the sound-pressure level difference is lower in equal room than in unequal room configurations.

The use of power flow methods for the assessment of sound transmission in building structures

Abstract A description is given of a versatile method of analysis of noise transmission in buildings. This method incorporates power flow techniques and has the advantage that a unified approach is possible to both the direct and indirect transmission paths and is therefore equally applicable to transmission between rooms which are adjacent or several rooms or floors apart. The first configuration only lies within the range of classical theories. In this analysis, the building is regarded as an assembly of plate elements. Vibrational energy flows to and from each plate across common structural junctions in a manner analogous to heat transfer between bodies at different temperatures, and can also flow to and from the enclosed air volume. The solution of sets of simultaneous energy-balance equations gives the steady-state vibrational energy of each plate at their common junction. The pressure wave energy generated within a room by plate (or wall) vibrations is also predicted. The vibrational energies are predicted from the modal density, loss factor, coupling loss factor, and radiation loss factors, which are assessed theoretically and experimentally. Numerical calculations, involving a computer, yield values of the transmission coefficients of bending and longitudinal wave at cross-junctions, T-junctions and corners of reinforced concrete plates. The theory is compared with measurements involving quarter-scale models over a frequency range of measurement of 400 Hz-12·5 kHz.

Sound transmission and mode coupling at junctions of thin plates, part I: Representation of the problem

A description is given of a versatile method of analysis of sound waves generated at a junction of plates as a result of a wave incident on any one of the plates. Thin plate theory is used in calculating mode coupling and energy intensity and the description is generalized in that the junction can be of two to four plates, the incident wave can have any mode of vibration, and dissipative loss factors can be included in the calculation. Results are compared with those of previous workers for the case of a bending wave incident at a cross junction and at a corner and the description is extended to include in-plane vibration incident at the same junctions.

Towards a structure-borne sound source characterization

Abstract Characterization of mechanical sources is complicated since up to six forces and moments contribute at a contact and the response at one contact is the result of forces and moments at all contacts. For the transmission process, the quantity of prime interest is the complex power, and, therefore, a proper source characterization should relate to it. The source descriptor does so and offers a consistent base for studies of source characterization. Theoretical and experimental progress on the source descriptor is reviewed. From inspection of the source descriptors and the mobilities from which they are composed, it is observed that the uncoupled component source descriptors can establish the bandwidth and dominant components of excitation. The ‘signatures’ of the point mobilities required allows contacts to be classified as mass-, stiffness- or resonance-controlled component, establishing simplifications with respect to source characterization.

The acoustic performance of building façades in hot climates: Part 2—Closed balconies

Abstract The acoustic protection afforded by a balcony open to the street but enclosed on all other sides was investigated by means of one-tenth scale models. The results indicate that the protection is slight and the frequency invariant when the receiver is exposed to a direct path. The protection increases with increased sound incident angle and balcony depth and becomes frequency dependent when the direct path between source and receiver is screened. In general, the protection measured by the author, using scale models, compared well with the full-scale measurements of other workers. The measured protection of a balcony is considerably improved when perforated screens of unusual geometry are installed. A maximum protection due to these screens is obtained at first and second floor levels and at balcony depths of 2 m and less.

The effect of construction material, contents and room geometry on the sound field in dwellings at low frequencies

Abstract An experimentally validated finite element method is used to model the sound level in rooms at low frequencies. It is demonstrated that the dimensions of rectangular rooms strongly influence the sound pressure level difference. Additional factors were investigated which are not normally considered in the frequency range where diffuse sound field conditions can be assumed. Three effects were investigated: room damping due to wall vibrations, furniture, the effect of small deviations from simple rectangular shapes. It is confirmed by field measurements that the vibrations of masonry walls and floors introduce less damping than surfaces of lightweight construction. Assigning to the FE model a damping equivalent to a surface absorption of 0.02 reproduces the effect of walls of heavyweight construction. Damping equivalent to a surface absorption of 0.15 reproduces the effects of plastered timber-frame walls, floors and ceilings. The work was briefly extended to a room pair built with heavyweight and lightweight material of construction. The modification of the shape of the room frequency response highlights well the effect of material of construction. In-situ and laboratory measurements show that furniture has little effect on steady-state room response below 100 Hz. Modelling a wall recess smaller than 0.5 m improved the agreement between prediction and measurements but the assumption of a simple rectangular room remains appropriate.

Sound transmission and mode coupling at junctions of thin plates, part II: Parametric survey

In the second part of this discussion of sound transmission at the junction of structural plates results are presented of a parametric survey conducted on the performance of a T-junction of concrete plates of various thickness, density, bending rigidity and loss factor. It is seen that alteration of bending rigidity and density produces greater variation in transmission loss than that of plate thickness or material loss factor. Indeed an increase in loss factor can lead to a detrimental increase in mode coupling between travelling bending waves and in-plane waves at all angles of incidence. In general in-plane vibration suffers less reduction in level at low frequencies at junctions of dissimilar plates than does bending vibration.

The derivation of eigenvalues and mode shapes for the bending motion of a damped beam with general end conditions

The eigenvalues and mode shapes for the bending motion of a beam with restricted classes of end support, and generally undamped, have been derived separately by a number of authors for use in particular applications. A general method is presented here for the derivation of the complex eigenvalues and eigenvectors for a uniform beam governed by the classical fourth order bending wave equation, the constraints of the supports to shear and rotation being expressed as complex impedances with any desired frequency dependence and damping (with any necessary frequency dependence) being included in the beam material as well as in the supports. The matrix describing the motion is written in a form which can be handled numerically on a computer to provide finite, complex, solutions for any set of dimensonless parameters (describing the magnitude of the support constraint in terms of the beam characteristics) each permitted to vary over the whole range from zero to infinity. The distinction between the effects of damping in the beam material and in the supports is considered, and it is noted that the effect of material damping is dependent on the end conditions.

The oblique incidence measurement of transmission loss by an impulse method

Abstract The oblique incidence transmission loss of a free standing panel has been determined experimentally with the use of short duration impulsive signals. The geometry of the source, panel and receiver is such that the direct signal can be isolated and, on subsequent analysis, the infinite panel response obtained. Agreement with mass law is good and the angular and spectral variation of coincidence is clearly seen. Closer inspection of the time history of the transmitted waves shows a signal which arrives after the direct signal has effectively finished and before the arrival of scattered waves from the edge of the plate. Frequency analysis of this component reveals a coincidence type dip which is independent of angle and frequency.