Home HomeLeif Akerlof, Ingemansson Technology AB (Sweden)
In Stockholm an old electric power station built in the 1920s is being converted into a combined theatre and apartment building. The four storey theatre contains 400 seats and it is flanked with apartments inside the power station on one side and in a new building connected to the old one on the other side. The foyer of the theatre and a restaurant are situated on the ground floor directly below some of the apartments. This poses a number of problems. Before and during the design of the building an extensive acoustical investigation was made in order to fulfil the tough acoustical demands. It is hard to avoid disturbances from neighbours in a common apartment building and in this case the sound levels of the disturbances are quite different. The activities in the theatre must not cause disturbances in the apartments even though there are performances late at night. Some of the considerations that had to be taken are presented and also some solutions that were used.
sv970444.pdf (Scanned)
TOP
Luis Bento Coelho, CAPS-Instituto Superior Tecnico (Portugal)
D. Alarcao, CAPS-Instituto Superior Tecnico (Portugal)
A technique is presented for the characterization of sound fields in enclosures by considering the three dimension random walks of phonons inside the room, Different wall materials as well as arbitrary geometrical arrangements can be dealt with. The different transition probabilities are calculated considering the solid angles of all walls. At every transition, the phonon density over every wall is changed. The sound intensity at a referrence point is then evaluated by considering diffuse wall radiation. The room impulse response, reverberation time and inteligibility indexes can be calculated by assuming that every transition takes place at a rate given by the mean relaxation time. The model was seen to yield results in good agreement with simple theories for a quasi-diffuse sound field in en enclosure with a homogeneous wall arrangement. For cases other than this, the simulation results show that the sound intensity increases near to a highly reflecting surface. The reverberation time and the inteligibility indexes depend on the reference point location for a non-diffuse field. Computation of the model is fairly fast. Comparisons between theory and experiment for different shapes of auditoria showed good agreement.
sv970405.pdf (Scanned) TOPJohn L. Davy, CSIRO DBCE (Australia)
The 1996 version of the draft international standard ISO/DIS 3741, "Acoustics - Determination of sound power levels of noise sources using sound pressure - Precision methods for reverberation rooms" deleted the room qualification procedure for the measurement of discrete frequency components. The alternative multiple source position method has been retained. This paper shows that there is an error in the constant in the equation for determining the number of source positions in the retained alternative method. It also shows that the multiple source position method is not sufficient at low modal overlap. Thus the room qualification procedure should be reinstated. The measurement variance can be split into source position, receiver position and room variance. The room variance depends on the distribution of modal spacings. Earlier theoretical and numerical calculations used the Poisson or "nearest neighbour" distributions. Both these distributions produce non-zero room variance. The Gaussian Orthogonal Ensemble (GOE) distribution, which is currently believed to be correct, produces zero room variance at high modal overlap. At low modal overlap, the GOE and "nearest neighbour" distributions produce room variance values which tend towards the non-zero values produced by the Poisson distribution.
sv970548.pdf (Scanned) TOPElizabeth A. Lindqvist, Royal Melbourne Institute of Technology (Australia)
A car floor is normally comprised of a decorated soft trim on a heavy layer with a net weight of 4 - 8 kg/m2 and 2 - 4 mm thickness, which is backed by a porous or fibrous material. This is either simply fitted to the steel floor pan or glued to it. Such a system essentially constitutes a double leaf partition with a high damping filler. In order to facilitate design of carpeting systems to provide optimal sound insulation properties, a good model of sound transmission through such a structure is required. Several models of sound propagation through porous materials are available. These models have been compared and in particular the Biot model has been studied. This model predicts that three distinct waves will propagate through a porous material. Under what circumstances must each of these waves be included in the model and when can we neglect one or more of them? Do porous plastics behave in a similar way to fibrous materials? As is often the case, evaluation of a theoretical model is hampered by lack of experimental values of parameters. We have therefore sought to relate measurable quantities to the theoretical parameters.
sv970361.pdf (Scanned) TOP
Antonio P.O. Carvalho, University of Porto (Portugal)
Antonio E.J. Morgado, University of Porto (Portugal)
This study reports on objective and subjective acoustical field measurements made in a major survey of 36 Catholic churches in Portugal built in the last 14 centuries. Monaural acoustical measurements (RT, EDT, C80, D50, TS and L) were taken at several source/receiver locations in each church and a group of college students was asked to judge the subjective quality of music. The listeners in each church evaluated live music performances at similar locations in each room. Evaluation sheets were used to grade the listeners' Overall Impressions of room acoustics qualities and also Loudness, Reverberance, Intimacy, Envelopment, Directionality, Balance, Clarity, Echoes, and Background Noise. This paper concentrates on the relationships of the subjective parameters with the objective room acoustics measures and with the architectural features of the churches. The results are graphed and analyzed by comparisons. Correlation analyses and statistical modeling identified significant relationships among the measures. For instance, linear correlation coefficients (|R|) of 0.8-0.9 were found for the relationships: Reverberance/RT and Clarity/C80; the maximum |R| found was 0.93 for Echoes. Regarding architectural features the maximum |R| found was 0.87 between Intimacy and Tonal Volume.
sv970319.pdf (Scanned) TOP
S.M. Dance, South Bank University
B.M. Shield, South Bank University
Previous researchers have all found abnormal sound fields in empty enclosed spaces. The sound appeared to propagate in such a way as to increase in level with increasing distance from the sound source, although it returned to its expected level at the farthest distances. These anomalies were caused by interference, which was dependent on phase shifts.A computer model, PHASE, was previously used to predict interference effects at two frequencies in two spaces. The model has been optimised, extended and modified to predict across the full frequency range efficiently. To validate the model sound levels in a reverberation chamber, as measured by Hodgson, were predicted at both a low and a high frequency. In addition the overall sound levels were predicted in an empty factory space to illustrate the accuracy of the model for practical purposes. To this end the predictions of an intensity based model, CISM, were used as a comparison.
sv970273.pdf (Scanned) TOPM. Asselineau, Peutz & Associates Acoustical Consultants (France)
One of the urban environment challenges is to manage to locate such leisure places as cinemas, restaurants, or even discotheques, as close as possible to the living quarters. While this can be reasonably achieved in brand new buildings, with careful acoustical and urban planning and engineering, it often. proves tricky, or even impossible, to practice in the kind of older buildings that usually are to be found at the core of European cities.This paper aims to submit and discuss a few case studies, ranging from real success to total failure, looking into the context of the situation as well as into the technical solutions and legal strategies that were applied.One of the main conclusions is that whenever any benefit is proved to result from the presence of such leisure places close to home, the neighborhood can much more readily accept the acoustical implications. However, when no thought is given to the acoustical problems, the technical solutions - and the relational efforts -needed to correct the situation, often prove to be beyond the capabilities of the operators.
sv970259.pdf (Scanned) TOPMark Debevc, Boral Plasterboard (Australia)
The method of directly attaching plasterboard by adhesive daubs to both sides of a masonry wall will result in increased sound transmission which can occur at low, mid, or high frequencies. The proximity and the magnitude of reduction in transmission loss over the frequency bands is dependent on combinations of the number of fixing points and the depth of the cavity behind the plasterboard. The effect on the determined Sound Transmission Class (STC) ranged from a 1 to 8 dB decrease over that of the bare wall. Measurements were conducted on a 110 mm hollow block wall with the plasterboard bonded by various configurations in the adhesive daub centres and thickness. This paper will discuss contrasts in measurement results of airborne sound insulation which show that relatively small structural changes can make large differences to the transmission losses, and hence the STC, when plasterboard is directly fixed to both sides of a block wall.
sv970035.pdf (Scanned) TOP