Prediction of the Sound Isolation of Multi-layered Small Sized Elements

S. Schoenwald, E. Gerretsen, H.J. Martin

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Often it is a problem to get a good estimate by prediction of the sound transmission loss R of multi-layered structures, like glazing, panels, etc.. A practical solution could be based on the transfer matrix approach for infinite plates for the prediction of the trans-mission coefficient t that is a function of the angle of inci-dence of sound. Since in reality a more or less diffuse sound field exists in rooms, we have random incidence. If an ideal diffuse sound field is assumed the prediction will underesti-mated grossly the sound transmission loss due to low values for grazing incidence. Beranek [1] suggests to limit the an-gles of incidence up to 78° for limp single layer panels. For other types of structures the cut-off angle usually varies between 70° to 85° to get a good estimate of the single num-ber rated value, but for the frequency dependence of stiffness controlled panels this ad-hoc correction does not give a satis-factory agreement at all. To improve the prediction, an ap-proach that considers the finite size of the structure is com-bined with a non-uniform directional distribution of incident power.
Original languageEnglish
Title of host publicationFortschritte der Akustik - 33. Jahrestagung fur Akustik, Stuttgart, Germany
EditorsS.-R. Mehra, P. Leistner
PublisherDeutsche Gesellschaft für Akustik
Pages253-254
Publication statusPublished - 2007
Event2007 DAGA: 33. Jahrestagung für Akustik, 19-22 Marz 2007, Stuttgart, Deutschland - Stuttgart, Germany
Duration: 19 Mar 200719 Mar 2007

Conference

Conference2007 DAGA: 33. Jahrestagung für Akustik, 19-22 Marz 2007, Stuttgart, Deutschland
Country/TerritoryGermany
CityStuttgart
Period19/03/0719/03/07
OtherDAGA 2007

Fingerprint

Dive into the research topics of 'Prediction of the Sound Isolation of Multi-layered Small Sized Elements'. Together they form a unique fingerprint.

Cite this