Conference_programme: 17: Building acoustics
Lecture: EFFICIENT SOUND INSULATION PREDICTION MODELS FOR HOLLOW CORE MASONRY WALLS
Author(s): Van Den Wyngaert Jan, Schevenels Mattias, Reynders Edwin
Summary:
Bricks are often perforated to decrease weight, resulting in a decreased material cost and an easier handling on site. However, walls made of perforated brick have a lower airborne sound insulation than solid walls with similar weight and thickness. This can be explained by thickness resonances and the complex brick geometry. Numerical sound insulation prediction models that include the detailed perforation geometry of all individual bricks are computationally demanding since a refined mesh is needed in order to properly capture this geometry. However, the length scale of the perforations is generally much smaller than the wavelength of deformation. Both length scales can therefore be considered separately: first the behavior of individual bricks is considered (micro-scale), upon which this behavior is employed for constructing a homogeneous solid model of the entire wall (macro-scale). The most simple macro-scale models only take into account the bending behavior of an isotropic thin plate. More advanced macro-scale models account for thickness resonances and orthotropic behavior. An example of such a model that has been proposed in literature is a TMM model with spatial windowing. This model does not include the modal behavior of the wall, leading to an overestimation of the sound insulation at low frequencies. In this paper, an alternative macro-scale sound transmission model that takes thickness, orthotropy and modal behavior into account, is developed within the hybrid deterministic-statistical energy analysis framework. The sound fields in both rooms are modelled as diffuse in order to reduce the problem size. The vibration field in the wall is modelled deterministically, using an analytical formulation of the eigenmodes of a thick, simply supported orthotropic plate to achieve high computational efficiency. The model is validated with experimental data found in the literature. The computed results show a good correspondence with the data, while retaining a low computation time.
Corresponding author
Name: Mr Jan Van den Wyngaert
e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Country: Belgium
