Refine
Document Type
Has Fulltext
- no (12)
Is part of the Bibliography
- no (12)
Keywords
- Schallmesstechnik (11)
- Holzbau (10)
- Dowel connections (1)
- Eurocode 5 (1)
- Modal analysis (1)
- Timber floor (1)
- Trockenbau (1)
- Vibration (1)
- Vibration serviceability (1)
Institute
This paper investigates the prediction of low-frequency impact sound insulation for a solid timber floor formed from dowel-connected joists with Swiss hardwood.
A finite element model was developed for the dynamic response of the floor as it is neither homogeneous nor isotropic and has complex connections.
With point force excitation, a FEM model for the dynamic response of the floor was validated using experimental modal analysis in the laboratory.
Two different FEM models were developed, one using spring connectors and the other using join connectors. Good agreement between FEM and measurements in terms of the Modal Assurance Criterion (MAC) and eigenfrequencies was achieved for the first 14 modes with the spring connector model and for the first 7 modes with the join connector model.
However, for the vibroacoustic analysis it was necessary to use the join connector model due to the computational instabilities of the spring model above 100 Hz when it was coupled to the acoustic medium.
With mechanical excitation the radiated sound power from the underside of the timber floor was measured with sound intensity in order to calculate the radiation efficiency. Comparison of measurements and FEM showed reasonable agreement between 20Hz and 200Hz
This paper concerns the development and validation of Finite Element Methods (FEM) to simulate the dynamic response of a dowelled-joist timber floor. This is a solid floor comprised of timber joists connected using timber dowels with individual assemblies connected using inclined metal screws. The focus is on the structural dynamics in the low-frequency range up to 200 Hz which is the relevant range for impact sound insulation and vibration serviceability. Dowel connections between the joists that formed each assembly were modelled using either rigid or spring connectors in the FEM models. The validation against experimental modal analysis showed that both approaches were valid in terms of the eigenfrequencies, Modal Assurance Criterion (MAC) and the spatial-average velocity with point excitation. Whilst the FEM model with spring connectors had a higher number of correlated modes in the MAC analysis, this required removal of many spurious modes before predicting the response. The validated models were used to demonstrate the potential in predicting assessment parameters for vibration serviceability that are contained in EN 1995-1-1 (Eurocode 5). This predictive approach to the evaluation of vibration serviceability has the advantage in that it can be used for non-standard timber floors with non-standard boundary conditions or floor plans.
In this contribution a method for the prediction of the radiated sound out of the velocity field of a vibrating ceiling is presented. The numerical method was validated via measurements on a real structure using a pp-probe. It is based on Integral Transform techniques and can be applied in the postprocessing of a FEM simulation
(harmonic analysis) [1]. The method was verified by measurements on the real structure and can be used for parametrical studies.
Rounding off the Parts 1 and 2 of this publication nomograms, based on dimensionless parameters, can be developed to predict the sound radiation of light weight slabs. Hereby different sets of geometry as well as different configurations of ceilings can be studied in order to describe the acoustical and dynamical behaviour of wooden ceilings.
Numerical models for the prediction of vibroacoustical characteristics of light-weighted ceilings
(2013)
In order to set up guidelines for the design of light-weighted ceilings for timber constructions to be used by engineers in practice, investigations based on both measurements and numerical models have been carried out within the cooperative research project “VibWood”.
In this contribution the setup and the calibration of the numerical model of the structure as well as the prediction of radiated sound are discussed, where a special focus is set on a dimensionless description in order to deduce information for a wide range of system’s specifications.
The structure, consisting of a timber slab, a floating floor and a suspended ceiling, is built up in a Finite Element model, where the material properties of wood and the characteristics of the system (e.g.support conditions, contact phenomena dynamic properties of individual parts) are considered.
The model is parameterized in order to enable computations with varying geometry and material parameters.
After calibrating the FE-model with the help of measurements using model updating techniques dimensionless parameters are defined based on the Buckingham-π- Theorem and computations are carried out in order to specify guidelines for various systems. The radiation of sound is computed in a post processing using Integral Transform Methods.
It is common for timber ceiling constructions to include a suspended ceilings made of plasterboard. In order to minimize the radiation from impact excitation, especially in the low frequency range below 100 Hz, a detailed analysis of the vibration behaviour was necessary.
An experimental modal analysis was carried out and the radiated sound power from the suspended ceiling was measured using a sweep sine excitation by a shaker connected to the top layer of the timber floor construction.
Since the experimental modal analysis was conducted using a high frequency resolution and with a non-stationary sound field, it was necessary to check the validity of the intensity measurements by comparing it with 1/3-octave band measurements. The results show very good agreement.
By this means the radiation efficiency can be measured simultaneously with an experimental modal analysis of a structure.
Die mithilfe der Abklingmethode ermittelten Verlustfaktoren sind bei mehreren Eigenfrequenzen pro Frequenzband zuverlässig. Die Ergebnisse der Resonanzbreitenmethode beziehen sich dagegen auf die Eigenfrequenzen und sind nur bei einer geringen Modenüberlappung zuverlässig. Beide Methoden sind gleichwertig, wenn mehrere Eigenschwingungen im Frequenzband vorhanden sind und die Modenüberlappung gleichzeitig gering ist. Für einseitig überlagerte Resonanzkurven wurde eine einfache Alternative der Resonanzbreitenmethode vorgeschlagen.
Die Ergebnisse wurden im Rahmen des gemeinsam von der TU München, der HS Rosenheim und dem ift Rosenheim durchgef¨uhrten DFG-AiF Forschungsvorhabens ”Vibroakustik im Planungsprozess für Holzbauten” erarbeitet. Das IGF-Vorhaben 17328N/1 der Forschungsvereinigung HS Rosenheim wurde über die AiF im Rahmen des Programms zur Förderung der Industriellen Gemeinschaftsforschung (IGF) vom Bundesministerium für Wirtschaft und Energie aufgrund eines Beschlusses des Deutschen Bundestages gefördert.
Within the scope of the research project "VibWood", operational vibration analyses of timber floor constructions were carried out. In order to understand the vibration behaviour of and the sound transmission through the layers of the constructions the coupling between them was investigated. The transfer function from the point of excitation to a pattern of receiving positions located on each of the layers of the construction was measured using a swept sine excitation. As a result, the frequency range of decoupling between the base floor and the floating floor and between the base floor and the suspended ceiling could be identified and the individual vibration behavior analyzed. The measurements also included the radiated sound power from the suspended ceiling. There is indication that there is not necessarily a correlation between the eigenmodes of the suspended ceiling and the maxima of the radiated sound power