Direct and flanking transmission across timber concrete composite floors with cross laminated timber walls

Churchill, CE (2018) Direct and flanking transmission across timber concrete composite floors with cross laminated timber walls. PhD thesis, University of Liverpool.

Text 200717725_Jan2018.pdf - Unspecified Download (13MB)

Abstract

Timber-concrete composite (HBV) floors are widely manufactured by many construction companies. Cross laminated timber (CLT) elements are an effective way to produce more homogenised timber building elements. In this thesis, the applicability of measurement-based prediction methods to calculate the apparent sound reduction index (R’) for these two types of elements was investigated. A prediction model using Statistical Energy Analysis (SEA) was developed and validated to calculate the airborne sound transmission of a HBV floor. There was additional complexity in modelling this floor system compared with other SEA models found in the literature therefore two types of model were compared. To determine the stiffness of the floor it was proposed that bending stiffness predicted using the theories of Huffington and Troitsky provide a more suitable and flexible approach than that of Kimura and Inoue. All SEA models predicted a weighted sound reduction index to within 2.0dB of the measurements. For CLT plates, an accurate measure of the elastic moduli was sought to determine direct sound transmission. A low-frequency stiffness can be determined by modal measurement and optimisation of the material constants (≤ 100Hz a thin plate model is adequate). At mid to high frequencies (>300Hz) sound transmission of CLT was predicted using a frequency-dependant modulus determined from directly measuring the bending wavespeed. Elastic moduli were extracted from wavespeed data (the Young’s modulus was measured to be approximately 50% less stiff than the low-frequency stiffness) and in the y-direction the value of the shear modulus Gyz determined that a thick plate model should be applied in this direction. Finite Element Method (FEM) models of CLT plates and junctions using the elastic moduli determined using the methods described above were validated using measurements of freely and simply supported plates and three simple junctions (L-junction, rotated L-junction and T-junction). The contribution of residual modes (higher than the frequency range of interest) to the calculated point mobility was assessed. The transition to thick plate theory occurs within the frequency range of interest (100-5000Hz). This resulted in a modal density which increases with frequency and some implications of this are discussed. The best agreement between FEM model and measured data was obtained for the simplest (unrotated) L-junction. Finally, measurement based prediction methods to assess the flanking performance of CLT combined with the HBV floor were investigated. Agreement was obtained between the structure-borne and airborne methodologies if a correction factor, which accounts both for the thick plate properties of the CLT and the fit of measured results to the HBV model, was used. The Df path was the strongest transmission path and the most accurately predicted and hence it was possible to predict the apparent sound reduction index (R’) by summing the paths.

Item Type:	Thesis (PhD)
Divisions:	Faculty of Humanities and Social Sciences > Faculty of Humanities and Social Sciences
Depositing User:	Symplectic Admin
Date Deposited:	09 Aug 2018 16:03
Last Modified:	19 Jan 2023 06:33
DOI:	10.17638/03021424
Supervisors:	Hopkins, Carl ORCID: 0000-0002-9716-0793
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3021424