Acoustic and thermal behaviour of cross-insulated timber panels

Abstract This paper presents an experimental study about the acoustic and thermal behaviour of a novel type of sandwich panel solution based on Cross-Laminated Timber (CLT). The layout of the new panel, named Cross-Insulated Timber (CIT), is similar to the one of a five-layer (CLT) panel, but the inner layer is made of polyurethane (PUR) rigid foam instead of timber, aiming to rationalize the wood volume and combine it with a low-density core layer for improved thermal insulation and reduced weight. For the acoustic behaviour, the study included the assessment of (i) the airborne and impact sound insulations of the developed panel, as well as of equivalent-thickness CLT solutions, and of (ii) the suitability of analytical models to describe such sound insulation. In general, the airborne and impact sound insulation of the CIT panels were found to be lower than that of CLT panels with equivalent thickness, namely due to their mass differences. Analytical models available in the literature to describe the airborne sound insulation of typical sandwich panels failed to describe the behaviour of the CIT panels. The Sharp model for homogeneous isotropic elements was able to provide a good description of the airborne sound insulation behaviour of the thicker CIT panels; however, it also failed to describe the behaviour of the thinner CIT panels. An adaptation of such model, by replacing the typical critical frequency due to bending by the dilatational frequency, enhanced the accuracy of predictions. An empirical model to estimate the impact sound insulation of both CIT and CLT panels was also proposed and fitted quite well the results. Regarding the thermal behaviour, some variability was found in the thermal resistance of the CLT and CIT panels, which was attributed to the intrinsic scatter in their material properties, namely for wood. Despite such variability on the coefficient of thermal conductivity of wood, its overall influence on the thermal resistance of the CIT panel solutions was actually marginal due to the considerably lower thermal conductivity of PUR. As expected, the analytical model (conventional heat conduction) was able to predict the thermal resistance of the CIT panels, with good accuracy.