Thermal transmittance of a composite lightweight wall panel with integrated load-bearing structure: Experimental versus numerical approach

Abstract

One of the most important parameters when it comes to heat losses in buildings is the thermal transmittance or U-value. Therefore, great importance should be given to the determination of U-values, especially for elements where there is a high thermal bridge effect, as is the case with lightweight steel frame (LSF) structures. Since LSF structures are usually geometrically more complex, especially when diagonal elements are present, the determination of the overall U-value of these elements is usually done either on scale models in the laboratory or by numerical methods. This paper compares different methods for determining the total U-value for four different polyurethane foam-filled LFS walls and a reference wall made of EPS. Analytical (ISO 6946), experimental (Guarded Hot Box Method and HFM) and numerical 2D and 3D calculations were used to determine the U-value. The aim of the comparison was to verify which methods can be used for more complex geometries of LSF walls when there is a stronger influence of point thermal bridges due to the additional diagonal bracing. All methods showed similar U-values with the highest absolute deviation of 17.18% between the HFM and the 3D numerical calculations. The analytical method for inhomogeneous building elements given in ISO 6946 agrees well with all methods with the maximum absolute deviation of 8.83% between the analytical and HFM method. The work showed the importance of the placement of the HFM sensor for the determination of the surface heat flux, as incorrect placement of the sensor can result in inadequate U-values that deviate up to 167% from the true value.