Computational characterisation of the heat flow meter method applied to moist bio-based insulating building materials

This paper investigates the role of moisture content on the measurement of the thermal conductivity of wood-based low-density fibreboards (LDFs) used for building insulation. To that purpose, a computational code of coupled transfers was used to simulate the widely used Heat Flow Meter (HFM) method. To widen the range of study, the thermal conductivity is obtained with a mixing law model to account for the properties and fractions of the constitutive phases. This mix law model was fitted from thermal conductivity values reported for commercial LDFs. The prediction of the heat flow density during the HFM test was compared with the measurements for dry and wet LDF. Through a series of case studies, the simulations showed that, in the case of moist products, the thermal conductivity determined by the HFM method was overestimated by up to 20%. This overestimate depends on the density distribution, thickness, moisture content, and measurement parameters such as time and the equilibrium criterion. Recommendations are made to limit this error.