Limit of validity of the log-linear model for determining thermal properties of light insulation materials with cylindrical hot probe

This work aims to contribute to the long term objective of developing in situ thermal characterization of building insulation material. Cylindrical hot probe can be considered the most adapted instrument but the heat flux is radial then unreliable results can be obtained if samples are not homogeneous or radially symmetric like common layered insulation materials. To avoid this difficulty, we have studied glass wool which has been performed by a “crimping” process in order to obtain a more isotropic structure. The main problem treated in this work concerns the identification procedure to determine thermal characteristics of insulation materials. The standard procedure to determine thermal conductivity requires linear thermogram representing temperature rise against the logarithm of the time. Sigmoid curve (not linear curve) has been obtained for insulation material. So we suggest to develop identification procedure applied from the whole thermogram and not only from the linear part. The interest of this procedure is also to determine simultaneously thermal conductivity and thermal capacity of the surrounding material. Analytical solution based on Jaeger model has been employed to resolve the time dependant problem of heat transfer. The mathematical approach of the model of Blackwell [22], Jaeger [23] and Devries [24] are very similar. However, none of these models described a proper way to calculate the thermal conductivity and thermal capacity simultaneously from thermogram. This work is referred to the work of van Haneghem [17] and Hutter [31] for granular materials which developed or applied revised model based on that of Jaeger. Inverse technique with optimization routines from Matlab has been used to minimize the difference between measured and estimated temperature thermograms.