Computer simulation of the measurement error with the guarded hot-plate system
Résumé
A guarded hot-plate system used for the measurement of the thermal conductivity of building materials is presented. This apparatus is in conformity with the experimental method described by the French standard Norme Française X 10-021 and with the international standard ISO 8302 for such testing. According to the standard, tolerance to the difference of temperatures between the central measurement zone and the first peripheral zone is 0.02 K. This condition is not severe enough to guarantee a good evaluation of the conductivity. The losses through the lateral faces of the system are very important, especially for insulation samples, and are accentuated besides by the absence of a thermal cut to the level of the hot plate. By means of a computer simulation in three dimensions, the errors involved in the experimental technique used have been determined, in the case of two symmetrical samples placed on each side of the guarded hot plate. These errors are estimated for two different materials, with various thicknesses, and various temperature gradients.