Hygrothermal phenomena in cellular concrete samples during the steady-state tests of thermal conductivity have been numerically simulated. It has been found that the most important factor for the accuracy of these tests is the initial moisture content of the test specimens. In the range of 70-85% RH the “heat pipe” mechanism is of importance, causing an additional latent heat transport, which increases apparent thermal conductivity. This phenomenon is strongly influenced by the material porosity and inner structure of pores, i.e., the shape of the sorption isotherms. Outside this range of relative humidity, errors caused by moisture transport and related heat effects are negligible. The error caused by moisture has been found for the cellular concrete with density of 600 kg/m3 to be smaller than for the 400 kg/m3 one. For smaller values of specimen thickness, the error is smaller. The smaller the temperature difference used between plates of the heat-flow-meter apparatus, the smaller is the moisture induced error observed. On the basis of the computer simulation results, graphs and tables with correction factors are developed to improve the accuracy of steady-state measurements of thermal conductivity for cellular concrete.

Some experimental results of steady-state measurements of thermal conductivity by use of the heat-flow-meter apparatus are presented. They concern three types of cellular concrete with densities of 400, 500, or 600 kg/m3 and additionally for two types of lightweight concrete, i.e., wood-concrete and EPS-concrete, at various moisture contents. These results qualitatively confirm the theoretical predictions presented in the first part of the paper.