Determination of Thermal Conductivity of Wood Exposed to Fire based on Small Scale Laboratory Trials for Finite Element Calculations

Abstract: This study describes an approach to determine the thermal conductivity of wood at elevated temperatures. The aim is to be able to use the developed conductivity as input in structural elements in finite element calculations. The conductivity of pine wood and glue laminated timber with different densities and moisture contents have been evaluated where small scale one-dimensional laboratory trials have been carried out in a cone calorimeter. Steel temperatures were measured behind the exposed wood samples. Obtained temperatures from the experimental trials have been compared with back calculated steel temperatures in the finite element program TASEF (Temperature Analysis in Structures Exposed to Fire). In the back calculations the conductivity at 100 °C, 300 °C and 500 °C was altered in order to achieve a best fit steel temperature curve as the measured ones during the experimental trials. At 20 °C the conductivity was taken from the literature. Between these temperature levels the conductivity was assumed to vary linearly. The dehydration of the moisture content in the wood samples have been considered by including it in the specific volumetric enthalpy, i.e. the integral over temperature of the density and specific heat as input in the temperature calculation program TASEF. Regarding the thermal degradation, recommended formulas in Eurocode 5: Design of timber structures – Part 1-2: General – Structural fire design, have been applied. The final back calculated conductivity values of the studied pine wood at specific temperatures (20 °C, 100 °C, 300 °C och 500 °C) were determined by the cone calorimeter test to be as follows; 0.09 W/mK, 0.07 W/mK, 0.05 W/mK and 0.35 W/mK. Comparing with presented conductivity of wood in Eurocode 5 the developed conductivity in this study are generally lower. Derived conductivity values from the back calculations in TASEF have been reconsidered for the glue laminated timber by taken account of differences in density and moisture content. By using a developed conversion factor, so called “conductivity ratio”, new conductivity values could be obtained which then has been used as an input in TASEF. As a result, good similarities between calculated steel temperatures and measured steel temperatures could be seen. The implemented method, consisting of simple one-dimensional laboratory trials for determining the thermal conductivity is deemed to be promising.  However, further studies are needed to be done in order to increase the accuracy of the method.