Surface temperature measurement on wood

Wood is a hygroscopic material, which exchanges moisture continuously with the surrounding environment. This process involves heat of sorption phenomenon which is either a releases or adsorption of heat, resulting in temperature change on wood surface, which can potentially be utilized to reduce indoor energy consumption. The main objective of this research was to determine a suitable method and develop a protocol to accurately measure the temperature change on wood surface during exchanging moisture with the ambient environment. While conventional invasive measurement techniques were not considered to be suitable for this purpose, thermography demonstrated the advantages to measure surface temperature and was thus employed as the main technique to measure the change in wood surface temperature during moisture sorption. The measured protocol was validated experimentally by recording temperature change on pine wood surface which is subjected to a change in relative humidity of the environment instantaneously from relatively 0 to 95%. From experimental results, the anisotropy property in wood clearly showed the influence on the temperature change on pine wood during adsorption. The maximum temperature rise on a transverse surface was recorded as approximately 3.6 ± 0.4oC, whereas it was only approximately 1.4 ± 0.4oC in cases of radial and tangential surfaces. The experimental uncertainty obtained from 10 repetitions for each direction surface was considerably lower compared with the estimated uncertainty from literature review as 1.17oC. Therefore, it confirmed the suitable ability of thermography to record temperature change on wood due to the heat of sorption phenomenon. The measurement protocol established from this study can be applied to investigate the temperature variation occurring from different initial moisture contents during sorption in different wood species. Furthermore, the surface temperature data obtained from the heat of sorption experiments will be significant inputs for numerical models describing the behaviour of a wooden building with higher accuracy