Temperature Measurements in Timber Exposed to Fire Using Thermocouples

Fire tests on non-loaded steel-timber composite elements and their numerical simulation are achieved. Differences between measured and simulated temperatures in timber can be significant, reaching several hundred kelvins. This observation leads us to examine various instrumentation techniques. Inconsistent temperatures are measured depending on whether the thermocouple is parallel or perpendicular to the heat flow, and whether it passes through the fire or in an environment that remains cold. The local thermal disturbance caused by the presence of thermocouples is simulated by the finite element method. This approach allows us to assert that the differences observed between the measured temperatures are due to the contrast between the thermal properties of the used sensors and those of the surrounding timber. The simulation of the thermal field in a timber sample without thermocouple constitutes a comparative basis. It allows us to make evident that thermocouples inserted parallel to isotherms are the only ones able to perform correct temperature measurements in timber. This observation is confirmed by the comparison of charring rates deduced from measured and simulated temperatures with typical combustion behavior of softwood in standard fire tests (ISO 834 heating curve).