The use of furnace tests to describe real fires of timber structures

Fire resistance is an important characteristic for all building structures regardless the building materials used. Methods for fire resistance testing were developed already before 1900 to measure the response of the structure in fire and compare different products. In the last decade, the increased popularity of timber buildings has led to a renewed interest in the performance of timber structures in fire and timber products were frequently tested in furnaces. Currently, some discussions question the validity of furnace test results for timber members which are carried out according to standards. Generally, it was stated that combustible and incombustible products are exposed to different thermal exposures when tested in furnaces or exposed in real fires. Additionally, some experts think that massive timber elements, e.g. cross-laminated timber (CLT), cannot be tested in furnaces as these products increase the fire load and, thus, statements in the framework of fire resistance testing are not possible. This paper investigates the validity of furnace resistance testing for combustible products and its limitations. It is shown that, firstly, the thermal exposure in fire resistance tests of incombustibles and combustibles is similar. Secondly, in addition to thermal exposure, the term fire exposure should be introduced where the oxygen concentration is described as the oxygen concentration significantly influences the behaviour of combustible material in fires. Thirdly, the furnace and compartment environment in flash-over fires is similar with respect to this fire exposure. Finally, it is not possible to directly use furnace test results to predict a compartment response in real fires including the cooling phase but recent investigations indicate that results from fire resistance tests can be used to predict burn-out when the mass loss of the timber specimen is measured.