A localized fire is a fire which in a compartment is unlikely to reach flash-over and uniform temperature distribution. Designing for localized fires is generally more difficult than for flash-over compartment fires because of the complexity of the problem. There is also a lack of experimental data. We report here on a full scale test series on a steel column exposed to localizedfires. The setup is a 6 meters tall hollow circular column, I= 200 mm with a steel thickness of 10 mm. The unloaded column was hanging centrally above different pool fires. Temperatures of gas and steel were measured by thermocouples, and adiabatic surface temperatures at the steel surface were measured by plate thermometers of various designs. The results are compared with estimates based on Eurocode 1991-1-2 which in all cases studied overestimate the thermal impact for this setup. The input from plate thermometers was used to compute the steel temperatures using finite element methods. Excellent agreement was found if the radiation exchange within the column due to asymmetry of the exposure was taken into account.

Purpose: The purpose of this paper is to propose a simplified representation of the fire load in computational fluid dynamics (CFD) to represent the effect of large-scale travelling fire and to highlight the relevance of such an approach whilst coupling the CFD results with finite element method (FEM) to evaluate related steel temperatures, comparing the numerical outcomes with experimental measurements. Design/methodology/approach: This paper presents the setup of the CFD simulations (FDS software), its corresponding assumptions and the calibration via two natural fire tests whilst focusing on gas temperatures and on steel temperatures measured on a central column. For the latter, two methods are presented: one based on EN 1993-1-2 and another linking CFD and FEM (SAFIR® software). Findings: This paper suggests that such an approach can allow for an acceptable representation of the travelling fire both in terms of fire spread and steel temperatures. The inevitable limitations inherent to the simplifications made during the CFD simulations are also discussed. Regarding steel temperatures, the two methods lead to quite similar results, but with the ones obtained via CFD–FEM coupling are closer to those measured. Originality/value: This work has revealed that the proposed simplified representation of the fire load appears to be appropriate to evaluate the temperature of steel structural elements within reasonable limits on computational time, making it potentially desirable for practical applications. This paper also presents the first comparisons of FDS–SAFIR® coupling with experimental results, highlighting promising outcomes. 

Purpose: In the frame of the European RFCS TRAFIR project, three large compartment fire tests involving steel structure were conducted by Ulster University, aiming at understanding in which conditions a travelling fire develops, as well as how it behaves and impacts the surrounding structure. Design/methodology/approach: During the experimental programme, the path and geometry of the travelling fire was studied and temperatures, heat fluxes and spread rates were measured. Influence of the travelling fire on the structural elements was also monitored during the travelling fire tests. Findings: This paper provides details related to the influence of travelling fires on a central structural steel column. Originality/value: The experimental data are presented in terms of the gas temperatures recorded in the test compartment near the column, as well as the temperatures recorded in the steel column at different levels. Because of the large data, only fire test one results are discussed in this paper.

Purpose – The purpose of this study is to investigate the influencing factors on the charring behaviour of timber, the char layer and the charring depth in non-standard fires. Design/methodology/approach – This paper summarizes outcomes of tests, investigating the influences on the charring behavior of timber by varying the oxygen content and the gas velocity in the compartment. Results show that charring is depending on the fire compartment temperature, but results show further that at higher oxygen flow, char contraction was observed affecting the protective function of the char layer. Findings – In particular, in the cooling phase, char contraction should be considered which may have a significant impact on performance-based design using non-standard temperature fire curves where the complete fire history including the cooling phase has to be taken into account. Originality/value – Up to now, some research on non-standard fire exposed timber member has been performed, mainly based on standard fire resistance tests where boundary conditions as gas flow and oxygen content especially in the decay phase are not measured or documented. The approach presented in this paper is the first documented fire tests with timber documenting the data required.