Fire dynamics inside a large and open‐plan compartment with exposed timber ceiling and columns: CodeRed #01

Abstract

SummaryThere is an increasing global demand to build from timber as it is a sustainable and attractive material. One of the key challenges associated with timber buildings is their performance in a fire, in particular, for medium‐ and high‐rise buildings and when timber is exposed. Research on this topic to date has been performed in compartments smaller than 84 m2 which does not capture the fire dynamics of large compartments. This paper presents the first in a series of experiments carried out inside a large, purpose‐built, open‐plan compartment with a floor area of 352 m2. The large‐scale compartment had a fully exposed, unloaded, cross‐laminated timber (CLT) ceiling and glued laminated timber (glulam) columns, made with adhesives that have been tested to not exhibit char fall‐off in fire. At 352 m2 floor area, this is currently the largest compartment fire experiment carried out globally. The compartment characteristics and the arrangement of the wood crib mimicked the previous series of experiments in Poland in a non‐combustible compartment by Rackauskaite et al. (2021). The experiment was instrumented with thermocouples measuring the gas temperature in the compartment, and above two of the openings, as well as the temperature within the timber members. Plate thermometers were also included to measure the heat flux to the ceiling and of the external flaming. Additionally, fixed and drone‐mounted cameras captured the fire. A protected steel column was added to calculate the equivalent fire exposure when compared to a standard fire resistance furnace test. The experiment was allowed to burn out without fire‐fighting intervention. Continuous measurements were taken for 48 h. The wood crib was ignited at one end of the compartment. Flames rapidly spread across the timber ceiling to the other end of the compartment with the rest of the wood crib being involved in the fire at 5 min 36 s from ignition. The total duration of the fire was 22 min 30 s, at which time visible flaming ceased. Significant external flaming was observed through the openings in the order of 3 m in height but varied with position as the fire travelled across the compartment. Analysis of the experimental results and camera footage showed that the presence of the timber structure approximately doubled the heat release rate (HRR) compared to the value expected from the wood crib alone. Smouldering combustion of the timber members continued for hours after the cessation of flaming, burning in several hotspots, resulting in holes through the CLT slab. This paper improves the understanding fire dynamics for open‐plan, exposed timber compartments and has shown that issues such as external flaming, smouldering and the increase in HRR need to be considered when deriving practical design solutions.