To better evaluate failure of building elements and spread of fire beyond the room of fire origin, this thesis developed and integrated into CUrisk a barrier failure model and a fire spread model.
The role and position of the Fire Spread submodel were analyzed and changes to the system model and some other submodels were undertaken. With these modifications, CUrisk can employ the Fire Spread submodel to predict the fire hazard conditions in a building fire, and to use the results to predict the life risk and fire damages. Through a comprehensive case study of fire risk assessment of a six-storey residential building using the improved CUrisk, the Fire Spread submodel demonstrated the impacts of fire spread level on building occupant safety and fire losses.
The model performance was verified by comparing with the fire test measurements, which demonstrated good agreements. Comparable results are also predicted regarding the fall-off behaviour of the fire-exposed gypsum board as well as the charring behaviour. In addition, an example calculation was made using the probabilistic barrier failure model. Finally, a fire risk analysis case study was conducted on a six-storey apartment building with the purpose of showing the effect of wall barriers on fire risk. Results indicated that CUrisk can evaluate the impact of fire barriers on the fire risk with the new Barrie Failure submodel.
This paper documents the findings of a series of full-scale room fire tests, which includes tests on fully protected, partially protected CLT rooms as well as light-frame timber/steel rooms under real natural fires, aiming to investigate the fire behaviour and performance of CLT panels as an increasingly popular engineered wood product and to compare it to the performance of more traditional construction methods. Results show that the CLT panels when left unprotected get involved in the room fire as part of the combustible contents, responsible for over 60% of total heat release in the fully unprotected CLT room and double the heat release rate of a fully protected room fire where the CLT does not contribute. Partially-protected CLT rooms also demonstrates various levels of fire contribution. The amount of CLT exposure is also related to the occurrence of re-ignition and a second flashover after all the movable fuels are consumed. The behaviour of CLT delamination and charring as well as the performance of gypsum boards in fire are also discussed.
As timber buildings are constructed taller, architects and building owners are asking for more timber to be exposed. Addressing how exposed timber and in particular cross laminated timber, influences a fully developed fire through to self-extinguishment is a current and complex fire safety issue. There is limited research available on how exposed timber alters heat release rate, temperatures and fire duration. This paper provides a summary of the relevant research to understand similarities in findings and how the results of fire tests can be applied. Research shows that large areas of exposed timber has a significant impact on heat release rate, but limited areas of exposed timber can be accommodated within a fire safe design. The location of exposed timber and avoiding two or more adjacent exposed surfaces, is an important finding. It is evident from the limited testing that a single exposed timber wall of approximately 20% of the total wall area has little impact on a compartment fire. The development of a calculation methodology to account for the change in compartment fire dynamics when two or more surfaces are exposed is the next step in the advancement of exposed timber fire safety engineering.