Fire field modelling (Computation Fluid Dynamics) has become more and more attractive as a critical design tool to meet Performance-based fire design on advanced modern buildings. This paper describes the application of Computational Fluid Dynamics (CFD) to predict velocities and temperature distributions induced by a fire in a Steckler’s experimental data [1]. The experimental data of different fire loads is taken as case study for present investigation. The experiments of Steckler’s compartment fire were conducted to investigate fire-induced flows through the opening in a compartment of size 2.8 m × 2.8 m × 2.18 m (height). The compartment has a doorway opening of 0.74 m × 1.83 m to account the ventilation condition. A porous gas burner is flushed at the floor in the centre of the room with the diameter of 0.3m in the compartment. With the above experimental data, simulation studies were performed with combustion modelling using commercial code of ANSYS CFX-5. The comparison of simulation results of fire field models with experimental domain for different strengths of fire 31.6, 62.9, 105.3 and 158.0 kW is reported. The boundary conditions of the simulation are kept constant, only fire strength is changed to see the performance of the CFD tool. The door centreline temperature, velocities and room corner temperatures are predicted and compared with experimental data as well as with FDS. The results are in good agreement with the experimental data.

View full abstract

With regard to reaction-to-fire tests for building materials in Japan, ISO 5660-1 (small-scale test cone calorimeter) is de facto the only method for evaluation, according to the current building standard law of Japan [1], which actually also designates ISO/TS 17431 (intermediate-scale test) as analternative not being implemented very often, however, bythe industry. It is noted that it is impossible to predict the fire performance of sandwich panels when they are actually used in real buildings only from small scale tests such as ISO 5660-1. This is not a deficiency regarding the ISO 5660-1 as a test method but it is difficult to use the small scale results on a horizontal surface (100 mm by 100 mm) in order to predict the fire performance of sandwich panels in real applications.. The reason is that in actual building fires, both ceilings and walls made of sandwich panels are heated from various directions and weak points are the joints and seals which can never be evaluated with a small-scale test. Therefore, in this study, the authors firstly modified ISO/TS 17431 model box test with free-standing specimens, referring to ISO 13784-1, and different types of sandwich panels were chosen to be the specimens, and the results are discussed comparing with ISO 5660-1 results.

View full abstract