Full scale experiments for evaluating theoretical fire wall models

Abstract

AbstractThe aim of the research described in this paper was to provide experimental results for the evaluation of theoretical models for predicting the behaviour and time‐to‐failure of loadbearing and non‐loadbearing wood framed walls in fire. References for thermal and mechanical properties of wood and gypsum board are given to provide comprehensive input for the evaluation of theoretical wall models. The scope of the research involved full‐scale uninsulated cavity walls with well‐controlled clearly known conditions including initial ambient vertical load capacity for benchmarking the reduction in capacity and stiffness, rotational stiffness of supports, eccentricity of vertical load, elastic moduli of wood and gypsum board in compression, stiffness of slip between gypsum board and studs and end stud effects. The experiments were repeated and they demonstrated that the controls led to high consistency in the results despite the inherent large variability of the mechanical properties of wood. The results include temperature distributions, initial vertical load capacity, load‐deflection plots and times‐to‐failure. The results show that the temperatures in the studs are approximately uniform until all the moisture is vaporized. Thermal properties of wood will not vary significantly for consistent density, moisture content and species of wood. The main structural actions that should be modelled for different loading regimes are deduced. Copyright © 2004 John Wiley & Sons, Ltd.