The unstable oscillatory behaviour, with frequency in the order of few mHz, that has been occasionally observed in mechanically-ventilated compartment fires, is studied experimentally and numerically. First, a series of experiments using a small-scale compartment have been conducted using heptane and dodecane as fuels. Results show that unstable and stable combustion regimes can occur depending on fuel type, pool size, air renewal rate of the compartment (ARR), and ventilation conditions. For a certain range of these factors, unstable low-frequency (LF) oscillatory combustion, accompanied by thermodynamic pressure and ventilation flow rate variations and displacement of the flame outside the pan, is observed. The occurrence and persistency of LF oscillations result from the competition between oxygen supply and fuel vapor supply due to the heat feedback from the flame and enclosure to the fuel tray. Whatever the fuel type, it is found that i) the range of ARR where LF oscillations appear and the oscillation amplitude increase with the pool size, and ii) the frequency increases, while amplitude decreases, with increasing ARR, independently of the pool size. It is also found that the more flammable the fuel, i) the smaller pool size for which LF oscillations appear and the higher the frequency for the same ventilation conditions, and ii) the wider the range of ARR where LF oscillations appear for a given pool size. The effects of air inlet position and blowing direction on the oscillations properties is also investigated. Second, predictive CFD simulations have been performed using the in-house SAFIR software. Although SAFIR does not correctly describe the displacement of the flame outside the fuel pan, it satisfactorily reproduces the LF oscillatory fire behaviour, especially its dominant frequency. Information about inaccessible or difficult-to-measure local quantities, such as the local evaporation rate, temperature and heat flux at the liquid surface, and species concentrations, are provided from the numerical simulation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.