Abstract

Flame necking-in is a fundamental behavior at the base of diffusive pool fires caused by the entrained flow approaching the flame induced by buoyancy (density difference), which is also the major generation source of flame periodic oscillatory instability. However, measurements have not been previously reported for the evolution of the flame necking-in dynamic characteristics and how they relate to flame instability behaviors. This paper quantifies the flame necking-in dynamics and instability motions in 0.04–0.25m diameter ethanol pool fires with different lip heights (0.3–2cm). Based on direct time-sequence analysis on flame photographs, the necking-in characteristic maximum depth (Dmax), average necking-in velocity (Unecking-in), maximum uprising height (Hmax), average uprising velocity (Uuprising), as well as the vortices shedding instability frequency (f) and characteristic vortices formation lift-time (τ) are quantified to find their evolutions with pool size and lip height along with their associated dominant instability motions. Three different flame instability motions are identified: short life Rayleigh–Taylor (R–T) instability, extended R–T instability and puffing instability. The dominant instability motion is found to transit from extended R–T instability to puffing instability with increase in pool size or lip height. The pumping capacity of large-scale vortices formation (which could be quantified by Dmax, Hmax, Unecking-in, Uuprising) is primarily associated with the extended R–T instability frequency. The frequency (f) of the necking-in extended R–T instability is found to be greater than the puffing frequency. The puffing frequency increases slightly with lip height and decreases with pool diameter, D, following the well-known pool fire square root law (f∼D−1/2, or non-dimensionally St∼Fr−1/2). Meanwhile the frequency (f) for Extended R–T instability is found to be well correlated by fR–T,extended∼(ℓfg2/Q̇)1/3. The characteristic life-times (τ) of the extended R–T instabilities increase with pool diameter while remaining smaller than the puffing life-times that scale by τ∼D1/2.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call