Fire Whirl Dynamics example essay topic

Introduction: Fire whirls are a typically rare but potentially catastrophic form of fire. They are observed during urban and forest fires, where fire "tornadoes" are characterized by large-scale whirling flames which rise in 2 to 360 m diameter vortices from 10 to 1200 m high. These fire whirls accelerate combustion, produce significant suction pressures and lifting forces, and can carry burning debris, logs and even buildings thousands of meters from the main fire. During the Great Chicago Fire in 1871 fire whirls carried burning debris up to 3/8 mile beyond fire lines rapidly extending the fire out of control.

In the same year in Peshtigo, Wisconsin, the largest and deadliest fire in US history fire tornadoes were reported to lift roofs off buildings and resulted in the death of 1,500 people. In the 1923 Kanto Earthquake fire in Tokyo 38,000 people were killed in 15 minutes when a huge fire whirl descended upon refugees sheltering in an open area adjacent to the urban fire. Recently, CFD calculations performed by the author of atria fires inside a proposed building produced very energetic fire whirls 5 m to 20 m diameter and 40 m tall which roared from one end of the 45 m open space to the other and back in less than a minute (Banks, Chang & Meroney, 2002). Unfortunately, as building atria get larger, attempts to control ventilation during fires in atria may introduce vorticity, which can thus generate "internal" fire whirls.

Fire Whirl Dynamics: The formation of fire whirls require a source of ambient vorticity, a concentrating mechanism, and a favorable environment for fire whirl stability and growth (augmentation physics). Ambient vorticity can be produced by ground level boundary layers generated by the wind, wind shear from nonuniform horizontal densities, the earth's rotation, nonhomogeneous surface heating, or wind shear produced as air passes over a ridge or hill. Concentrating mechanisms include rising air in a buoyant column from unstable layers forming over sun-heated ground, the presence of a storm front, or hot gases from a fire. The concentrating mechanisms rotate the horizontal vorticity into the vertical and stretch the vortex tubes. Through conservation of angular momentum the stretched tubes induce more rapid rotation resulting in lower axial pressures, which in turn encourages further entrainment of ground level vortex-rich air.

Finally, the rotational structure of the vortex induces centrifugal forces which dampen turbulence near the vortex core; thus, reducing any tendency for the fire whirl plume to diffuse outward from the core. Laboratory Fire Whirls: Byram and Martin (1962) used external vertical cylinders with tangential slots oriented to produce rotating flow about a fire source. They examined two sets of equipment of diameters and heights, 33 and 183 cm, or 66 and 335 cm, respectively. Burning alcohol pools within their apparatus, they reported visible fire whirls up to 300 cm tall with inner fire tube columns 2 cm in diameter. They observed horizontal velocities at the surface of the inner column of about 9 m / sec (~6000 rpm) and vertical velocities to 18 m / sec.

Emmons and Ying (1966) used a rotating-screen apparatus to systematically evaluate the effects of angular rotation (Ross by number) and plume buoyancy (Froude number) on fire whirl dynamics. They reported that turbulent mixing coefficient decreases with increasing angular momentum, and increases with elevation above the ground. Later Chigier et al. (1970) reproduced their apparatus, but they used a turbulent jet diffusion flame rather than a fuel pan. More recently Satoh and Yang (1996, 1997) produced laboratory scale fire whirls by adjusting symmetrical vertical gaps separating the square vertical bounding walls surrounding a central fire pan. They examined the effect of gap size, wall height, fuel size, and heat load on the fire whirl.

They determined that there is a critical gap size which is not so large or small that it inhibits the entrainment of air needed to sustain the fire. Stable whirls were generally associated with flame heights smaller than the wall height of the square enclosure. Flame temperatures were primarily affected by the magnitude of the volumetric heat source. Large scale fire whirl simulations have been produced for video and movie effects by combining shrouded helicopter blades and ancillary fans to produce vortices 12 m (40 ft) high and core diameters of 30 cm (1 ft) by Reel Efx (1995) for car commercials and adventure movies (Volvo-850 commercial (1995) and Twister).

Simulating Fire Whirls by CFD: Mur gai and Emmons (1960) and Emmons and Ying (1966) describe integral plume models which are calibrated with experimental data. Satoh and Yang (1997) used the Notre Dame UNSAFE code with associated 3d, compressible, buoyant, and constant turbulent viscosity specifications. Ten cases were considered which included validation exercises and parameter sensitivity studies. Battaglia et al. (2000) simulated the laboratory experiments of Emmons and Ying (1966), Chigier et al. (1970), and Satoh and Yang (1997), which included cases for fixed circulation and variable fire strength, fixed fire strength and variable circulation, and jointly varied fire strength and circulation.

The numerical code used was the NIST shareware FDS (Fire Dynamics Simulator) which includes 3d, compressible, buoyant and LES turbulent models (Baum et al., 1996). This author has reproduced cases from Emmons and Ying (1966) and Satoh and Yang (1997) with both Fluent 6.0 and FDS. Finally, several case studies were examined for fire whirls growing within hypothetical building atrium's. The validation runs were considered satisfactory; hence, case study results should be trustworthy. Building fire whirls were found to quickly form when cross-ventilation produced horizontal wind shear about the base of a fire.

Bibliography

Banks, D., Chang, C.H. and Meroney, R.N. (2002), "FD Simulation of the Progress of Fires in Building Atria", Session on Computational Evaluation of Wind Effects on Buildings, Proceedings of ASCE 2002 Structures Congress, Denver, CO, April 4-6, 2002, 2 pp.
Battaglia, F., McGrattan, K.B., Rehm, R.G., and Baum, H.R. (2000), "Simulating Fie Whirls", Combustion Theory Modelling, Vol.
4: 123-138 Baum, H.R., McGrattan, K.B., and Rehm, R.G. (1996), "Three Dimensional Simulations of Fire Plume Dynamics", Fire Safety Science- Proceedings of the 5th Int.
Symposium, pp. 511-522. Byram, G.M. and Martin, R.E. (1962), "Fire Whirlwinds in the Laboratory", Fire Control Notes, Vol.
33 (1): 13-17. Chigier, N.A., Beer, J.M., Greco v, D. and Bassin dale, K. (1970), "Jet Flames in Rotating Flow Fields", Combust.
Flame, Vol. 14: 171-180. Emmons, H.W. and Ying, S.J. (1966), "The Fire Whirl", 11th International Combustion Symposium, pp.
475-488. Reel Efx (1995), Simulation of Fire Tube, North Hollywood, CA, web web, October 2002.
Satoh, K. and Yang K.T. (1996), "Experimental Observations of Swirling Fires", Proceedings of ASME Heat Transfer Division, HTD-Vol.
335, Vol. 4: 393-400. Satoh, K. and Yang, K.T. (1997), "Simulations of Swirling Fires Controlled by Channeled Self-generated Entrainment Flows", Fire Safety Science- Proceedings of the 5th Int.