Elsevier

Combustion and Flame

Volume 213, March 2020, Pages 107-116
Combustion and Flame

Experimental investigation of compartment fires with circular opening: From the aspects of internal temperature and facade flame

https://doi.org/10.1016/j.combustflame.2019.11.027Get rights and content

Abstract

This paper experimentally investigated the internal temperature and the facade flame for compartment fires involving circular openings. A series of experiments were carried out by using a reduced-scale model which composed of a cube compartment with various circular openings and a vertical fireproofing board (used as building facade). A porous burner with propane fuel was set inside the compartment as fire source. The internal temperature profile was measured by vertical thermocouple arrays at the inner and outer corners, and the facade flame was captured by a CCD camera from a side view. This paper revealed that: (1) With the same area, maximum heat release rate inside the compartment with circular opening is larger than that with square opening. (2) For relatively larger circular openings, facade flames tend to ceiling jet flame extension. With the increasing of heat release rate, facade flame induced by ceiling jet flame extension will convert to spill fire plume. (3) A dimensionless number R, which represents the ratio of total heat release rate to internal maximum heat release rate, is proposed to distinguish spill fire plume and ceiling jet flame extension when facade flames appear at exterior. (4) A non-dimensional correlation is developed to characterize the facade flame heights for both ceiling jet flame extension and spill fire plume. These findings and results can provide fundamental knowledge for understanding compartment fire with a circular opening.

Introduction

In high rise buildings fires, flame ejection usually occurs through windows when a compartment fire reaches a fully developed state after the phase of ceiling jet [1,2]. Fire may propagate to upper floors if ejected flames extend along the facade wall of building to ignite the floors above the fire compartment, which will cause a serious risk to urban safety and result in a catastrophic loss of life and property. Understanding the characteristics of flame ejection is extremely important as the facade flames can alert occupants to escape rapidly by triggering the sprinkler and the alarm system.

Yokoi [3] started to examine this phenomenon in 1960 and proposed a length scale r0 to characterize the temperature distribution of ejected thermal plume. After that, a significant number of researches have been conducted to analyze and quantify the characteristics of spill fire plume for various window dimensions without and with different external conditions. These characteristics include facade flame height [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], temperature profile outside and inside enclosure [6,7,10,11,[14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], heat flux intensity [10,11,17,22,25], internal combustion [17], [18], [19] as well as flame ejected behavior [10,23,26] and so on.

Related researches which focused on facade flame height have attracted more and more interests in the recent year. Webster et al. [4,5] and Seigel [6] studied facade flame height from compartment fire and revealed its evolution. Tang et al. [7] and Oleszkiewicz [8], respectively conducted a series of reduced- and full-scale experiments to discuss and characterize facade flame height with various dimensions and heat release rates. Later, Hu et al. [12,13] took external wind into account and developed non-dimensional correlations to predict facade flame height. Particularly, Delichatsios [10] and Lee [11] proposed characteristic length scales ℓ1(1=(AH)5/2) and ℓ2(2=(AH2)1/4) in relation to windows dimensions to describe the characteristics of spill fire plume. Then, a following non-dimensional equation was established to predict the mean flame height from ventilation-controlled compartment fires [10,11]:Zf1=f(Q˙ex*)=f(Q˙exρcpTg15/2)in whichQ˙ex=Q˙Q˙inside=Q˙1500AHwhere Zf is mean flame height obtained when the intermittency is 0.5, ℓ1 is characteristic length for rectangular opening, A is area of opening, H is height of opening, ρ is ambient air density, cP is specific heat of ambient air at constant pressure, T is ambient temperature, g is acceleration of gravity, Q˙ex is excess heat release rate by reason of excess unburned fuel spilled outside, Q˙ is total heat release rate of fuel supplied by the gas burner and Q˙inside is the heat release rate of combustion inside the compartment.

Meanwhile, both the external- and internal temperatures [6,7,10,11,[14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24] at diverse conditions have been analyzed. Based on Yokoi's research [3] and combining virtual origin theory [27], Tang et al. [7] modified a dimensionless correlation of facade vertical temperature utilizing the characteristic length ℓ1 for spill fire plume under ventilation-controlled condition. For the gas temperature inside the compartment, Delichatsios [16] proposed a predicting formula for the growing period of enclosure fire. McCafreey and Quintiere [15] developed a simple method to estimate the gas temperature and predict the likelihood of flashover in a fire room by energy balance theory. Moreover, the two-layer zone models were established [17,18] and used for estimating the gas temperature profile inside fire compartment in engineering. Considering the effect of external wind, Hu et al. [19] revealed the internal temperature evolution with compartment fire growth from stratified phase to well-mixed phase. Additionally, Hu et al. [26] quantitatively studied the critical condition of flame ejection and the intermittent behavior of ejected flame for compartment fire from over-ventilated condition to under-ventilated condition. They found that the intermittency of flame ejection depends on the heat release rate, ventilation geometry and heat loss.

It can be noticed that the above-mentioned researches mainly concentrated on compartment fire with a rectangular opening. However, there are plenty of buildings with circular windows (as Fig. 1 shows) in urban area with the development of modern building industry. Unfortunately, little attention has been focused on these non-rectangular opening building fires, and the basic knowledge about compartment fire of circular window is still deficient. In the meantime, the classical theory of spill fire plume, which concentrated on the investigation of flame ejection involving rectangular opening, may not be applicable to compartment fire with circular opening. Therefore, it is of significant necessity to carry out investigations on compartment fire with circular opening for providing supplementary knowledge over previous spill fire plume researches.

Meanwhile, there is another burning phenomenon called ceiling jet flame extension [1] in compartment fires. Aside from the spill fire plume, ceiling jet flame extension also can cause the appearance of facade flame for an over-ventilated compartment fire as reported in Refs. [28,29]. Ceiling jet flame extension is also likely to ignite external combustible and induce fire propagation at exterior wall. Ohmiya et al. [28] reported the data of facade flames driven by both ceiling jet flame extension and spill fire plume. Hurley [29] pointed out that some differences existed between ceiling jet flame extension and spill fire plume. However, there is still little work concerning facade flame caused by ceiling jet flame extension. Knowledge about the relation between spill fire plume and ceiling jet flame extension is so limited that it desiderates to be supplemented.

In this paper, a series of experiments were conducted to investigate the characteristics of compartment fires with circular opening. Temperature evolution inside the compartment was measured by thermocouples and the facade flame was recorded by a CCD camera. A qualitative analysis of internal temperature was performed, and a non-dimensional number was proposed to distinguish facade flame generated by spill fire plume and ceiling jet flame extension. The empirical equation of facade flame height for both spill fire plume and ceiling jet flame extension was extended to the circular opening condition based on scaling analysis of the experimental data. These results may help investigators and engineers to correctly understand and further properly handle compartment fires with circular openings.

Section snippets

Experimental setup

Figure 2 shows the schematic diagram of the experimental apparatus used in this study. A reduced-scale experimental model, consisting of a fire compartment and a vertical board attached to the compartment window, was applied to investigate compartment fires with different circular openings. The fire compartment is a 0.4 m cube and is inner-lined with a 0.03 m ceramic fiber board for thermal insulation. The ceramic fiber board has a density of 285 kg/m3, a thermal conductivity of 0.18 W/(m · K)

Internal temperature of compartment fire

To reveal the law of compartment fire growth, Hu et al. [19] experimentally investigated the transition from a stratified phase to a well-mixed phase by analyzing the gas temperature evolution inside a fire compartment with a rectangular opening. They established a mathematical model to estimate the critical heat release rate for reaching the well-mixed phase as the following:Q˙wellmixed1500AH=0.47UwgH+1.7where Q˙wellmixed is critical heat release rate for reaching the well-mixed phase, Uw

Conclusions

As the classical models established for rectangular opening could not be applied to the cases of circular opening directly, this paper investigated internal temperature and facade flame for compartment fire with circular opening. Reduced-scale experiments were carried out to obtain experimental results for various circular openings and different heat release rates. Major results are summarized as follows:

  • (1)

    With the same area, maximum heat release rate inside the fire compartment with circular

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by National Natural Science Foundation of China (51506032), National Key Research and Development Program of China (2018YFB1501500) and Natural Science Foundation of Guangdong Province (2019A1515011846).

References (38)

Cited by (38)

View all citing articles on Scopus
View full text