《sfpe handbook of Fire Protection Engineering：Section Two：Fire Dynamics.pdf》由会员分享，可在线阅读，更多相关《sfpe handbook of Fire Protection Engineering：Section Two：Fire Dynamics.pdf（320页珍藏版）》请在三一文库上搜索。
1、SFPE Handbook of Fire Protection Engineering Third Edition Editorial Staff Philip J. DiNenno, P.E. (Hughes Associates, Inc.), Editor-in-Chief Dougal Drysdale, PhD. (University of Edinburgh), Section 1 Craig L. Beyler, PhD. (Hughes Associates, Inc.), Section 2 W. Douglas Walton, P.E. (National Instit
2、ute of Standards and Technology), Section 3 Richard L. P. Custer (Arup Fire USA), Section 4 John R. Hall, Jr., PhD. (National Fire Protection Association), Section 5 John M. Watts, Jr., PhD. (The Fire Safety Institute), Section 5 National Fire Protection Association Quincy, Massachusetts Society of
3、Fire Protection Engineers Bethesda, Maryland FM.QXD 3/3/2003 4:26 PM Page iii Section Two Fire Dynamics 2-OPENER.QXD 11/14/2001 12:02 PM Page a Chapter 21Fire Plumes, Flame Height, and Air Entrainment Introduction2-1 Fire Plume Features2-1 Calculation Methods2-2 Plumes in Temperature-Stratified Ambi
4、ents2-8 Illustration2-13 Additional Flame Topics2-14 Data Sources2-15 Nomenclature2-15 References Cited2-16 Chapter 22Ceiling Jet Flows Introduction2-18 Steady Fires2-18 Convective Heat Transfer to the Ceiling2-22 Sloped Ceilings2-23 Time-Dependent Fires2-23 Confined Ceilings2-25 Ceiling Jet Develop
5、ment2-28 Summary2-29 Nomenclature2-29 References Cited2-30 Chapter 23Vent Flows Introduction2-32 Calculation Methods for Nonbuoyant Flows2-32 Vents as Part of the Building Flow Network2-41 Nomenclature2-41 References Cited2-41 Chapter 24Visibility and Human Behavior in Fire Smoke Background2-42 Visi
6、bility in Fire Smoke2-42 Human Behavior in Fire Smoke2-46 Intensive System for Escape Guidance2-49 Conclusion2-52 References Cited2-52 Chapter 25Effect of Combustion Conditions on Species Production Introduction2-54 Basic Concepts2-55 Species Production within Fire Compartments2-59 Fire Plume Effect
7、s2-69 Transient Conditions2-70 Species Transport to Adjacent Spaces2-71 Engineering Methodology2-77 Nomenclature2-81 References Cited2-82 Chapter 26Toxicity Assessment of Combustion Products Introduction2-83 Dose/Response Relationships and Dose Estimation in the Evaluation of Toxicity2-86 Allowance
8、for Margins of Safety and Variations in Susceptibility of Human Populations2-89 Fractional Effective Dose Hazard Assessments and Toxic Potency2-90 Asphyxiation by Fire Gases and Prediction of Time to Incapacitation2-99 Irritant Fire Products2-111 Chemical Composition and Toxicity of Combustion Produ
9、ct Atmospheres2-122 Smoke2-124 The Exposure of Fire Victims to Heat2-125 Worked Example of a Simplified Life Threat Hazard Analysis2-132 Fire Scenarios and Victim Incapacitation2-133 The Use of Small-Scale Combustion Product Toxicity Tests for Estimating Toxic Potency and Toxic Hazard in Fires2-144
10、The Conduct and Application of Small-Scale Tests in the Assessment of Toxicity and Toxic Hazard2-158 Summary of Toxic and Physical Hazard Assessment Model2-159 Appendix 26A2-164 Appendix 26B2-165 Appendix 26C2-165 References Cited2-168 Additional Reading2-171 Chapter 27Flammability Limits of Premixe
11、d and Diffusion Flames Introduction2-172 Premixed Combustion2-172 Diffusion Flame Limits2-183 Nomenclature2-186 References Cited2-187 Chapter 28Ignition of Liquid Fuels Introduction2-188 Vaporization: AContrast between Liquid and Solid Combustibles2-188 Mixing of Vapors with Air2-189 Ignition of the
12、 Mixture2-189 Some Experimental Techniques and Definitions2-189 Example Data2-190 Theory and Discussion2-191 Concluding Remarks2-198 Nomenclature2-198 References Cited2-199 Chapter 29Smoldering Combustion Introduction2-200 Self-Sustained Smolder Propagation2-201 Conclusion2-209 References Cited2-209
13、 Chapter 210Spontaneous Combustion and Self-Heating Introduction2-211 The Literature2-213 The Concept of Criticality2-213 The Semenov (Well-Stirred) Theory of Thermal Ignition2-215 Extension to Complex Chemistry and CSTRs2-218 The Frank-Kamenetskii Theory of Criticality2-219 Experimental Testing Met
14、hods2-220 Special Cases Requiring Correction2-221 Finite Biot Number2-222 Times to Ignition (Induction Periods)2-223 Investigation of Cause of Possible Spontaneous Ignition Fires2-224 The Aftermath2-225 Case Histories and Examples2-225 Nomenclature2-227 References Cited2-227 Section 2 Fire Dynamics
15、2-OPENER.QXD 11/14/2001 12:02 PM Page b Chapter 211Flaming Ignition of Solid Fuels Introduction2-229 The Process of Ignition2-229 Conduction-Controlled Spontaneous Ignition of Cellulose Due to Radiant HeatingMartins Map2-230 AQualitative Description2-231 Conservation Equations2-233 Ignition Criteria
16、2-234 Solid Conduction-Controlled Ignition2-235 Role of the Gas Phase Processes2-239 Some Practical Issues2-241 APractical Illustration2-242 Conclusion2-243 Explicit Forms of Equation 14 for Some Limiting Cases2-243 Nomenclature2-244 References Cited2-245 Chapter 212Surface Flame Spread Introduction
17、2-246 Background2-247 Flame Spread over Solids2-247 Flame Spread over Liquids2-254 Flame Spread in Forests2-255 Flame Spread in Microgravity2-256 Concluding Remarks2-256 Nomenclature2-256 References Cited2-256 Chapter 213Smoke Production and Properties Introduction2-258 Smoke Production2-258 Size Di
18、stribution2-259 Smoke Properties2-263 Nomenclature2-268 References Cited2-268 Chapter 214Heat Fluxes from Fires to Surfaces Introduction2-269 General Topics2-269 Exposure Fires2-270 Burning Walls and Ceilings2-281 Exposure Fires and Burning Walls and Ceilings2-291 Fires from Windows2-292 Effects of
19、Other Variables2-293 Nomenclature2-294 References Cited2-294 Chapter 215Liquid Fuel Fires Introduction2-297 Spill or Pool Size2-297 Fire Growth Rate2-300 Fire Size2-308 Nomenclature2-315 References Cited2-315 2-OPENER.QXD 11/14/2001 12:02 PM Page c 21 Introduction Practically all fires go through an
20、 important, initial stage in which a coherent, buoyant gas stream rises above a localized volume undergoing combustion into surround- ing space of essentially uncontaminated air. This stage be- gins at ignition, continues through a possible smoldering interval, into a flaming interval, and may be sa
21、id to end pri- or to flashover. The buoyant gas stream is generally turbu- lent, except when the fire source is very small. The buoyant flow, including any flames, is referred to as a fire plume. Combustion may be the result of pyrolysis of solid ma- terials or evaporation of liquids because of heat
22、 feedback from the combustion volume, or of pressurized release of flammable gas. Other combustion situations may involve discharge of liquid sprays and aerosols, both liquid and solid, but these will not be discussed here. In the case of high-pressure releases, the momentum of the release may be im
23、portant. Flames in these situations are usually re- ferred to as diffusion flames, being the result of combustible vapor or gas mixing or diffusing into an ambient oxidant, usually air, as opposed to being premixed with an oxidant. The properties of fire plumes are important in deal- ing with proble
24、ms related to fire detection, fire heating of building structures, smoke filling rates, fire venting, and so forth. They can also be important in fire suppression system design. This chapter deals with axisymmetric, turbulent fire plumes and reviews some relations for predicting the properties of su
25、ch plumes. It is assumed throughout the chapter that the surrounding air is uncontaminated by fire products and that it is uniform in temperature, except where specifically treated as temperature stratified. Re- lease of gas from a pressurized source is assumed to be vertical. The relations cease to
26、 be valid at elevations where the plume enters a smoke layer. Main topics are flame heights, plume temperatures and velocities, virtual origin, air entrainment, and effects of ambient temperature stratifications. At the end of the chapter, a few additional aspects of diffusion flames are touched on
27、briefly, including flame pulsations, wall/cor- ner effects, and wind effects. Fire Plume Features Figure 2-1.1 shows a schematic representation of a tur- bulent fire plume originating at a flaming source, which may be solid or liquid. Volatiles driven off from the com- bustible, by heat fed back fro
28、m the fire mix with the sur- rounding air and form a diffusion flame. Laboratory simulations often employ controlled release of flammable gas through a horizontal, porous surface. The mean height of the flame is L.Surrounding the flame and extending up- ward is a boundary (broken lines) that confine
29、s the entire buoyant flow of combustion products and entrained air. The air is entrained across this boundary, which instan- taneously is very sharp, highly convoluted, and easily discernible in smoky fires. The flow profile could be the time-averaged temperature rise above the ambient SECTION TWO C
30、HAPTER 1 Fire Plumes, Flame Height, and Air Entrainment Gunnar Heskestad Dr. Gunnar Heskestad is principal research scientist at Factory Mu- tual Research, specializing in fluid mechanics and heat transfer of fire, with applications to fire protection. Flame Entrained flowFlow profile ZZ 0 T0u0 T0;u
31、0 L Figure 2-1.1.Features of a turbulent fire plume, includ- ing axial variations on the centerline of mean excess temperature,g gT0, and mean velocity, u0.34 02-01.QXD 11/14/2001 10:56 AM Page 1 temperature, or the concentration of a gas (such as CO2) generated by the fire, or the axial velocity in
32、 the fire plume. Figure 2-1.1 suggests qualitatively, based on experi- mental observations,15how the temperature rise on the centerline, !T0,and the velocity on the centerline, u0,might behave along the plume axis. In this example of a relatively tall flame, the temperatures are nearly constant in t
33、he lower portion of the flame. Temperatures begin to decay in the in- termittent, upper portion of the flame as the combustion re- actions trail off and air entrained from the surroundings cools the flow. The centerline velocities, u0,tend to have their maxima slightly below the mean flame height an
34、d al- ways decay toward higher elevations. If the combustible is porous and supports internal combustion, there may not be as pronounced a falloff in the gas velocity toward the top of the combustible, as suggested in Figure 2-1.1.3 The total heat release rate of a fire source, Q g, is either convec
35、ted, Qc g ,or radiated, Qr g ,away from the combustion region. In a fire deep in a porous combustible pile (e.g., a stack of wood pallets), some of the total heat generated is trapped by and stored in the not yet burning material; the rest escapes from the combustible array as either convec- tive or
36、 radiative energy flux. If most of the volatiles re- leased undergo combustion above the fuel array, as in pool fires of liquids and other horizontal-surface fires, and even in well-developed porous pile fires, then the convective fraction of the total heat release rate is rarely measured at less th
37、an 60 to 70 percent of the total heat re- lease rate.6,7The convective flux, Qc g ,is carried away by the plume above the flames, while the remainder of the total heat liberated, Qr g ,is radiated away in all directions. The total heat release rate, Q g, is often assumed to be equal to the theoretic
38、al heat release rate, which is based on complete combustion of the burning material. The theo- retical heat release rate in kW is evaluated as the mass burning rate in kg/s multiplied by the lower heat of com- plete combustion in kJ/kg. The ratio of the total heat re- lease rate to the theoretical h
39、eat release rate, which is the combustion efficiency, is indeed close to unity for some fire sources (e.g., methanol and heptane pools),6but may deviate significantly from unity for others (e.g., a poly- styrene fire, for which the combustion efficiency is about 45 percent,7and a fully involved stac
40、k of wood pallets, for which the combustion efficiency is 63 percent6). Calculation Methods Flame Heights The visible flames above a fire source contain the combustion reactions. Tamanini8has investigated the manner in which combustion approaches completion with respect to height in diffusion flames
41、. Typically, the luminosity of the lower part of the flaming region appears fairly steady, while the upper part appears to be intermittent. Sometimes vortex structures, more or less pronounced, can be observed to form near the base of the flame and shed upward.9,10 Figure 2-1.2 helps to define the m
42、ean flame height, L.10It shows schematically the variation of flame inter- mittency, I, versus distance above the fire source, z, where I(z) is defined as the fraction of time that at least part of the flame lies above the elevation, z. The intermittency de- creases from unity deep in the flame to s
43、maller values in the intermittent flame region, eventually reaching zero. The mean flame height, L, is the distance above the fire source where the intermittency has declined to 0.5. Objec- tive determinations of mean flame height according to in- termittency measurements are fairly consistent with
44、(although tending to be slightly lower than) flame heights that are averaged by the human eye.10 The mean flame height is an important quantity that marks the level where the combustion reactions are es- sentially complete and the inert plume can be considered to begin. Several expressions for mean
45、flame height have been proposed. Figure 2-1.3, taken from McCaffrey,11 shows normalized flame heights, L/D, as a function of a Froude number, Q? g (represented as Q?2/5 g to compress the horizontal scale), from data correlations available in the literature. This Froude number is defined Q? g C Q g :
46、cpTgDD2 (1) where Q g C total heat release rate (given in terms of the mass burning rate, mfg,as m gfHc) :and TC ambient density and temperature, respectively cpC specific heat of air at constant pressure g C acceleration of gravity D C diameter of the fire source Quoting McCaffrey with respect to t
47、his figure: “On the left are pool-configured fires with flame heights of the same order of magnitude as the base dimension D. In the middle is the intermediate regime where all flames are similar and the Q2/5 g is seen as a 45-degree line in the fig- ure. Finally, in the upper right is the high Frou
48、de number, high-momentum jet flame regime where flame height ceases to vary with fuel flow rate and is several hundred times the size of the source diameter.” 22Fire Dynamics L 0 0.5 1.0 I z (arbitrary units) Figure 2-1.2.Definition by Zukoski et al.10of mean flame height, L, from measurements of in
49、termittency, I. 02-01.QXD 11/14/2001 10:56 AM Page 2 Buoyancy regime:The correlation by Heskestad (H) represented in Figure 2-1.3 covers the entire Q? g range ex- cept the momentum regime and has the following form given by McCaffrey:11 L D C 1.02 = 3.7Q?2/5 g (2) Actually, this correlation was originally presented in the form:14 L D C 1.02 = 15.6N1/5(3) As before, D is the diameter of the fire source (or effective diameter for noncircular fire sources such that 9D2/4C area of fire source) and N is the nondimensional parame- ter defined by NC ? ? cpT g:2 (Hc/r