1、CH-06-3-1 Overview of Tenability Analyses in Smoke Management Applications James A. Milke, PhD, PE Bryan L. Hoskins ABSTRACT While any smoke from a fire is undesirable, tenability analyses are conducted to assess whether a particular set of smoke conditions can be tolerated. The purpose of this pape
2、r is to provide an overview of tenability criteria and analytical methods that are used to evaluate the level of life safety provided in buildings. The basis of these criteria will be reviewed. Procedures for analysis of tenability will be outlined relative to the scope and limitations of the method
3、s. Tenability criteria relative to visibility, heat exposure, and smoke toxicity are outlined through a review of case studies. INTRODUCTION Tenability analyses have become more frequent in recent years as relatively elementary correlations to conduct an anal- ysis have become available. One of the
4、more frequent uses of the tenability analysis methods is in association with perfor- mance-based design. In these applications, a tenability analy- sis is conducted to estimate the magnitude or concentration of combustion products that can be compared to tolerance levels by building occupants,.as de
5、termined from empirical models (Milke 2002). Forensic investigations may also apply a tena- bility analysis, perhaps estimating when people became inca- pacitated. A third application is in research studies. A tenability analysis was employed in a recent study to assess the performance of household
6、smoke alarms (Bukowski et al. 2004). Principal Steps of Tenability Analysis Any tenability analysis involves three steps. The first step involves describing the source term for the smoke in order to James P. Carroll Diana E. Hugue assess the production of combustion products. Next, a trans- port ana
7、lysis is required to determine if and when occupants become exposed to the smoke. Finally, the effect of the expo- sure needs to be examined. This review will begin with the third part, with the discussion of the first two steps being described via case studies. Methodologies to Assess Physiological
8、 Impact Endpoint Criteria. Endpoint criteria are available in the literature to estimate lethality, incapacitation, and visibility reduction. This paper will emphasize tenability analyses conducted to assess impairment of an individuals ability to evacuate without assistance. In order to provide a c
9、onservative design, incapacitation and visibility reduction are the preferred endpoints in design applications rather than lethal- ity. As such, this review presents information only for tenabil- ity criteria related to incapacitation and visibility reduction. One important aspect concerning tenabil
10、ity criteria is the lack of a single threshold number for exposure to any partic- ular condition produced by fire. Instead, the effect of an expo- sure to a particular condition, or mixture of conditions, is expressed as a combination of exposure time and concentra- tion (or magnitude) of the condit
11、ion. Where a single value is presented in the literature, it is properly done only by assuming a very short exposure, e.g., 5 s or less for exposures involving heat or one or two breaths if inhalation of a gas is considered. Temperature/Heat. Occupants who are evacuating in close proximity to the fl
12、aming fuel source, under a heated smoke layer or within the smoke layer, may be influenced by thermal effects. Being exposed to high temperatures can result in pain, blistering, hyperthermia (heat stroke), skin burns, and respiratory tract burns. James A. Milke is an associate professor and associat
13、e chair, James P. Carroll is a teaching assistant, and Bryan L. Hoskins is a research assistant in the Department of Fire Protection Engineering at the University of Maryland, College Park, Md. Diana E. Hugue is a fire protection engineer with Koffel Associates, Ellicott City, Md. 02006 ASHRAE. 379
14、Within codes and standards of the National Fire Protec- tion Association, NFPA 130 (NFPA 2003b) and NFPA 502 (NFPA 2004) stipulate that the maximum air temperature be 60C for short exposures (i.e., a few seconds), averaging 49C or less for the first six minutes of the exposure and decreasing thereaf
15、ter. NFPA 101 (NFPA 2003a) defines tenable condi- tions for an area of refuge depending on the height of the smoke layer. If the smoke is more than 1.5 m above the floor, the temperature in the area of refuge must be less than 93C. However, if the smoke descends below the 1.5 m level, the temperatur
16、e must be less than 49C. The International Standards Organization (ISO) TS 13571 (IS0 2002) notes that different effects result for expo- sure ofpeople to dry air at temperatures above orbelow 120C. Exposure to lesser temperatures may lead to hyperthermia, while greater temperature additionally lead
17、s to pain from skin bums. A radiant heat flux of 2.5 kW/m2 is the tolerance limit for short exposures without unbearable pain (IS0 2002). NFPA 130 sets a tenability limit of 2.5 kW/m2 for an exposure of 30 minutes, while the tenability limits for radiant heat flux noted in NFPA 502 are 6.3 kW/m2 for
18、 a few seconds, averag- ing 1.58 kW/m2 or less for the first 6 minutes of the exposure and averaging 0.95 kW/m2 for longer exposures. Gas Inhalation. NFPA 130 and 502 spei the carbon monoxide (CO) tenability limit as being 800 ppm based on a 30-minute evacuation period. NFPA 101 specifies a CO tena-
19、 bility limit in terms of an integrated dose of 30,000 ppm-min. Visibility. The reduction of visibility in a fire due to smoke obscuration is an important consideration in tenability analyses. Jin has shown that the walking speed of individuals decreases, perhaps to zero, as the visibility distance
20、is reduced (Jin 1981, 2002). When exposed to irritating smoke with an extinction coefficient of 0.4 m-?, the participants? walking speed decreased from approximately 1.2 m/s to 0.8 m/s. Currently, the work by Jin (1976, 1978) is often cited as a basis for critical visibility levels. In this study, i
21、ndividual subjects walked down a 20 m long corridor containing smoke. The occupants? walking speed was graphed versus the extinc- tion coefficient of the smoke. Walking speed in complete dark- ness was determined to be 0.3 mfs and the optical density corresponding to this walking speed was considere
22、d to be the minimum visibility for people familiar with the building to safely egress. For people unfamiliar with the building, the assumed minimum visibility occurred when the walking speed was unchanged from the normal walking speed. In another study, Jin (1 98 1) determined the smoke densi- ties
23、that caused emotional fluctuations in subjects both familiar and unfamiliar with the test facility. The subjects were seated and had to push a stylus through different sized holes without touching the edges. Jin assumed that the emotional fluctuations of the subjects correlated to the smoke densitie
24、s that would allow for safe escape. These densities were O. 15 m-? for people unfamiliar with the building and 0.5 m-? for people familiar with the building. In contrast to Jin?s studies, Kawagoe and Saito (1967), Shern (LAFD 1961), Rasbash (1967), and Kingman (1953) determined the minimum visibilit
25、y to be 20 m, 13.5 m, 4.5 m, and 1.2 m, respectively. Shern found that an extinction coef- ficient of 0.2 m-? prevents safe egress. Given the discrepancy in results from the various researchers, additional work needs to be conducted to address this issue. For well-ventilated fires, occupants become
26、disoriented at a fuel mass loss concentration of 20 g/m3 (Vaught et al. 2000; IS0 2002). Depending on the familiarity of the occupant with the surroundings, a CO level between 10 and 35 ppm is equivalent to the critical level of smoke visibility (Vaught et al. 2000). NFPA 130 andNFPA 502 require tha
27、t smoke obscura- tion levels be continuously maintained so that a sign illumi- nated at 80 lx is discernible at 30 m. Doors and walls are discernible at 10 m. Assessment Methods IS0 TS 13571 identifies the following methods that can be applied to conduct a tenability analysis: = mass loss fractional
28、 effective dose (FED), fractional effective con- centration (FEC) visibility reduction Mass Loss. The mass loss method used by IS0 TS 1357 1 is the same as Purser?s method of toxic potency (Purser 2002). Mass loss is an alternative method to the FED calculations for determining the effect of exposur
29、e to fire gases. The mass loss method takes into account the combined effects of irritant gases, asphyxiant gases, and lung inflammation. The following simplifying assumptions are made: CO calculations are based on a 70 kg human performing light aerobic work. This assumed mass impacts the aver- ages
30、 for respiratory minute volume ?(RMV) and circula- tion. Many of the toxicity data come from animal testing (Neviaser and Gann 2004). The testing considers ?average? individuals; thus, if the population were more or less healthy, the results would differ. The mass loss or toxic potency method is a m
31、eans of determining the toxicity of fire gases. The advantage of this method over FED analysis is that the input is based solely on the material being burned and does not require a knowledge of the specific combustion products generated nor their respec- tive yields. However, the limitation of this
32、approach is that not many materials have been tested for their toxic potency, so data may not be available to apply this method for a particular fuel of interest. 380 ASHRAE Transactions: Symposia A toxic potency analysis requires the following informa- Description of fire scenario (smoldering, earl
33、y flaming, or post-flashover). The mass loss/dispersal volume-time curve for the fire. The rodent LCt, (product of LC, and exposure time), determined under the same conditions as those in the postu- lated fire scenario. The mass of the combustion products in the smoke is equal to the mass loss of th
34、e fuel. The mass concentration of combustion products is equal to the mass of combustion products divided by the volume of the smoke, as indicated in Equation 1 (IS0 2002). tion: 1. 2. 3. Irritant Gas HCI Am c. = - V Fi (PPm) Irritant Gas Fi (PPN 1 O00 Formaldehvde 250 The exposure dose (g/m3.min) i
35、s equal to the product of the mass concentration and the time step. The FED is then defined as HF NO? n 12 (2) FED = CExposureDose = Am At Toxic Potency =qqj6ij i= 1 i, where (CJ = % (LC) (g.rnp3,min) . FEDIFEC. Irritant gases can inhibit the ability of an indi- vidual to evacuate. The effects of ir
36、ritant gases can be calcu- lated as an FEC. The FEC approach adds the effect of multiple irritant substances that can be produced in smoke, as indicated in Equation 3 (IS0 2002). FEC = +- - HCf I HBr I HF + + acrolein FHCI FHB FHF FSO, FNO, Facroiein Vormaldehyde + + Fforrnaldehyde FG (3) The variab
37、les noted in the square brackets are the irritant gas concentrations (in ppm) generated by the exposure and the variables in the denominators are the irritant gas concentra- tions (ppm) that are expected to seriously compromise an occupants ability to take effective action in order to escape. The fi
38、nal term refers to the concentration of all other irritant 500 Acrolein 301 250 gases (indicated by ir and FG is the limiting concentration for escape of such a mixture. Values of Fi for common irritant gases in smoke are presented in Table 1. Asphyxiant gases impair an individuals ability to self-
39、evacuate by decreasing the amount of oxygen available, caus- ing disorientation and possibly unconsciousness. These hypoxic effects can damage both the central nervous and cardiovascular systems. Exposure to CO leads to the produc- tion of carboxyhemoglobin (COHb) in the blood. The princi- pal effec
40、t of COHb is a decrease in the oxygen-carrying ability of the blood. The fractional effective dose (FED) of asphyxiant gases is estimated by ci At n 12 FED = E-. (et) ,60 i= 1 1, (4) In this case, (Ct)i is the specific exposure dose in ppm*min that would result in incapacitation and, thus, prevent s
41、afe escape. The doses are summed for different species over the duration of the exposure. Incapacitation is expected for an FED in excess of 0.3 (NFPA 2003a, 2003b). An FED approach typically assumes for an FED = 1 .O that 50% of the population will be affected. Consequently, for an FED = 0.3, appro
42、xi- mately 1 1.4% of the population will be affected, assuming the sensitivity of the population is normally distributed. Because the asphyxiant gases that have the most signifi- cant effect on available time for escape are carbon monoxide and hydrogen cyanide, Equation 4 becomes (IS0 2002) I“ i“ Th
43、e gas concentrations in Equation 5 are the average concentrations over the selected time increment, At. If the concentration of carbon dioxide is greater than 2% by volume, an individuals respiratory rate is expected to increase. To account for this effect, both factors in Equation 5 should be incre
44、ased by the following factor: Table 1. Fi for Common Irritant Gases In Smoke I HBr 1 O00 150 - ASHRAE Transactions: Symposia 381 As an altemative approach, Purser (2002) developed an equation for calculating carboxyhemoglobin levels as %COHb = (5.53*10-9*C06*RMV*t. (7) For light levels of activity,
45、e.g., walking, the RMV is expected to be approximately 25 L/min. Using a steady-state approximation for levels of carbon monoxide, the COHb level can be related to the exposure time: %COHb = (5.53*10-9)*(78.5).036”25*t = 0.00122”t. (8) Purser (2002) suggests a COHb limit of 30% to approxi- mate inca
46、pacitation for the average person involved in “light activiy.” TemperaturelHeat While most fire deaths are caused by smoke inhalation, occupants who are in close proximity to the fire, walking under a heated smoke layer, or completely submerged in the smoke layer, may be affected by thermal effects.
47、 Being exposed to high temperatures can result in pain, blistering, hyperthermia (heat stroke), skin burns, and respiratory tract bums. If the heated air contains less than 10% by volume of water vapor, both respiratory tract burns and skin burns are likely. Correlations developed by Wieczorek and D
48、embsey (2001) based on a semi-infinite slab approximation for skin can be applied to estimate the time for the onset of pain (Equa- tion 9) and time for second-degree bums (Equation 10) as a function of heat flux. (9) NFPA 130 and IS0 TS 13571 provide an FED model for increased temperatures using sl
49、ightly different correlations. Assuming black-body radiation, the heat flux can be replaced by the temperature of the smoke layer. 1.63 10 tirad = (T + 273).35 Considering the time to incapacitation for convective heating: Equation 15 may be applied to the results from IS0 TS 13571 or NFPA 130 to estimate the FED from the combined effect of both radiative and convective heating for someone submerged in the smoke layer. (15) FED = x(-+-)At 1 1 Ilrad tIconv A method is not available currently to determine the impact of a combined exposure of both the heated
