Available online at www.sciencedirect.com
Procedia Engineering 45 (2012) 695 – 699
2012 International Symposium on Safety Science and Technology
Numerical studies on effects of cavity width on smoke spread in doubleskin facade LI Junmeia,b,*, XING Xuefeia,b, HU Chengc, LI Yanfenga,b, YIN Chenchena,b, LIU Shanshana,b a
The Key Laboratory of Urban Security and Disaster Engineering , Ministry of Education, Beijing University of Technology, Beijing 100124, China b College of Architecture and Civil Engineering, Beijing University of Technology, Beijing 100124, China c Safety & Emergency Management Lab, Beijing Municipal Institute of Labour Protection, Beijing 100054, China
Abstract Double-skin façade is more and more popular in modern buildings in recently years due to it can provide the building with improved thermal and sound insulation compared with a traditional glazed facade. But fire safety is a problem, the inner glass wall broken during a fire would lead to the smoke and flame spread to the adjacent levels, this might be dangerous for the occupants in those areas. Many such projects failed to comply with the fire safety codes at present. Engineering approach similar to performance-based fire codes practicing in some countries was applied to demonstrate the design is safe. Fire hazard assessment should be supported by the further investigations on the smoke and fire spread in DSF. Cavity depth effects, one of the key factors which can influence the smoke flow in DSF would be studied by numerical method in this paper. By examining the results for cavity depth of 0.5, 1.0 and 1.5 m, it is found that a wider cavity might give better fire safety under the scenarios studied. The outer glass panel would be broken rapidly for the cavity of 0.5m deep. a Cavity depth of 1.0 m might be the most risky design comparing with the other two cavity depths studied, the inner glass panel might be broken before the outer panel, this might lead to the fire spread to the adjacent upper levels, high strength glasses should be used or other protection measures should be taken for the DSF design under this condition. Further studied should be required for an in-depth understanding the smoke and fire spread in DSF in the further.
© 2012 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the Beijing Institute of Technology. Open access under CC BY-NC-ND license. Keywords: double-skin façade; smoke movement; cavity depth; numerical study
1. Introduction Double-skin façades have become very popular in our townscape in recently years. A double skin facade of a building consists of an inner and outer skin. It can offer several advantages compared with a traditional glazed façade[1-2]. It provides an additional layer that helps to reduce the acoustic impact on a building. Its cavities provide a space for positioning advanced sunshading devices that reduce heat gain but allow in natural daylight. Allowing natural daylight into a building for lighting appears to reduce the heat load for artificial lighting on air conditioning[3-4]. Finally, the buoyancy flow in double-skin facade cavities may reduce solar heat gain and also supports HVAC (heating, ventilation and airconditioning) systems. This can help to minimize the size of such systems, thereby reducing the building’s energy[5]. However, this kind of structure of DSF might increase the fire risk as shown in Fig. 1[1], if the inner glass wall of a DSF is broken during a fire, smoke and flame moving out to the cavity might spread to the adjacent levels. This is the main fire safety concern for DSF[1]. Many new construction projects with DSF failed to comply with the fire safety codes.
* Corresponding author. Tel.: +86-10-67391147-103 E-mail address:
[email protected] Supported by the Funding Project for Academic Human Resources Development in Institutions of Higher Learning Under the Jurisdiction of Beijing Municipality and the Innovation Team Plan of Beijing Institute of Science and Technology(IG201206N)
1877-7058 © 2012 Published by Elsevier Ltd. Open access under CC BY-NC-ND license. doi:10.1016/j.proeng.2012.08.225
696
LI Junmei et al. / Procedia Engineering 45 (2012) 695 – 699
Engineering approach similar to performance-based fire codes practicing in some countries was applied to demonstrate the design is safe. Therefore, fire behavior of double-skin facade should be studied more carefully[6-7]. The depth of the cavity is identified as a one of the key parameters in fire spreading in buildings with DSF, smoke movement leading to glass damages in double-skinned facade has been studied by full-scale tests by Chow et al.[6-7], surface temperature and heat flux received were recorded. However, these data are not enough for analyzing the smoke flow pattern in DSF. In this paper, smoke movement inside the cavity between the two layers of glass in a DSF will be studied by the numerical method one the physical of Chow et al.[6-7], more detail results such as temperature distributions and smoke flow patterns would be reported for analyzing the outer or inner glass broken behavior. The effect of the cavity depth on smoke movement would be discussed.
Fig. 1. Propagation and penetration of the fire along a double skin façade.
2. Numerical procedure The CFD code FDS (Fire Dynamics Simulator, version 4.05) based on Large Eddy Simulation (LES)[8], developed by National Institute of Standards and Technology (NIST) is implanted to carry out the numerical simulations in this paper. The transient conservation equations of mass, momentum, energy, and species for low-speed motion of a gas are solved numerically. The three-dimensional space is divided into rectangular volumes in which the gas variables are assumed to be uniform, but changing with time. The Navier-Strokes equations are solved in FDS using large eddy simulation to account for subgrid turbulence. The mixture fraction model is used for combustion reactions. Heat transfer to solid surfaces and convection within the fluid are taken into account. In addition, the radiative transport equation for an absorbing, emitting and scattering medium is also solved. 2.1. Physical model of DSF The physical model of the DSF is from the full burning test rig in literature[6-7] as shown in Fig. 2. It was constructed to give an L-shape configuration. There was a fire room and a cavity space with two levels. The cavity space was 2m wide and 4m high, formed by two glass panels. The fire chamber, taken as the room which a fire broken out, was constructed of bricks with width 2m, depth 1.5m and height 2m. An opening to the cavity of 1.5m 0.8m was assigned, say resulted from breaking of the inner glass panel. A 0.5m diameter pool fire at the center of the chamber was set up, the fire was set as the medium Q-t2 fire[9], and the maximum heat release rate was 550 kW. The back vent of 0.5m 2m of the fire chamber was open for supplying air required for combustion.
(a) Fig. 2. (a) Schematic diagram and (b) sectional view of the experimental setup.
(b)
697
LI Junmei et al. / Procedia Engineering 45 (2012) 695 – 699
2.2. Numerical details Thermocouple trees were set in the simulation to measure the surface temperature as shown in Fig. 3. 7 thermocouples (TC1–TC7) were positioned close to the outer curtain wall, 3 thermocouples (TC8–TC10) were positioned close to the inner curtain wall, and other two thermocouples (TC11–TC12) were positioned at the ceiling of the fire room. The vertical distance between the thermocouples was 0.5m. The ambient temperature in the simulation was 30 . Outer curtain wall
Inner curtain wall
Fire room
Fig. 3. Position of the thermocouples in the numerical simulation.
3. Results and discussion 3.1. Surface temperature distribution
TC2 TC9
400
600
t/st(s)
800
1000 1200
T / ooC 0
70
40 30
(b)
T(oC)
T/ C
50
o
T(oC)
T / oC
60
TC2 TC9
t/s
800 t(s)
1000 1200
200
80
60
50
50
40 30
10
600
0
70
10 400
800 1000 1200
60
20 200
600
t/st(s)
TC1 TC8
70
20 0
400
80
TC3 TC10
80
200
T/ C
200
260 240 220 200 180 160 140 120 100 80 60 40 20
o
0
(a)
T( C)
T(oC)
260 240 220 200 180 160 140 120 100 80 60 40 20
T(oC)
TC3 TC10
260 240 220 200 180 160 140 120 100 80 60 40 20
T / oC
T / oC
T(oC)
Comparison of the temperature on the outer curtain wall and inner curtain wall with different cavity depth of 0.5m, 1.0m and 1.5m at the same height were shown in Fig. 4. Higher temperature differences between the outer skin and the inner skin were found when the cavity depth was small. When the cavity depth became large, lower curtain wall surface temperature could be found, as shown in Fig. 5b, due to more cool air being entrained by the hot smoke in the cavity. For the narrow cavity, outer surface temperature at the measuring height was higher than that of the inner surface. For the wide cavity, inner surface temperature was found to be higher than that of the out surface at the lower height, but will be lower than that of the out surface with the height increasing.
400
600
t(s) t/s
800
1000 1200
TC1 TC8
40 30 20
0
200
400
600 800 1000 1200 t(s)
t/s
10 0
200 400 600 800 1000 1200 t(s)
t/s
Fig. 4. Compaison of the temperature at the same height on otuter and inner curtain wall for cavity depth (a) d = 0.5m and (b) d = 1.5m.
698
LI Junmei et al. / Procedia Engineering 45 (2012) 695 – 699
Variation of the temperature fields in the cavity with time were shown in Fig. 5. Temperature higher than 110oC were found close to the inner curtain surface in the initial time e.g. t = 300s when the cavity depth d = 0.5m, as shown in Fig. 5a, this means that the outer glass facade might crack or break following the past research works by Shields[10-11] and Chow[6,7]. For the cavity depth d =1.0m, higher temperature zone (temperature greater than 110 oC) were found close to the inner surface before the fire lasted for 500s, as shown in Fig. 5b, inner curtain wall might be break down first. For cavity depth d = 1.5m, higher temperature zone were found close to the out curtain surface at the beginning stage of the fire, outer curtain surface might break first under this situation. Breakage of glass on either skin was undesirable, but inner breakage seems to be more risky as occupants in the adjacent upper level would be endangered. Although outer curtain breakage is not dangerous to the upper adjacent levels, outer panes fallen down would also put the pedestrians at risk. 10
5
0
1 0 0
0 0 1
12
10
0
0 0 1
1 5 0
5 7
80
t = 500s
15 0 0 0 2 2 50 25 0 300
1 8 0
0 0 3
140 22 00
t = 900s
2 5 0
6 0
1 2 0 1 0 0
t = 300s
0 5 1
4 0
1 25 0 20
2 5 0
40
10
0
140
135
0 5
50
35
(a)
5 7
1 5 0 1 2 5
0 0 2 0 5 2
60
0 10
60
20 0
0 0 2
10 0
80 0
0
0
40
30 0
6 0
0 0 2 0 20
0 0 1
80
60
140
100 80 60
20 0
1 0 0
200
10
0 80 0 0 8
120
10 0
t = 900s
135 200
0 16 0 8 1
0 60
0
10 0
140
200
500
t = 500s 10
135
0 30 300
0 12
t = 300s
0 10
60
14
0 6
12
20 1 1 0 0 5 0 0
1 4 0 1 1 6 0 0 0 8 0
24
50 70 0 0
60
60
35
30
30
(b)
30
10
12 0
250
50
15 0
0 100
70
50
0 0 0 5 55
30
30
(c)
0 10
0 45
0 35
40
40
100
400
1 2 10 50 00 0
10 0
320
60
40
t = 900s
25 0 300
0 26 0 26
t = 500s
0 35
0
0 25
0
12
0
250
0 18 0 12
t = 300s
22
10
150 70
200
300 250
10 0
60
100
200
40
30
30
Fig. 5. Variation of the temperature distribution in the cavity for cavity depth (a) d = 0.5m, (b) d = 1.0m, and d = 1.5m.
LI Junmei et al. / Procedia Engineering 45 (2012) 695 – 699
699
4. Conclusions Effects of the cavity depth on the smoke movement were studied by the numerical method. By examining the results for cavity depth of 0.5, 1.0 and 1.5 m, it is found that a deeper cavity might give better safety under the scenarios studied. The outer glass panel would be broken rapidly for the cavity of 0.5 m deep. Cavity depth of 1 m might be the most risky design comparing the other two different cavity depths studied. The inner glass panel might be broken before the outer panel, might lead to the fire spread to the adjacent upper levels, this would give an undesirable outcome. To protect against the collapse of the double skins, higher strength glass or other protection methods should be used. Smoke movement in DSF depends on a lot of factors, only the cavity depth effects were studied in this paper, but the height of DSF is limited in this study, stack effects on the smoke flow in real DSF would not be involved. Further investigations should be required for an in-depth understanding the smoke and fire behavior in the cavity space of DSF in the future.
References [1] Åke Blomsterberg, 2007. Best practice for double skin facades, EIE/04/135/S07.38652, WP5 Best Practice Guidelines, 2007 [2] Oesterle, E., Lieb, R.-D., Lutz, G., and Heusler, B. Double-skin facades : integrated planning : building physics, construction, aerophysics, airconditioning, economic viability, Prestel, Munich, 2001 [3] Von Grabe, J, 2002. A prediction tool for the temperature field of double facades. Energy and Buildings, 34(9), p. 891-899 [4] Bodart, M. A. and de Herde, A, 2002. Global energy savings in offices by the use of daylight. Energy and Buildings, 34(5), p. 421-429 [5] Andersen, K. T, 2003. Theory for natural ventilation by thermal buoyancy in one zone with uniform temperature. Building and Environment, 38(11), p. 1281-1289 [6] Chow, W.K., Hung, W.Y, 2006. Effect of cavity depth on smoke spreading of double-skin façade. Building and Environment, 41, p. 970-979 [7] Chow, W.K., Hung, W.Y., Gao, Y., Zou, G., Dong, H, 2009. Experimental study on smoke movement leading to glass damages in double-skinned façade, Construction and Building Materials, 21 p. 556-566 [8] McGrattan, K, 2006. Fire Dynamics Simulator (Version 4), Technical Reference Guide. NIST Special Publication 1018. National Institute of Standards and Technology, Gaithersburg, Maryland, USA, 2006 [9] NFPA 92B: Guide for smoke management systems in malls, atria and large spaces. National Fire Protection Association, Quincy, Massachusetts, USA 2005 [10] Shields, T. J., Silcock G. W. H., Flood M. F, 2001. Performance of a single glazing assembly exposed to enclosure corner fires of increasing severity. Fire and Materials, 25, p. 123- 52. [11] Shields, T.J., Silcock, G.W.H., Hassani, S.K.S, 1997. The behavior of double glazing in an enclosure fire. Journal of Applied Fire Science, 7(3), p. 267-286
