Ventilation improvement in fire smoke control

Transcription

1 Environmental Health Risk II 41 Ventilation improvement fire smoke control L. Santarpia, F. ugliermetti &. Zori Department of Fisica Tecnica, Rome University La Sapienza, Italy Abstract Thermal gradients, wd pressure and mechanical ventilation systems determe a natural airflow travellg the two atmospheres. A fire ignition a confed space creases thermal gradients and determes a growg smoke flux towards adjacent atmospheres. A free air jet from an air curta system reduces the exchange airflow and improves fire and smoke compartimentation. Two theoretical analyses are used for the air curta device dimensiong emergency conditions. The results provide guideles to improve the project quality. Keywords: air curta, confed fire, smoke control, ventilation device. 1 Introduction Smoke management methods can modify smoke movement to protect occupants and fire fighters, and to reduce property damage. Barriers and smoke vents, supplyg and exhaustg big air quantities (purgg dilution) of air fire space, are traditionally used. Doors coupled with mechanical fans are a very common system smoke control, while as the sgle smoke purgg is unable to provided the needed air flow attracted through the open door by pressure differences. In many practical applications escape routes towards refuge areas are not equipped with doors, especially large common-space. In this case the use of air curtas, with the barrier function, could be useful reducg or delayg smoke filtration towards escape ways. Atria commercial multi-stories buildgs, exhibition halls, sports arenas, railway stations are typical examples of large commonplaces where is suggested to apply an air curta system. There are a lot of works and guideles (NFPA 1995 [1], Klote and Milke 199 [], Tamura 1995 [3], Yamana and Tanaka 1985 [4], Hansell and Morgan 1994 [5]), standards and codes devoted to the design of smoke controls, but few papers regards on air curta system used as a smoke barriers. Prelimary

2 4 Environmental Health Risk II studies about air flow rates and pressure field analyses produced by air curta systems can be fd ugliermetti and Santarpia [6, 7, 8] This paper presents a practical application of smoke control by an air curta systems focused to the emergency ventilation system controllg smoke diffusion an underground subway station. Background Smoke diffusion a burng confed space can be analysed function of the fire dynamic and the exhaust ventilation efficiency [9]. A possible approach for an air curta device project and to achieve the fire smoke control a railway deep subway station is to consider a steady fire with an upper layer exhaust [10, 11]. The stationary fire condition determes a constant smoke production rate. This choice volves a security condition, assumg a constant heat release rate and a well fire oxygenation (maximum heat release rate). In this case the air curta system can operates protectg emergency exits and reducg evacuation time. The high temperature fire floats toward the ceilg smoke, makg a smoke layer whose thickness is z f. The outlet airflow, out, stabilizes the smoke layer thickness; its value is carried out by a mass balance the smoke layer control volume (Fig. 1). Outlet airflow, out z f Smoke layer thickness Control volume Clear layer height Plume flow p z c Plume volume Smoke flow, f Entraed airflow, a Heat release of the fire, Q g Heat loss, Q t Figure 1: Smoke layer mass balance. The smoke layer thickness, z f, (and consequently the height of lower clear layer, z c,) is constant the time when: out p. The p flow, comg from the

3 Environmental Health Risk II 43 plume towards the smoke layer, is a mixture of entraed air, a, and smoke, f. This plume flow generated, p, can be estimated as: ( Q Q ) z z + C ( Q Q ) p C1 g t 0 ) where: p mass flow plume at height z, kg/s; Q g heat release rate of fire (HRR), kw; Q t fire heat radiative loss (RHRR), kw; z height above top of the fuel, m; z 0 plume virtual orig, m; C ; C ( (1) The amount ( Qg Q t ) symbolizes the convective heat release rate, Q c, of the plume (CHRR). The virtual orig, z 0, is a function of the fire surface and acts shiftg the plume orig above the top of the fuel (z 0 > 0) or below the top of the fuel (z 0 < 0). The Heskestad s relationship for the virtual orig is: where: D f diameter of fire, m; C z C Q g 1. 0D f () The effective diameter of a fire, when the surface is rectangular, can be expressed by: g t where: S surface of fire, m². S D f (3) π When the fire is startg and developg, before reachg steady conditions (Fig. 1), the low smoke temperature not allows the defed upper and lowers layer creation. In this case the outlet airflow, out, is calculated supposg a perfect smoke mixg the confed space where the fire is developed. The smoke mass balance the confed volume, V, is: V dc dt C( t) dt (4) where: dt fitesimal time terval, s; f smoke mass production, kg/s; C(t) time dependent smoke mass concentration, kg smoke /kg total ; f out

4 44 Environmental Health Risk II Separatg the variables: V dc V d( C( t) f / out ) dt (5) C t) ( C( t) / ) out ( f out f out Integratg and settg C C 0 when t 0, we can obta: ( C( t) out C 0 f ) e out V out t + The outlet airflow, out, can be properly estimated if maximum smoke concentration is lower than security alert level, C s, durg the evacuation from the subway station. 3 Ventilation system project 3.1 Description The depth subway stations are characterized by tunnels (for tra transit), central docks and ascendg scales. The emergency ventilation system consists a metallic duct stalled at the ceilg connected to the airshaft. The ventilation duct takes the air and the smoke from several longitudal grilles disposed with step of.5 m. The surface subway stations are characterized by a rectangular section. In this case the exhaust ventilation system takes the air and the smoke from two localized airshafts (Fig. ). f (6) Air curta Air curta Airshaft Tra Dock Airshaft Fire Figure : Surface subway station layout. An air curta system is stalled at the start of each ascendg scale (emergency exit). This ventilation device starts to operate when the fire is

5 Environmental Health Risk II 45 begng, and the emergency conditions are announced by alert signal, to improve the exit practicability smoke presence. Two different operatg settgs are used to carry out the let and outlet ventilation both normal and emergency conditions. 3. Evacuation time Experimental results, published the scientific literature, [1] shown a temperature value above the fire surface lower than 00 C. A sprkler plant (the activation temperature is set to 68 C) assure an effective temperature value lower 00 C the first five mutes (durg the evacuation phase). Past 5 mutes, the developg fire becomes uncontrollable and determes not acceptable risks, and the evacuation phase must be completed before this period. The emergency action sequences when a fire starts are shown Tab. 1. Table 1: Event-action sequence. Time Event Action Temperature 0 sec. Start fire Start evacuation 5 C 30 sec. Fire detection Start emergency ventilation system 50 C 1 m. Sprkler head breakg Emergency ventilation full operative 80 C 3 m. Developg fire 75% of occupants evacuated 150 C 4 m. Smoke concentration growth Evacuation complete 00 C 3.3 Project design procedure Outlet ventilation system The mimum outlet airflow is determed from the number of the air changes per hour (ACH), parameter fixed by technical provisions, both for normal as for emergency operational conditions (Tab. ). Table : Air changes per hour (ACH). ACH (h -1 ) Normal Emergency 8 The correct outlet airflow must be determed to avoid danger caused by the high door pollution level (smoke, CO, CO, ) and by the temperature Developg fire A smoke mass balance the subway station premise can be used to determe the smoke concentration vs. time durg the evacuation phase, C(t). Settg the maximum smoke concentration level, C s 5%, and the smoke production rate, f 16 kg/s (that corresponds to HRR 3 MW), the smoke concentration, C(t), is calculated with eqn. (5). The operatg phase, shown tab. 3, are considered to compute the smoke concentration.

6 46 Environmental Health Risk II Table 3: Operatg phase. Phase Time Description 1 - itial phase 0 < t < 1 m. time gap between normal and emergency operatg state (normal outlet airflow and maximum smoke release rate) - dilution phase 1 < t < 10 m. outlet airflow set to emergency state and maximum smoke release rate. 3 - fal phase t > 10 m. outlet airflow set to emergency state and zero smoke release rate. The smoke concentration vs. time are synthesized tab. 4 and shown fig. 3. The emergency outlet airflow, out, durg the fire developg is m³/h for a depth station and m³/h for a surface station. Such disproportion (about 30% of airflow) is due, for the surface stations, to the directly air extraction from the airshaft, this case air mixed to smoke outlet airflow is duced from the adjacent tunnel. Table 4: Concentration vs. time. Phase 1 3 Time (m) Depth station (V 8000 m³) Ventilation (m 3 /h) Smoke concentration (%) Surface station (V 7000 m³) Ventilation (m 3 /h) Smoke concentration (%) Steady fire When steady conditions are reached the dimensiong the extraction system can be carried out by the model of the plume illustrated ( ). The dimensiong is obtaed directly through the esteem of the HRR and it

7 Environmental Health Risk II 47 leads to the same result obtaed followg the methodology illustrated Assumg the followg values for the calculations: HRR 3MW Convective heat fraction (Q c /Q t ) 0,7 Fire surface 30 m² A value for virtual orig, z 0-4,6 m, and an outlet airflow out 40 m³/s ( m³/h), are obtaed. Such values are sufficient to determe a free height from smoke, zc, constant the time, of approximately 5 m above the dock. Therefore the outlet airflow ( m³/h) is consistent order to guarantee the ventilation emergency to complete the evacuation of the premises (about 5 m.). For the surface stations the outlet grille airflow must be 30% greater to compensate the air-smoke entraed from the tunnels. The air changes per hour carried out with the present model are 19, approximately 60% greater than the provision values suggested as mimum value (Tab. ). Smoke concentration (%) Time (m) B A Figure 3: Depth (A) and Surface (B) stations smoke concentration vs. time Inlet airflow The outlet airflow gog out to the premises captures external renewal air through the galleries and the exodus way. This airflow, flowg opposite exodus versus, contrasts the premise evacuation (the difference of pressure correspondence of the emergency exits could terfere with the openg of the fire doors). A solution at this problem is the let airflow system. To avoid the smoke diffusion exodus way a localized let grille is stalled on the exit proximity. The vertical let airflow width creates a pressure gradient contrastg the air jet flowg on the communicatg doorway. The air jet can be considered as a lear vertical air curta system operatg with not recirculated airflow [13].

8 48 Environmental Health Risk II The project design can be guided by ASHRAE procedure, therefore, emergency operatg conditions, it can be useful applied the procedure presented Tab. 5 by Santarpia et al. [14]. The procedure is based on a momentum ratio, r M M/M 0, where M is the airflow momentum on the exodus way when the air curta is OFF and M 0 is the air curta jet momentum. Table 5: Design procedure. Settg: Width of push grille m d Door height m H d Outdoor temperature C t e Indoor temperature C t a Velocity coefficient φ 0.7 Width of openg m L Momentum ratio - M r M M 0 Carryg out: Pressure gradient (ρ e outdoor air density ρ a door air density) The average speed of the air flow through the openg when the air curta is OFF Air curta vertical discharge velocity Air curta airflow Pa m/s m/s m³/h p v v max m ( ρ ρ ) e p φ 3 ρ, tot r M a H vm d ρ H g d m max ρ m 3600 d L v The prcipal technical dimensiong parameter is the momentum ratio; normal operatg condition r M can be set to 10, while emergency operatg condition r M can be set to 3. In Tab. 6 are shown the dimensiong data carried out both with ASHRAE method and with the Tab. 5 procedure. On the Authors procedure (tab. 5), the air curta airflow,,tot, can be carried out function of the door steady temperature T a, door height H d, and air curta grille d, with:, tot θ a L 3600 H d rm gφ d 1 L (7) 9 θ g is the gravity acceleration, θ a and θ e are the door and outdoor Kelv temperature. e d

9 Environmental Health Risk II 49 4 Remarks The Authors air curta airflow dimensiong procedure is more precautionary than ASHRAE method emergency condition (airflow 0-30% greater). Moreover the let airflow balances the outlet airflow and improves security exodus ways. Table 6: Design results. v ASHRAE Authors model Settg: Depth Surface Depth Surface Unit E f d m r M H m Te C Ta C Tem C Carryg out: H d d f m ρ e kg/m³ ρ a ρ kg/m³ ρ m kg/m³ p Pa max p f ρ E β m/s v vm v max m/s 3 H d rm vm ρ m d ρ m/s 3600 v d m³/h x m L 4 7, 4 7, m, tot m³/h out m³/h out, tot m³/h

10 50 Environmental Health Risk II Acknowledgement This work has been supported with the economical contribute of the Fondazione Cassa di Risparmio di Roma the frame of a national program devoted to the human health protection. References [1] NFPA, NFPA 9 B. uide for smoke management systems malls, atria and large areas. Qucy, Mass: National Fire Protection Association [] Klote J.H. and J.A. Milke, 199. Design of smoke management systems. Atlanta, ASHRAE: American Society of heatg, Refrigeratg and Air- Conditiong Engeers Inc. [3] Tamura.T., Smoke movement and control high-rise buildgs. Qucy, Mass, NFPA: National Fire Protection Association. [4] Yamana T. and T. Tanaka, Smoke control large spaces. Fire science 5: [5] Hansell.O. and H.P. Morgan, Design approaches for smoke control atrium buildgs. BR-58. arston, U.K., BRE: Buildg Research Establishment. [6] Santarpia L., ugliermetti F., 00. Smoke movement and management larges spaces. Heat and Technology (Vol.1). - International Journal of heat and technology. [7] Santarpia L, ugliermetti F., 000. Air curtas to reduce outdoor pollutants filtration through buildgs aperture. Proc of Intren Conf. HB000, Helski, August [8] Santarpia L. ugliermetti F., A phenomenological approach to the performance of shutter type air curtas Proc. Of Air pollution 99, 7-9 July, San Francisco. [9] Klote J.H., Method of predictg smoke movement atria with application to smoke management. NIST aithersburg, NIST: National Institute of Standards and Technology. [10] Klote J.H., 1997/1. Prediction of smoke movement Atria: Part I- Physical concepts. Ashrae Transactions: 103(). [11] Klote J.H., 1997/. Prediction of smoke movement Atria: Part II- Application to smoke management. Ashrae Transactions: 103(). [1] Heskestad., Fire Plumes- Handbook of fire protection engeerg. Boston, Mass, SFPE: Society of Fire Protection Engeers [13] ASHRAE HVAC Application Handbook, Handbook Editor, ASHRAE, Tullie Circle, Atlanta. [14] Santarpia L., ugliermetti F., Zori. Dynamic efficiency of air curta systems. HEFAT005, 4th International Conference on Heat Transfer, Fluid Mechanics and Thermodynamics September 005, Cairo, Egypt.