Report of the Committee on

Smoke Management Systems

James A. Milke, Chair University of Maryland, MD [SE]

Daniel L. Arnold, RolfJensen & Assoc., Inc., GA [SE] Donald W. Belles, Koffel Assoc., Inc., TN [M]

Rep. American Architectural Mfrs. Assn. Jack B. Bucldey, Houston, TX [SE] Paul J. Carrafa, Building Inspection Underwriters, Inc., PA [E] Elmer F. Chapman, Fire Dept. New York, NY [E] Michael Earl Dillon, Dillon Consulting Engr, Inc., CA [SE] S. E. Egesdal, Honeywell Inc., MN [M]

Rep. Nat'l Electrical Mfrs. Assn. Douglas H. Evans; Clark County Building Dept., NV [E] Gunnar Heskestad, Factory Mutual Research Corp., MA [I] Winfield T. Irwin, Irwin Services, PA [M]

Rep. North American Insulation Mfrs. Assn. Daniel J. Kaiser, Underwriters Laboratories Inc., IL [RT] Gary D. Lougheed, Nat'l Research Council of Canada, ON,

Canada [RT] Francis J. McCabe, Prefco Products, PA [M] Gregory IL Miller, Code Consultants Inc., MO [SE] Stacy R~ Neldhart, Marriott Int'l, Inc., DC [U] Harold E. Nelson, Hughes Assoc. Inc., MD [SE] Zenon A. Pihut, Texas Dept. of Health, TX [E] Dale Rammien, Home Ventilating Inst., IL [M]

Rep. Air Movement & Control Assn. Inc. James Edward Richardson, Colt Int'l Ltd., England [M] Todd E. Schumann, HSB Indusu'ial Risk Insurers, IL [I] ~ Brooks Semple, Smoke/Fire risk Mgmt. Inc., VA [SE]

aul Simony, Conspec Systems, lnc., NJ [M] Paul G. Turnbull, Landis & Gyr Powers, Inc., IL [M]

Alternates

Craig Beyler, Hughes Assoc. Inc., MD [SE] (Alt. to H. E. Nelson)

Richard J. Davis, Factory Mutual Research Corp., MA [I] (Alt. to G. Heskestad)

Victor L. Dubrowskl, Code Consultants Inc., MO [SE] (Alt. to G. R. Miller)

Geraldine Massey, Dillon Consulting Engr, Inc., CA [SE] (Alt. to M. E. Dillon)

Jayendra S. Parikh, Underwriters Laboratories Inc., IL [RT] (Alt. to D.J. Kaiser)

James S. Slater, Pittway systems Technology Group, IL [M] (Alt. to S. E. Egesdal)

Randolph W. Tucker, RolfJensen & Assoc., Inc., TX [SE] (Alt. to D. L. Arnold)

PeterJ. Gore Willse, HSB Industrial Risk Insurers, CT [I] (Alt. to T. E. Schumann)

Michael L. Wolf, Greenheck, WI [M] (Alt. to D. Rammien)

Nonvoting

Bent A. Borresen, Techno Consultant, Norway (Air. to C. N. Madsen)

E. G. Butcher, Fire Check Consultants, England (Alt. to A~ (3. Parnell)

John H. Klote, John H. Klote, Inc., VA [SE] Christian Norgaard Madsen, Techno Consultant, Norway Alan G. Parnell, Fire Check Consultants, England

Staff Liaison: Gregory E. Harrington

Committee Scope: This Committee shall have primary responsibility for documents on the design, installation, testing, operation, and maintenance of systems for the control, removal, or venting of heat or smoke from fires in buildings.

This list represents the membership at the time the Committee was balloted on the text of this edition. Since that time, changes in the membership may have occurred. A key to classifications is found at the front of this book.

The Technical Committee on Smoke Management Systems is presenting two Reports for adoption, as follows:

Report h The Technical Committee proposes for adoption a complete revision of NFPA 92A-1996, Recommended Practice for Smoke-Control Systems. NFPA 92A-1996 is published in Volume 10 of the 1999 National Fire Codes and in separate pamphlet form.

NFPA 92A has been submitted to letter ballot of the Technical Committee on Smoke Management Systems, which consists of 24 voting members. The results of the balloting, after circt~lation of any negative votes, can be found in the report.

Report Ih The Technical Committee proposes for adoption a complete revision of NFPA 92B-1995, Guide for Smoke Management Systems in Malls, Atria, and Large Areas. NFPA 92B- 1995 is published in Volume 10 of the 1999 National Fire Codes and in separate pamphlet form.

NFPA 92B has been submitted to letter ballot of the Technical Committee on Smoke Management Systems, which consists of 24 voting members. The results of the balloting, after circulation of any negative votes, can be found in the report.

586

N F P A 92A - - MAY 2 0 0 0 R O P

N ~ A 9 ~

(Log #CP1) 92A- 1 - (Entire Document) : Accept SUBMITTER: Technic 'd Commit tee on Smoke Management Systems

I RECOMMENDATION: The Technical Committee on Smoke Management Systems proposes a complete revision o[ the 1996 edit ion of NFPA 92A, R e c o m m e n d e d Practice for Smoke-Control Systems, as shown in the draft at the end of this report. SUBSTANTIATION: A task group which included four members of the technical commit tee (Dan Arnold, John Klote, Gary Lougheed, and Paul Turnbull) , as well as J o h n Kampmeyer prepared this proposal, as modif ied by actions taken at the technical commit tee ' s March 1999 ROP meeting, to refine and update the document . Information based on recent research is provided for the design and tesdng of smoke control systems for areas of refuge, elevatoc lobbies and hoistways, and vestibules. A new chapter on computer models for use in the design of smoke control systems was added. In addition, the sections on control systems and the fire fighter 's control station were ref ined and clarified. COMMITTEE ACTION: Accept. NUMBER OF COMMITTEE MEMBERS ELIGIBLE TO VOTE: 24 VOTE ON COMMITTEE ACTION:

AFFIRMATIVE: 21 NOT RETURNED: 3 Carrafa, Chapman, Pihut

COMMENT ON AFFIRMATIVE: EGESDAL: The r ecommended testing frequency for dedicated

smoke control systems in Paragraph 5-4.3.1 should be weekly, not semiannually. The required testing frequency for engine-driven generators (NFPA 72-1996, Table 7-3.2, Item 3) is weekly. Dedicated smoke control systems, like engine-driven generators, sit idle with no way to moni tor the ability of the equ ipment to function. While the availability of power can be moni tored, there are many mechanical componen t s that cannot be moni tored. The present testing frequen=y required as part of the Underwriters Laboratories listing for dedicated smoke control equ ipment "?.-~i~ (UUKL) is weekly. There has been no feedback indicating that a weekly test f requency for dedicated smoke control systems is too frequent, or not f requent enough.

Additionally, A-3-4.6(e) should be dele ted for two r e a s o ~ : already ment ioned in A-3-4.6(a) and 5-4.3.1. ...-.% ~:~4-~!!!!~-.

1. Section 1-1, lines 10-12: The only chan~]]ouid'-'.~'::::'- -~":":"~ MILKE: to insert the words "limit migration of fire gases" b e ~ ' ~ a l ~ , . ~ a tenable environment.. .". All that is noted for d e l e ~ o n ' ~ ' ~ - -~*:':':"" ~ '~ ;{ . -~ being noted as being added. .....:.~.......:... x;~::::.-;::..

2. Section A-l-7, last line: "Gasses" should ~ ; ~ : . :~'~""~!.::. 3. Section B-I.I: Shouldn ' t the author ' s ~ ~ . ' I ' a m

be noted in the publication, ' " " " S oke v .":" "=~.~lt inlet: ." Hi~h-rise Buildings. 19947 ~:: ":::~.i

RICHARDSON: 1. Revision of 92B is usin~i~ :.,.-.%:

- ~ u r a l v ~ d l a t o r " instead of "gravity ven61ator," this documen t sh i~do . . . . . .~e same.

2. We are not convinced by the pressure v e r s u s ~ ' : ' ~ i ~ correlation in Table 2-2.1 and agree with the previous comme .n~-~in this.

SCHUMANN: Page ~1 (Editorial): All of the 4s ' | t the chapter title and at the start o f each paragraph need to be l ined out. This is now Chapter 5 and so marked.

Page 48 (Editorial): Revise A-4-5 to A- Table 4-5 since the * is on the table on Page 30 mid no t the paragraph. Or you can move the • on Page 30 from the table to Paragraph 4-5.

Page 48, A-5-4: I had a note f rom the ROP meet ing that an equation was to be added to the paragraph. None was added and I have no idea of what equation it was.

TURNBULL: The majority of the documen t has been substantially improved by this revision. However, I have concerns that the changes to Paragraphs 3-4.6 and A-3-4.6 have significantly reduced the integrity of dedicated systems.

The 1996 edit ion of this documen t r e c o m m e n d e d supervision (now called verification) for all dedicated equ ipmen t using methods that automatically verify proper operation each time the equ ipment is activated. The r e c o m m e n d e d methods did no t rely on manual intervention. Chapter 4 (now Chapter 5) contained additional recommendat ions for Periodic Testing of dedicated smoke control systems on a semi-annual basis.

The proposed 2000 edition of this documen t removed the distinction between dedicated and non-dedicated equ ipment in Paragraph 3-4.6, meaning that this paragraph now applies to all smoke control equipment . At the same time, a new item (e) "Periodic acceptance testing in accordance with Chapter 5" was added to the list of verification methods in Paragraph A-3-4.6. This change suggests that equ ipment should be verified by automatic means, such as those described in items (a) - (d) OR through

semi-annual manual testing. In o ther words, a non-binding agreement to test the system twice per year removes the need for automatic verification. This seems entirely inappropriate, since system reliability and readiness is a desired factor.

Fur thermore , combining the discussions of verification methods for dedicated and non-dedicated equ ipment may result in the recommendat ions of this d o c u m e n t becoming unclear. Many non- dedicated componen t s are opera ted daily for purposes such as comfort control, and therefore failures of these components are generally not iced quickly. Equipment opera ted in this manne r would no t normally need the type of verification discussed in Paragraph A-$-4.6, Items (a) - (d), to provide assurance that the equ ipment will operate when activated for smoke control. In contrast, dedicated equ ipmen t and some non-dedicated components are infrequendy or never opera ted dur ing normal building conditions. In these cases, automatic verification methods, such as those descr ibed in Items (a) - (d), would be appropriate. In both cases, the periodic testing described in Chapter 5 should be performed.

To remedy the situation described above, I r e c o m m e n d the following changes to the proposed document :

1) Restore the word "dedicated ~ to Paragraph 3-4.6 so that it reads:

"3-4.6* Control System Verification and Instrumentat ion. Every system sl-kQ.uld have means of ensuring it will operate if

activated. The ~ . . . . ~ d frequency will vary according to the complexity an~: : [ "~por~ce of the system."

2) Delete ~ ' . . ' ! ~ ) in Paragraph A-$-4.6.

.#.

N ~ A 9 ~

Practice for Smoke-Control Systems

2000 Edition

Chapter 1 General Informat ion

~.,~" Introduction. All fires p roduce smoke that, if no t controlled, i~ll spread throughout the building or port ions of the building, thereby d " " endanger ing life

A smoke-control system should be des igned to inhibit the flow of smoke into means of egress, exit passageways, areas of refuge, or o ther similar areas of a building. Limiting fire size by providing automatic sprinklers or o ther means of automatic suppression will generally be necessary for effective and economic control of smoke in most occupancies. Other =¢chn'q'.=e= ~ can be appropria te for specialized occupancies or existing facilities. Where smoke-control systems are provided, they should be activated dur ing the early stages of a fire emergency in order to m--nt=2n ~ tz==~!c cn;-rc.nmz:=t "~ ".~hz = rz~ to. ~¢ pr~=zz:z~_.fi~t mieration of fire ~ases and to maintain a tenable environment in the areas to be orotected . The smoke-control system should be functional during the per iod of evacuation of the areas protected by the system. Such systems are in tended to control the migration of smoke tc ~..~2='.22n tcn=5!c cc=~.~.n: "n into protec ted areas so as to nrovide areas of refuee or additional t ime for e trress, but it shoulcl no t be expected that such areas would be completely free of smoke. Smoke-control systems should be engineered for the specific occupancy and building design. Additionally, the smoke- control system design should be coordinated with other life safety systems so that they complement , ra ther than counteract, each other.

1-2 Scope. This r e c o m m e n d e d practice applies to the design, installation, testing, operation, and main tenance of new and retrofi t ted mechanical air- c c . ~ : i c ~ : : . g ~':~ ;~='5]=~.~ systems fc.r Lhz zc::.:r~! ~ f ~ smoke-control systems. This pracdce also applies to systems dedicated :v!cl)' tv ~k.c zc='.=c.! c.f smoke-control systems. (See NFPA 90A. Standard for the Installation of Air-Conditio~in~ and Ventilating ~3stems. for requ~rem~t, for the shutdown of smoke-control s~stems and the use of smoke comOartmentation, i The problem of maintaining tenable conditions witlain large zones of fire origin, such as atria and shopping malls,

587

N F P A 92A - - MAY 2000 R O P

is not addressed by this document . (See NFPA 92B. Gu/de for Smoke Management Sostems in Malls. Atria. and Large Areas. "for maintaining tenable conditions within large zones of tire origin, and NFPA 204~, Guide for Smoke and Heat Venting.)

1-3 Purpose. The purpose of this r e c o m m e n d e d practice is to provide ~uidance in implement ing systems using pressure d:.fferea"_z!~ differeuc¢~ to a ccompl i shone or more of the following:

(1) Inhibit smoke from enter ing stairwells, means of egress, areas of refuge, elevator shafts, or similar areas

(~20 Maintain a tenable envi ronment in thei~l'¢~s of refuge and means of egress during the time required for evacuation

(~.$.) C~.n'.rv.! =n~ rc~ucc ~ the migration of smoke from the gre-ac-ea~ smoke zgne

( t J0 Provide conditions outside the fire zone that enable emergency response personnel to conduct search and rescue operations a n d t o locate and control the fire

(d.5..) Contr ibute to the protect ion of life and reduct ion of property loss

1-4 Definitions. For the purposes of this r e c o m m e n d e d practice, the following terms will have the meanings given in this chapter.

Approved.* Acceptable to the authority having jurisdiction.

Area 9f Refuge. An area of the building separated from other spaces by fire-rated smoke barriers in whfch a tenable envi ronment is mainta ined for the oer iod of t ime that such areas may need to be occupied at t ime of fire.

Authority Having Jurisdiction.* The organization, office, or individual responsible for approving equipment , an installation, or a procedure . .'~':"- . ~ : : : ; : : : : 9 ~i::~:-':-~

been activated, such as dur ing smoke control, testing, or manual overridg operations. Failure or cessation of such oositive confirmation results in an off-normal indication.

Fire Fighters ' Smoke-Control Station (FSCS). A system that provides graphical moni tor ing and manual overriding

capability over smoke-control systems and equ ipment l ~ i d o ~ a t designated location(s) within the building for the use of the fire depar tment . Other fire f ighters ' systems (such as voice alarm, public address, fire depa r tmen t communicat ion, and elevator status and controls) are no t covered in this document .

Pressurized Sm!~o' .¢c~ ~ A type of smoke-control system in which stair shafts are mechanically pressurized, with resoect to the fire area. with outdoor air to keep smoke f rom contaminat ing t h e m dur ing a fire i ndden t .

R e c o m m e n d e d Practice. A d o c u m e n t that is similar in content and structure to a code or s tandard but that contains only nonmanda tory provisions using the word "should" to indicate recommendat ions in the body of the text.

Should. Indicates a recommenda t ion or that which is advised but no t required.

Smoke. The ai rborne solid and liquid particulates and gases evolved when a material undergoes pyrolysis or combustion,

together with the quantity of air that is entrained or otherwise mixed into the mass.

Smoke Barrier. A cont inuous membrane , ei ther vertical or horizontal, such as a wall, floor, or ceiling assembly, that is des igned and constructed to restrict the movement of smoke. A smoke barrier maymig[g_ or may might no t have a fire resistance rating. ~mok-eSuch barriers maYmigl~ have ~ openings protected ~)" . . . . . . . ~'~-:-~, dz:~.zz: ~r :dcq~atc --rfl~w:.

Smoke-Control Mode. A predef ined operat ional configuration of a system or device for the purpose of smoke control.

Smoke-Control System. An eng inee red system that uses mechanical fans to produce ~."fi..~;;~ m=d pressure differences across smoke barriers to l'm".t ~nd ~ r c c t ~ smoke movement.

Smoke-Control Zone. A space within a building enclosed by smoke barriers, including the top and bottom, that is part o f a zoned smoke-control system.

Smoke Damper. A device that meets the requirements of UL 555S, Standard for Safe O Leakage Rated Dampers for Use in Smoke Control Systems, designed to resist the passage of air or smoke. A combinat ion fire a n ~ o l t o k e damper should mee t the requirements of ~L 555, s t a , a ~ , C f ~ j e 9 Fi~e Dar~rs, and UL 555S.

Smoke E x h ~ # ~ . A mechanical or gravity system in tended to move s m o l ~ ~ s m o k e zone to the exterior of the building,..~]~fl~u~01.ng srm~. rag, and venting systems,

eqlq~ ~ i o n d~ Krem. oval, purging, :

as wel l~ to reduce the ~ s t fans utilized p r e ~ ' / ~ in:..it smoke zone, ~.~,ufintenance of a tenable environment m ~ o l ~ ' . ~ o n e is not~ ' th in the capability of these systems.

S m o k ' ~ " ~ e . :Of he smoke-control zone in which the fire is located.

q 'he vertical airflow within buildings caused by the differences between the building

exterior or between two interior snaces.

Environment. An envi ronment in which .t. . . . . . . . : . . . . smoke and heat is l imited or otherwise restricted to

on occupants to a level that is no t life threatening.

Zoned Smoke Control. A smoke-control system that includes smoke exhaust for the smoke zone and pressurization for all contiguous smoke-control zones. The remaining smoke-control zones in the building also may be pressurized.

1-5 Principles of Smoke Control.

1-5.1 Basic Principles. Frequently smoke flow follows the overall air movement within a building. Al though a fire may be confined within a fire-resistive compar tment , smoke can readily spread to adjacent areas through openings such as construct ion cracks, pipe penetrat ions, ducts, and open doors. The principal factors that cause smoke to spread to areas outside a compar tmen t are as follows:

(1) Stack effect

(2) Tempera ture effect of fire

(3) Weather conditions, particularly wind and tempera ture

(4) Mechanical air-handling systems

The factors listed in 1-5.1(1) th rough (4) cause pressure differences across partitions, walls, and floors that can result in the spread of smoke. The movement of smoke can be control led by altering these pressure differences. Building components and equ ipmen t such as walls, floors, doors, dampers , and smokeproof :tr2rt~.'::c= ~ can all be utilized along with the heating, ventilating, and air-condit ioning (HVAC) systems to aid in the control of the movement of smoke. Proper overall building design and tight construction are essential to smoke control. "

The dilution of smoke in the fire area of a compar tmented building is no t a means of achieving smoke control. Smoke

588

N F P A 92A ~ MAY 2000 R O P

control cannot be achieved simply by supplying air to and exhausting air f rom the compartment .

Smoke control can be stated in two basic principles as follows:

(1) Air pressure differences of sufficient magnitude acting across barriers will control smoke movement .

(2) Airflow by itself ~.11 control smoke movement if the average air velocity is of sufficient magnitude.

1-5.2 Pressurization. The primary means of controlling smoke movement is by creating air pressure differences across partitions, floors, and o ther building components . The basic concept of building pressurization ts to establish a h igher pressure in adjacent spaces than in the smoke zone. In this way, air moves into the smoke zone f rom adjacent areas and smoke is inhibi ted f rom dispersing th roughout tae building.

1-5.3" Airflow. Airflow at sufficient velocity can bc u :cd tc :t~.p restrict smoke movement t!'.rcugh - ~ v . . . . . . . . . . This principle is most commonly used to control smoke movement through openjllgr~ a . . . . . . . . . . . . . . . . . ~ . The flow of.alr through the opening into the smoke zone must be of sufficient velocity to pr-event limit migration of smoke from kmM-ng- that zone through such openings. The doors in these donnings are not onen for loner neriods of time. so this Fepresents a transient condit ion that is necessary in order to provide egress from. or access to. the smoke zone (See NFPA 92B. G~ide for Smoke Mana~,ement S~stems in Malls. Atria. and Laree

Since "~c qua='2"Scz cf - r rcqu ' rc~ =re l----'gc, --rfi.cv: - ne t

1-6 Design Parameters.

1-6.1 General. Dez:.g:z ,~',^.z . . . . . :~ .~ :~ .t. . . . . . . . : . . . . a . . . . t~.c stz.'zd.~r?~= rcfcrcncc~ !z 7 uhcm :hcu!d ~ . . . . . . . . . . . -~ u . . . . . . . . . . ~ . . . . . . , ~.~z~'~"

!ncludc :.-= An unders tanding with the authority h a v i n g j u r i s ~ : : of the expected performance of the system and the a c c e p ~ c e t ~ procedures sh0~lql, be established early in the design. ~ l e d ..-'~.~: engineer ing design information is contained in ~ ~ publication, Design of Smoke Management Systems, . :~

1-6.2 Leakage Area&.

/ca-} Small openings in smoke barriers, such as c o n ~ c t i o n joints, cracks, closed-door gaps, and similar c l e a r a n ~ s , should be addressed in terms of maintaining an adequate pressure difference across the smoke barrier, with the positive pressure outside of the smoke zone. Tvoical leakage areas are listed in Table 4-5.

Ogt- Large openings in smoke barriers, such as doors in tended to be open and other sizable openings, should be addressed. These o n e n i n ~ should be e~duated based on geometr ic area. "~n tc.."m: e f

1-6.3_* Weather Data. The tempera ture differences between the exterior and interior of the building cause stack effect and de termine its direction and magnitude. The effect of temperature and wind velocity will ~ r y with building height, configurauon, leakage, and openings in wall and floor construction. The system designer reouires summer and winter design temoeratures. For full analvsls, wind data also needs to be considered.

1-6.4 Pressure Differences. The maximum and min imum allowable pressure differences across the boundar ies of smoke- control zones should be considered. The maximum allowable pressure difference should no t result in door-opening forces that exceed the requi rements of NFPA 101®, Life Safely Code@, or local codes and regulations. The min imum allowable pressure difference should be such that there will be no significant smoke leakage during building evacuation. For the system to be effective,

the pressure needs to be enough that it is no t overcome by the forces of wind, stack effect, or buoyancy of ho t smoke.

1-6.,5 Airflow. Airflow can be used to limit smoke migration when doors in smoke-control barriers are open. The design velodty through an open door should be sufficient to p r - e v e n ~ smoke backflow dur ing building evacuation. The design velocity should take into consideration the same variables as used in the "selection of design pressure differences. " " " "s provided in ASHRAE/SFPE. Design of Smoke Management S~stems.

1-6.6 Number o f Doors Open. The number of doors that could be open simultaneously should be considered. This number will depend largely on the building occupancy and the type of smoke- control system. In some systems, doors will most likely be open for only short per iods of time and the smoke leakage will be negligible. (For the n~mber of doors oben in a stairwell Oressurlzation s~stem~ see

1-7_* Fire Suppression Systems. Automatic sprinkler and other fire suppression systems are an integral part of many fire protect ion designs and the reliability and efficiency of such s~ tems in controlling building fires is well documented . It Is impor tant to recognize that the functions of both suppression and smoke- control systems are j ~ . o r t a n t . Automatic suppression systems can extinguish a fire ~ T ~ r ~ s growth, thereby eliminating additional

Dn the other hand, smoke g e n e r a 6 ~ well-designed smoke- control systet~" . ~ n t a i n a tenable envi ronment along critical egress r o u ~ . . ~ [ ~ r i n g ~ m e it takes the fire suppression system or fire s e rv i~e4~ ' . .~ne l f ~ i e g e final extinguishment.

In ~tio~.. to the fact ~ ' ; ~ e systems perform different f u . ~ i~.~.~mportant~o consider the interaction between the

" k ~ n d fire suppression systems. For example, in a s m o

fully s p : ~ . e d building, pressure differences and airflows needed to control ~ : .~ . .~ovement may be less than in an unsprinklered

.'.:~lk'&.lding due ~.~:-xne likelihood that the maximum fire size will be ~ f l i ~ : * #~dle r than in an unspr inklered building.

!!..::.~ Control system can adversely affect t h e p e r f o r m a n c e of ~ e o u s agent, such as the clean a~ents as d e f i n e d i n NFPA 2001. "i.;..( - - ' . " - . " " . H a l ° n ' or C O , r

systems where the systems axe located in a ~.~ommon space. In the event that both systems are activated ' concurrently, the smoke-control system might dilute the gaseous agent in the space. Because gaseous suppression systems commonly provide only one application of the agent, the potential arises for renewed growth of the fire. Gaseous suppression systems and smoke-control systems cannot perform their in tended functions simultaneously when they are located within the same space.

Chapter 2 Smoke-Control Systems and Applicability

2-1 Introduction.

2-1.1 Purpose. This chapter discusses various types of smoke- control systems cu.':enfl7 : ;--!a~Ic and reviews the advantages and disadvantages of each type.

Determinat ion of system objectives and performance criteria should be made prior to design or construction.

2-1.2 Dedicated and Nondedicated Systems.

2-1.2.1 Dedicated Systems. Dedicated smoke-control systems are i ~ - e t ~ e d i n s ~ l e d for the sole purpose of p.Lq_vi.0j~ smoke control o~Pf. They are separate systems of air-moving and distribution equipment that do no t funct ion under normal building operating conditions. Upon activation, these systems operate specifically to per form the smoke-control function.

AdvaIltages of dedicated systems h: ; 'z ".b.: f~l!v: '-ng :d ' . xn '~g~ include the following:

(1) Modification of thes_.~Lt_gm controls dur ing ~):tc.'.. m^.2~,tcn^.~qzc after installation is less likely t-o--oc-eu~.

589

N F P A 92A ~ MAY 2000 R O P

(2) Opera t ion and control of system general ly simpler.

(3) w,.~., are !e~ !!kc!y t~ ~c : f lee ted Reliance on or imnac t by . . . . 1

"~c mc '~--Sca ' -~ c f o ther bui ld ing systems is limited.

-r,.^.. i. . . . . . ~'^ ~ ' ~ " " - ~ Disadvantages of dedica ted systems include the following:

(~ j , ) -" . . . . . . . . ,~: . . . . . . . . . . . v . . . . . . . . . . . . . . System i m n a i r m e n t s may go

~Bdiscovered between periodic ~ests or m ~ n t e n a n c ~ activities.

(~J 2) Systems ~ e ~ r e ~ d y can require more b u i l d i t ~ space.

2-1.2.2 Nonded l ca t ed Systems. Nonded ica t ed systems are those tha t share c o m p o n e n t s with some o ther system(s) such as the bui ld ing HVAC system. Activation causes the system to change its m o d e of operat ion in order to achieve the smoke-control objectives.

Advantages of nonded ica t ed systems h"-:'¢ ~he fc!!c-:-.'ng ~-.a.a:~-:xage: inc lude the following:

( 1 ) " . . . . . . . . . . . . . . . . . . . " . . . . . v . . . . . . . . . . . . . . . . l m p m r m e n l s to sha red e q u i p m e n t requi red for no rma l bui ld ing opera t ion are tess-likely to r-emai~ be uB-corrected ~ .

(2) E q u ' ~ m e a t cozY, may ~c lower.

( g ~ ~ addit ional space for smoke-control e q u i p m e n t may is necessary.

T~cy ~ave u~.c fcllc;;'i=g Disadvantages of nonded i ca t ed systems include the following:

( l ) System control may become elaborate. ,.~.:-.:-:':-~g .:.~:." ~?.~-~'.,.

(2) .'=~,.:==:=: =o~::=--:= ~ o f ~ ~ : : or controls affec-tng can imnal r smoke-control f ~ " " ' ~ g , * ~ ! : ~ more !!kcl 7 tc occur . ~"::g.:j~:'.:....

. . . . ~ . . . . . . . ~;^~ .-:::::" "::::i:.<:-.~.,

2-1.3 Basic System T~pes. Systems for c o n t r o l ~ > # m o k e ~i! m o v e m e n t in a bui ld ing can general ly be d i v i d e d : : ~ , tw o ~ p a r a t e types: shaft protect ion and floor protection. Shaft"~*.~.¢~[on can be fur ther divided into s m i e t o ~ e F s ~ i ~ e l l p r e s s u r i z a i ~ l systems and elevator hoistway systems. Floor protec t ion enO$i ~npasses several variations of zoned smoke control. Use of a part icular system or combina t ion of systems is d e p e n d e n t on bui ld ing and fire code requ i rements , as well as the specific occupancy and life safety r equ i rements of the si tuation be ing considered.

2-1.4 Tenab le Envi ronment . A n o n s m o k e zone of a zoned smoke- control system can be used as an area i n t ended to protec t occupants for the per iod of t ime needed for evacuation or can be used to nrovlde an area of refu~e. T h e c :mcc~t ~f a - c : ~f t c :ab!e

2-1.5 System Integrity. Smoke-control systems shou ld be des igned, installed, and ma in t a ined such tha t the system will r ema in effective durin.~ evacuation of the protec ted areas. O t he r cons idera t ions may thctate tha t a system shou ld remain effective for longer per iods of t ime. I tems tha t shou ld be cons idered are as follows:

(1) Reliability of power source(s)

(2) A r r a n g e m e n t of power dis t r ibut ion

(3) Me thod and pro tec t ion of controls and system mon i to r i ng

(5) Bui lding occupancy

2-2 Pressure D i f f e r e n c e s .

2-2.1" Table 2-2.1 presents sugges ted m i n i m u m des ign pressure dif ferences developed for gas t empera tu re o f 1700°F (92~°C) n e x t to the smoke barrier. Thes~ pressure d i f f e r e n c ~ are r e c o m m e n d e d for desit, n s tha t are based on main ta in in~ m i n i m u m nressure differences between snecif ied snaces.

If it is des i red to calculate pressure dif ferences for gas t empera tu res o ther t han 1700°F (925°C), the m e t h o d described in A-2-2.1 can be used. Pressure dif ferences p roduced by smoke- control systems t end to f luctuate due to the wind, fan pulsations, doors opening , doors closing, and o ther factors. Shor t - term deviations f rom the sugges ted m i n i m u m design pressure difference may no t have a serious effect on the protect ion provided by a smoke-control system. The re is no clear-cut allowable value of this deviation. It d epends on t ightness of doors, t ightness of construct ion, toxicity of smoke, airflow rates, a n d on the volumes of spaces. In te rmi t ten t deviations up to 50 pe rcen t of the sugges ted m i n i m u m des ign pressure di f ference are cons idered tolerable in mos t cases.

Table 2-2.1 e s t ed M i n i m u m D e s i g n P r e s s u r e s A c r o s s S m o k e Bar f f er s I

Notes:

19~.~ 0.10 0.14

21 ft 0.18

~-~..' For d e ~ $ i purposes , a smoke-control system sh o u ld -~:?:.i! ~-.~-. . . . . ~ ~ - ~ ' % s e r a m , m u m pressure differences u n d e r ~Y!~" ..::~;;!iR'~nditions--~=:~- of stack effect or wind.

• z A S : ' - sprinklered, NS - - nonspr inklered . $ -

..:~g...-. For zoned smoke-contro l systems, the pressure ~!!gi~ifference is m e a s u r e d between the smoke zone a n d

:: .¢#" adjacent spaces while the affected areas are in the smoke- s:" control mode .

2-2.2_* Similar to the pressure differences across smoke barriers, the pressure differences across doors shou ld no t exceed the values given in Table 2-2.2, so tha t the doors can be opera ted while the pressurizat ion system is operat ing. These pressure di f ference values are based on the 30-1bf (133-N) m a x i m u m force pe rmi t t ed to begin open ing the door s t ipulated in NFPA 101, Life Safety Code.

2-3 S'wArtw.':cr ~ Pressur izat ion Systems.

2-3.1 General . T h e pe r fo rmance goal of pressur ized :*~.rtw::er: is to provide a tenable env i ronmen t within the

in the event of a bui ld ing fire. A secondary objective is to provide a s taging area for fire fighters. O n the fire floor, a pressur ized ~ s ~ e f l needs to ma in ta in a pressure di f ference across a closed s m l e t - o ~ ~ door so tha t smoke infiltration is limited. The stairwell nressur izat ion svstem shou ld be des igned to mee t or exceed the m i n i m u m design nressure differences Liven in Table 2-2.1 bu t shou ld n o t e~ceecl th~ m a x i m u m nressure differences Liven in Table 2-2,2, (Refer ~ Section 2-6-when stairwell bressu~ization s~stems are used in combina~on with o~her smoke-control ~ s t e ~ , )

2-3.2 N o o c o m p e n s a t e d a nd C o m p e n s a t e d S y s t e m s .

(4) E q u i p m e n t materials and cons t ruc t ion

59O

N F P A 92A ~ MAY 2000 R O P

Table 2-2.2 Maximum Pressure Differences Across Doors t'~,s,*

Door Closer Door Width ~n. w.g./"

Force ~ (Ibf) 32 ~8 44 48 6 0.45 0.40 0.37 0.34 0.31 8 0.41 0.37 0.34 0.31 0.28 10 0.37 0.34 0.30 0.28 0.26 12 0.34 0.30 0.27 0.25 0.23 14 0.30 0.27 0.24 0.22 0.21

For SI units, 1 Ibf = 4.4 N; 1 in. = 25.4 mm; 0.1 in. w.g. = 25 Pa. Total door opening force is 30 lbf. Door height ~s 7 ft. The distance from the doorknob to the knob side of the door is 3 in. For other door-opening forces, other door sizes, or hardware other than a knob, for example, panic hardware, use the

calculation procedure provided in the ASHRAE/SFPE publication, Design of Smoke Management Systems. Many door closers require less force in the initial portion of the opening cycle than that required' to bring the door to the fully

open position. The combined impact of the door closer and the imposed pressure combine only until the door is opened enough to allow air to'pass freely through the opening. The force imposed by a closing device to close the door is often different from that imposed on opening.

boo r widths apply only if door is hinged at one end; otherwise, use the calculation procedure provided in ASHRAE/SFPE. Design of Smoke Management Systems.

2-3.2.1 In a noncompettsated system, supply air is injected into the ~ s t a i ~ e l l by actuating a single-speed fan, thus providing one pressure difference with all doors closed, another difference with one door open, and so on.

2-3.2.2 Compensated systems adjust to various combinations of doors that are open and. closed, while maintaining positive pressure differences across such openings. Systems compensate for changing conditions by either modulating supply airflows or by relieving excess pressure from the m ~ w e ~ s ~ i ~ e l l .

The response time of the control system should be closely evaluated to ensure that pressures do not fall below the short-term values given in Table 2-2.1. The location of the exhaust inlet

relative to the w~.'~;~" ~ from the s~wc-e,wer- ~,~_.gl]. supply outlet(s) ~ s ~ i ~ e l / o c c u r . should be such that short-circuiting ~:.. ~!!!..,

2-3.2.2.1 Modulating Supply Airflow. In a modulating s u ~ ~ airflow system, the capa,.nty of the supply fan is "ovit least the minimum air velocity when the desig ~dc ~ are open. Figure 2-3.2.,,.1 illustrates such a<..~. :~':"' of air into the ~ s ~ i ~ e l l is varie~g~/:~ g dampers, which are controlled by one or mot ess~..'e sensors that sense the pressure difference betwe, ~ e F s~rwel l and the building. When all the ~ . ~ , L doors are closed, the pressure difference increases and the pass damper opens to increase the bypass air and decreas~ ae flow of supply air to the ~ s ~ r w e l l . In this manner, excessive pressure differences between the ~air-t-oweestai~ell and the building are prevented. The same effect can be achieved by the use of relief dampers on the supply duct when the fan is located outside the building. Supply airflow modulation may also be accomplished by varying fan speed, inlet vanes, variable pitch fan blades, or number of fans operating. Response times of the controls with any system should be considered.

2-3.2.2.20verpressure Relief. Compensated system operation can also be accomplished by overpressure relief. In this instance, pressure buildup in the ~ s m i ~ e l l as doors close is relieved direcdy from the s t a i r - t o w e e ~ to the outside. The amount of air relieved varies with the number of doors open, thus attempting to achieve an essentially constant pressure in the s ~ 4 t o ~ a ~ s ~ i ~ e l l . Exterior relief openings can be subject to adverse effects from the wind so windbreaks or windshields are recommended.

I~ c:d~t'~g ~u!!~ng~, ~ overpressure relief m a ? ~ i s to be discharged into the building,,~-h~ the effects of this on the integrity of the z~'rto;;':.'r= ~ and the interaction with other building HVAC systems should be closely studied ~efvre prcFv=:'ng th'z m..z'~o~. Systems using this principle should have combination fire/smoke dampers in the ~ a i e t o ~ - ~ wall penetrations.

Roof

~" Fan Notes: 1. Fan bypass controlled by one or more static pressure sensors

Iocatedbetween the stairtower and the building interior. 2. A ground-level supply fan is shown; however, tan(s) could be

located at any level.

Figure 2-3.2.2.1 S'.c~cwc:" ~ pressurization with bypass around supply fan.

Overpressure relief may be accomplished by one of the following four methods.

(a) Barometric dampers with adjustable counterweights can be used to allow the damper to open when the maximum interior pressure is reached. This represents the simplest, least expensive method of overpressure relief because there is no physical interconnection between the dampers and the fan. The location of the dampers needs to be chosen carefully because dampers located too close to tile supply openings can operate too quickly and not allow the system to meet the pressure requirements throughout the ~ s t a i ~ e l l . The dampers can be subject to chattering during operation. Figure 2-3.2.2.2 illustrates overpressure relief using barometric dampers.

(b) Motor-operated dampers with pneumatic or electric motor operators are another option for overpressure relief. These dampers are to be controlled by differential pressure controls located in the ~ , ~ , ~ w e ~ . This method provides more positive control over the ~ s t a i ~ e l l pressures than barometric dampers. It requires more control than the barometric dampers and therefore is more complicated and cosily.

(c) An alternate method of venting a ~ s ~ i ~ e l l is through an automatic-opening ~ ~ door or vent to the outside at ground level. Under normal conditions this door

591

N F P A 92A - - MAY 2 0 0 0 R O P

would be closed and, in most cases, locked for security reasons. Provisions need to be made to ensure that this lock does no t conflict with the automatic operation of the system.

Possible adverse wind effects are also a concern with a system that uses an c.pcn aa ~ '~c a-.cc.r 9pening to the exterior at g round level as a vent. Occasionally, h igh local wind velocities develop near the ex te r io rs ta i r - tow~s ta i~e l l door. Such local winds are difficult to estimate in the vicinity of new buildings without expensive modeling. Adjacent objects can act as windbreaks or windshields. Systems udlizin~ vents to the outside of g round level are more effective under cold conditions with the stack effect assisting the stair pressurization system for stairwells primarily abgve grade,

(d) An exhaust fan can be used to prevent excessive pressure when all ~ ~ doors are closed. The fan should be control led by a differential pressure sensor so that it will no t operate when the pressure difference between the staietow~e

and the building falls below a specified level. This should prevent the fan f rom pulling smoke into the ~tk-toa~ee s ~ i ~ e l l when a number of open doors have reduced s t i f i r - t ~ s ~ s ~ r w e l l pressurization. Such an exhaust fan should be specifically sized so that the pressurization system will perform within design limits. To achieve the desired performance, it is believed that the exhaust fan control should be o f a modula t ing type as opposed to an on-off type. Because an exhaust fan will be adversely affected by the wind, a windshield is r ecommended•

Roof l e v e l

Note: Supply fan could be located at i i n y I ~ . ¥::~ ~i.':'~.':" ~-x._ ~:"'":':'::" A -'-';

2-3.2.2.2 s.-'__-:==,o: vent to the outside. :~.~-x:.~':"

2-3.3 Supply Air Source Locat ion.

2-3.3.1 The supply air intake should be separated f rom all building exhausts, outlets from smoke shafts and roof smoke and heat vents, open vents f rom elevator shafts, and o ther building openings that might expel smoke from the building in a fire. This separation should be as great as is practicable• Because ho t smoke rises, consideration should be given to locating supply air intakes below such critical openings. However, ou tdoor smoke movement that might result in smoke feedback depends on location of the fire, location of points of smoke leakage f rom the building, wind speed and direction, and the tempera ture difference between the smoke and the outside air. At present, sufficient information is no t available about such outdoor smoke movement to warrant general recommendat ions favoring ground-level intakes rather than roof- level intakes.

2-3.3.2 With a n y ~ s t a i ~ e l l pressurization system, there is a potential for smoke feedback into the pressurized ~a i r - to¢~

from smoke enter ing the ~ i i - r t o , a e~s~ i~e l l through the pressurization fan intake. Therefore , the capability of automatic shutdown in the event o f smoke feedback should be considered.

2-5.4 Supply Air Fans.

2-3.4.1 Propel ler Fans. Advantages and fimitatious on the use of propel ler fans are descr ibed in 2-~.4.1.1 through 2-~.4.1.3.

2-3.4.1.1 Simple single-point injection systems such as that illustrated in Figure 2-3.4.1.1 can use a roof or exterior wall- moun ted propel ler fan to supply air to stairwells. The use of propel ler fans without windshields is not r e c o m m e n d e d because of the ext reme effect wind can have on the performance of such fans.

2-3.4.1.2 One major advantage of using propel ler fans for stairwell pressurization is that they have a relatively flat pressure response curve with respect to varying flow• Therefore, propel ler fans quickly respond to airflow changes in the stak-towe~ s ta i~e l l as doors are opened and closed without major pressure fluctuations• A second advantage of using propel ler fans is that they are less costly than other types of fans and can provide adequate smoke control with lower installation costs.

2-3.4.1.3 A disadvantage of using propel ler fans is that they often require windshields at the intake because they operate at low pressures and are readily affected by the wind pressure on the building. This is less critical on roofs where the-fans are often protected by parapets and where the direct ion of the wind is at right angles to the axis o f the fan.

Propel ler fans moun t ed on walls pose the greatest susceptibility to the adverse effects of...~.'nd pressures. The adverse effect will be at a maximum when w i ~ e c t i o n is in direct opposit ion to the fan airflow, r e s u l t i n g ~ a IoC'~r intake pressure and thus significantly

• rv.¢::: . . . . reducmg fan e ~ n e s s . Winds that are variable m mtenslty and direction a l s ~ o s e ' ~ " ~ e a t to the ability of the system to maintain control ova .? 7-Kg'.I. stai "~.# . . ta t ic pressure. ~:.~: .,. ..:.~:~.......,

x~,~:.:- .~':~ ~:- '.::::::~::~. ~::- ,~-#- , ~... .:-~.~:i'~ :~.-: - ~'.-:.-'.~. P r o p e l l e r

5 "

Roof level

, .%upply _ air

I x . . _

| • |

Figure 2-3.4.1. I S'~--~vv:== ~ pressurization by r o o f - m o u n t e d prope l l er fan.

2-3.4.2 Other Types o f Fans. Other single-injection systems and multiple-injection systems might require the use of a centrifugal or an in-line axial fan to overcome the increased resistance to flow in the supply ductwork to the staif- toweestai~ell .

2-3.5 Single and Mult iple Injection.

2-3.5.1 S ingle Inject ion.

2-3.5.1.1 A single-injection sTstem is one that has pressurization air supplied to the ~ s t a J ~ e l l at one location. The most common injection point is at the top of the stairwell, as illustrated in Figure 2-3,5.1.

592

N F P A 9 2 A ~ MAY 2000 R O P

Centrifugal ~ . . . . . . j ~ R o o f level

Roof level

. , , - - - r ' ~ - ~ . / C e n t r i f u g a l

aft

Figure 2-3.5.1 8mir-tower-S~i~ell pressurization by top injection.

2-3.5.1.2 Single-injection systems can fail when a few doors are open near the air supply injection point. All of the pressurization air can be lost through these open doors, and the system will then fail to maintain positive pressures across doors farther from the injection point.

2-3.5.1.3 Because a ground-level o t a g v t o w ~ - s ~ e l l door is likely to be in the open position much of the time, a single-bottom- injection system is especially prone to failure. Consideration of this specific situation as well as overall careful design analysis is required for all single-bottom-injection systems, and for all other single-injectlon systems for :m'rt~wc~ ~ in excess of 100 ft (30.5 m) in height.

2-3.5.2 Multiple Injection.

2-3.5.2.1 A multiple in~ection system is one in which air is supplied to the stairwell at muinpie points. Figures 2-3.5.2.1(a) and~_...~.'.'..'~.~ 2-3.5.2.1(b) are two examples of many possible m u l t i p l e - i ~ systems that can be used to overcome the limitations of~iigle- ~-.'~ injection systems. The pressurization fans can be I o ~ f g g r o . ~ . % ~ ...$: • level, roof level, or at any location in between. ..~-~.-..~.

R o o f [ level

. ~ / Duct

x ~ J I I fan

,

Figure 2-3.5.2°I (a) S m i F t e w e e S t ~ e l l pressurization by multiple injection with the fan located at ground level.

2-$.5.2.2 In Figures ', shown in a sepa ra t~ have eliminated t ~ ' ~ supply duct in .. .t~t~ that the d u c t . ~ e s " ~

open neces three

Figure 2-$.5.2.1(b) S'^--~c"-'cr S t ~ e f l pressurization by multiple injection with roof-mounted fan.

~$.5.2.1 (a) and 2-3.5.2.1 (b), the supply duct is :at~l~7-. However, systems have been built that ~t~"~dx~g~se of a separate duct shaft by locating the

in ~ r enclosure itself. Care needs to be taken so ~ e s " ~ e d u c e the required exit width or become an

obstruct io~.~der~?ng__, ,~,~ ~ evacuation.

2-$.5.~*J~" M a I ~ m u l t i p l ~ o n systems have been bu,lt with supl~[air i t ~cu o n " "" each floor. These systems represent t l ~ a t g ~ . l a r e v e n t i n g loss of pressurization air through a few

~ ' l i ~ e v e r , that many injection points might not be with injection points more than cessar ~..r system desi[gns yam mje,

ee stor ~ the designer should use a computer analysis ~[a as the o._f~_'~=n. ASHRAE/ SFPE, Design of Smite Management 3"y~ '~9~tgana lys i s is to ensure that loss of pressurization air ~ ~ W open doors does not lead to substantial loss of . ~ s ~ e l l pressurization.

t~ Vestibules. ~m;rt~wz~ S ~ e l l s that do not have vestthules .aga be pressurized adequately using ~ currently available ~ - h n i q u ~ . Some buildings are constructed with vestibules because of building code requirements. These vestibules may be either nonpressurized or pressurized.

2-$.6.1 Nonpressurized Vestibules. S::~,¢,;;'cr= ~ that have nonpressurized vestibules can have applications in existing buildings. With both vestibule doors open, the two doors in series provide an increased resistance to airflow compared to a single door. This increased resistance will reduce the required airflow so as to produce a given pressure in the ~ s t a i ~ e l l . This subject is discussed in detail in ASHRAE/SFPE. Design of Smoke Management Systems.

In buildings with low occupant loads, it is possible that one of the two vestibule doors may be closed, or at least partially closed, during the evacuation period. This will further reduce the required airflow to produce a given pressure.

2-3.6.2 Pressurized Vestibules and Stairwells. To minimize the ~roount of smoke that enters a vestibule and stairwell, both the vestibule and stairwell can be pressurized. The combined svstem will enhance the effectiveness of the stairwell Dressurization swstem. AIs9, the pressurized vestibule can nrovide a temvorarv area of ur_e_fu~

2-$.6.3 Pressurized Vestibules. With both doors closed, the smoke entering a vestibule can be limited so as to vrovide a tenable environment as an area of refuge. The adjacent stairwell is indirectly pressurized bv airflow from the pressurized vestibule. However. this t~ressurizafion can be lost if the exterior door is open. Also. smoke can flow into the stairwell through any leakage openings iO the stairwell walls adjacent to the floor snace. Such walls should be constructed to minimize leakages for-a stairwell vrotected bv a vressurized vestibule svstem.

593

NFPA 92A - - MAY 2000 ROP

23.6.4 Purged or Vented Vestibules. Purged or vented vestibule systems fall outside the scone of this document . A hazard analysis would be reuuired using the Drocedures provided in the SFPE H~ndbqqk of-Fire Protection EnMneerin~. Prep-curl)" there arc no ,mca.qa a':ailable An en~ineerimt analysis should be ner formed to a t a = a t y z e - ~ the benefits,-if any, of pressurizing, purging, or exhausting vestibules on the ~

2 ~.7 FL-e ~ c c r Exha'_'='..

. . . . . . . . . . . . . . . . . . . . . . . . . . . ~ . . . . . . . . . . . [~ . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . I" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ] . . . . . . . . . . . . . . . . . . .

p c ~ r m c d ~cfore c¢za 'dcr ing th'~ concept.

2-3.7* N u m b e r o f Doors Onen.

F0F ~ stairwfll pressurization system that has not been designed to accommodate the ooenin~ of doors, nressurization will d rop when ally d99rs open. and ~mol~e mav infiltrate the stairwell. For-a buildin~ of low occupant density, the onenin~ and closin~ of a few doors dur ing evacuation will have little e f f ec ton the system. For a

• building with a high occupant densitv and total building ¢vacuation. it can be exnected that most o f the doors will be onen at som¢ time qiuring evacuation. The methods nrovided in ASHRAE/SFPE. Desia-a of Smot~, Management S~steras. can be used to desima systems to accommodate anywhere f rom a few onen doors go almgsr all ~ e door8 being onen. The effect of ooenin~ a door g9 the outside is usually m u d a greater than that of onening- interior doors. When systems are des igned for onen doors and- total building evacuation, the number of oven doors needs to include the exterior stairwell door.

2-4 Elevator Smoke Control. .~:# ~iii 2-4.1 Historically, elevator hoistways have proved to b ~ _ ~ ~i y~{.}.'i:.:.~ available condui t for the movement of smoke throughSut :!.5: }.x~ -~" buildings. This is because the elevator doors have R.~ t.beer {$~ii~.~}:. t ightfittmg and elevat°r h°istways have been Pr~'~" " ~ "; t ~ ~! openings in their tops. The building stack e f f ~ has ~ :1 t ~ driving force that has readily moved s m o k % ~ , and ~ le ;.-':! loosely const ructed elevator hoistways. S ~ g e t h ~ :Is of~ correcting this problem have been proposed a ~ ] ~ y , ttigati

~:2.i$'?.:. ..4 These methods include the following: -%~

(1) Exhaust of the fire floor

(2) Pressurization of enc losed elevator lobbies

(3) Construct ion of smoket ight elevator lobbies

(4) Pressurization of the elevator hoistway

(5)* Closing of elevator doors after automatic recall

2-4.2 The methods listed in 2-4.1(1) th rough (5) have been employed ei ther singly or in combination. However, their application to a particular project, including the effect o f any vents in the elevator hoistway, should be closely evaluated. The open vent at the top of the elevator hoistway may have an undesirable effect on elevator smoke-control systems.

24.3_* Fires have shown the tendency of smoke to migrate into elevator hoistways. Therefore, the use of elevators for egress

D . . . . . . I ~ : . . . . . . . . I . . . . . . 4 . . . . . . . ^ purposes has no t been favored . . . . . . . . . . . . . . . . . . . . ~ . . . . . . . ;_~ w

gm__c -u . :..g d . . . . Research has shown that use of an elevator durin~ a fire is feasible provided the elevator system is nrotec ted aeainst heat. flame, smoke, loss of electrical t3ower, loss of elevator machine room cooline, water intrusion, and- inadver tent activi~ti011 of controls.

2-5.1 General.

2-5.1.1 The pressurized atz.!rtcwcr~ ~ discussed in Section 2-3 are in tended to control smoke to the extent that they inhibit smoke infiltration into the s * = a i r - t o ~ ~ . However, in a building with jus t a pressurized stsfr-to,a~ ~ smoke can flow through cracks in floors and partit ions and through other shafts to threaten life and to damage property at locations remote from the fire. The concept of zoned smoke control discussed in this section is in tended to limit this type of smoke movement within a building.

2-5.1.2 Limiting fire size (mass burn ing rate) increases the reliability and viability of smoke-control systems. Fire size can be limited by fuel control, compar tmentadon , or automatic sprinklers. It is possible to provide smoke control in buildings no t having fire- limiting features, but in those instances, careful considerat ion must be given to fire pressure, high temperatures, mass burn ing rates, accumulat ion of unburned fuels, and o ther outputs resulting from uncontro l led fires.

2-5.2 Smoke-Control Zones.

2-5.2.1 Some buildings can be divided into a number of smoke- control zones, each zone separated f rom the others by partitions, floors, and doors t h ~ . . ~ be closed to inhibit the movement of smoke. A s m o k e - q ~ ' ~ . . z o n e can consist of one or more floors,

~t of'~'ne or or a floor can c9 ~ - more smoke-control zones. A r r a n g e m e n ~ : ~ . 8 smoke-control zones are illustrated in Figure 2 - 5 . 2 . ~ ' : . - - < [ ~"

In Figt ~i~.~.2.1~ the smoke zone is indicated by a minus sign and : d ~ ' ~ are indicated by a plus sign. Each floor can be :ol ~ r zone as in (a) and (h), or a smoke zone can ~ . ~ e than one floor as in (c) and (d). All the ~,zones in a building could be pressurized as in (a) and fly , nousmoke zones adjacent to the smoke zone could be ~d as in (b) and (d). A smoke zone can also be limited of a floor as in (e).

i.'2-5.2.2 In the event of a fire, pressure differences and airflows p roduced by mechanical fans can be used to limit the smoke spread to the zone in which the fire initiated. The concentrat ion of smoke in this smoke zone might r ender it untenable. Accordingly, in zoned smoke-control systems, building occupants should evacuate the smoke zone as soon as possible after fire detection.

2-5.2.3_* Smoke-eomr-ot zones should be kept as small as practicable so that evacuation f rom these zones can be readily achieved and so that the quantity of air required to pressurize the sur rounding spaces will be kept to a manageable level. However, these zones should be large enough so that heat buildup f rom the fire will become sufficiently di luted with sur rounding air so as to prevent failure of major componen t s of the smoke control system.

2-5.2.4 When a fire occurs, all of the nonsmoke zones in the buildin can be pressurized as shown in Figure 2-5.2.1, parts (a), (c), and (e). This system requires large quantit ies of outside air. The comments concern ing location of supply air inlets o f pressurized 8=a2r:owc.~ s ~ e l l s (see 2-3.3) also apply to the supply air inlets for nonsmoke zones.

2-5.2.5 In cold climates, the introduct ion of large quantities o f outside air can cause serious damage to building systems. Therefore, serious considerat ion should be given to emergency preheat systems that will t emper the incoming air and help to avoid or limit damage. Alternatively, pressurizing only those zones immediately adjacent to the smoke zones could limit the quantity of outside air required, as in Figure 2-5.2.1, parts (b) and (d);

2-5 Zoned S m o k e Control .

594

N F P A 9 2 A - - M A Y 2 0 0 0 R O P

(c)

Smoke zone

L } Smoke z o n e

{

+

+

+

(a)

4" +

+

+

+

+

+

+

+

+

+

+

+

+

(b)

(d)

can be pressurized stairwells tha t are also connec ted to the area of

An example of a s imnle system is when the re are onlv nressur ized Ct~rwells in the building. Even then . the interact ion between stairwells t h r o u g h the b-uildint,, oart icularlv when doors are o n e n e d a n d closed, m u s t be considered.

Often these systems are des igned indeoendenf lv to onerate u n d e r the dynamic forces they will encoun t e r (for examole, buoyancy. stack effect, wind). Once the des ign is comnleteci, it is necessary to ~tudy the imnac t the smoke-control system(s) will have on each other. For example , an exhaus ted smoke zone opera t ing in conjunc t ion with a stairwell pressurizat ion system can t end to improve the pe r fo rmance of the stair pressurizat ion system. At the same time, it could increase the pressure di f ference across the door, causing difficulty in open ing the door into the stairwell. For comolex systems, it is r e c o m m e n d e d tha t a network mgdel , such as those discussed in Chan te r 4. be used for the a m ! z ~

2-6 7,~ Fire Floor Exhaust . Exhaus t ing the fire floor can imnrove

F i g u r e 2-5.2.1

+ +

+ + + _ } S m o k e

z o n e + +

+ +

+ +

(el

Arrangemen t s o f smoke<on t ro l zones .

however, the disadvantage of this l imited approach is tha t it is possible to have smoke flow t h r o u g h shafts past the pressurized zone and into unpressur ized spaces. W h e n this alternative is ....~.:~.:~ considered, a careful examina t ion of the potential smoke f involved needs to be accompl i shed and de t e rmi ned a c c e ~ ~"'% ~;:~ ' " ~ L 2-5.2.6 Signals f rom pr:;tccd-'c z'g==ar::g fire a l u m s y s ' ~ ~.~e~,~ used to activate the appropr ia te zoned smoke-control syste ~ .

the alarm zones be a r r anged to coincide with.....~'e s m ~ • 0, 7 zones so as to avoid acavation of the ~ o n g ~ o k e - c o n t r ~ i system(s) . -%::..~::. i

• . . ~ a&<-'k 2-5.2.7 Unless vent ing or exhaus t Is p rowded m t-~.i~re z ~ , the pressure differences will no t be developed and e v e ~ r e s s u r e equalizat ion between the fire zone and the u n a f f e c t e ~ Les will become establ ished and there will be no t h i ng to ~ , ~ j ~ smoke spread into all the zones.

2-6" Areas o f Refu~e. Smoke control for areas of refu~e can be provided by pressurization. For areas of refu~e adiacent to stairwells or elevators, provisions need to be made- to prevent loss of pressure 0r ¢~¢eessive pressures due to the interact ion o f the area 9;[ r e f u ~ smoke control and the shaft smoke control.

2-67 Combina t ion o f Systems.

2-67.1 General . There m l g ~ - b e a r e occasions when

. . . . . . . . n . . . . . ~, ~.~., . . . . . . . more t han one smoke-control system ; ' - '~ Frc~zur'zed ::z2rtcwerc will be o n e r a d n g simultaneously.

t~ t~e "~=errel='-c~ ~f the ccmpe.'=e=t :)-.te:n..=. !t .~..'g~t =c t be . . . . . . . . . . . . . . . . . . . . . . . ~ o 7 . . . . . . . . . . . . . . . . . . v . . . . .

indzpzndent z;~'.cn-..~ zz:d cc..'nSinc "..Scmta. =zh!z':c .% ;'::r'.zz.5!e ~ - - ' - : - - , ; . . . . . . . . For example , pressur ized ~tairwells can . . . . . . . . . . . . . . . ! . . . . . . .

c o n n e c t t o f l o o r a r e a s I h a t a r e D a r t o f a z o n e d smoke-control System. ~[evator hoist~ays tha t are Dart of an elevator smoke- control system can connec t to floor areas that are har t of a zoned smoke-cootrol svstem. T he elevator smoke-control system can be connec ted to areas of retiree tha t in tu rn are connec ted with floor areas ~ha~ ~r~; par t of a zoned smoke-control system. Further . there

3 B u i l d i n g E q u i p m e n t a n d C o n t r o l s

"" ~ . . . ' . . " ' " ____" _ With some modif icat ion, ~ b m ' l d ' m g HVAC c q u l p m c : : systems can be used to ~i~ovide:~uilding smoke control. Various types o f bui ld ing i~ lu jpmen t are discussed in this chapter; however, it is impractical t ~ v e r all types. This chapter provides genera l i n f o r m a u o n on ~ u i p m e n t and controls a n d provides guidel ines that can be used ~o adap t the majority o f systems encountered•

3-2 Heat ing , V e n t i l a t i n g , a n d Air-Condi t ioning (HVAC) E q u i p m e n t .

3-2.1 General . Heat ing, ventilating, a n d a i r -condi t ioning (HVAC) cqu'Fmczt .%~te_[~ normal ly provide a m e a n s of supplying, re turn ing , and exhaus t ing air f rom a condi t ioned space. Th e HVAC e q u i p m e n t can be located within the condi t ioned space, within ad jacent spaces, or within remote mechan ica l e q u i p m e n t rooms. Most HVAC eq-!F.-'..ent Systems in bui ldings where smoke control is cons idered can be adap ted for zoned smoke control.

3-2.2 ~ It is necessary to have the capability of providing adequa te outside air for supply so tha t suff icient d:~ffcrcnd~2 prcz~ure;'. ~ can be achieved across

to ~ ~ migra t ion of smoke into uninvolved areas. Mechanical exhaus t to the outside f rom the smoke zone is also necessary. Some HVAC systems have this capab!lity without a need for modificat ion. When supply and re turn are in te rconnec ted as par t of no rma l HVAC operat ion, smoke d a m p e r s ne ed to be provided for separa t ing the supply and exhaus t du r ing smoke- control operat ion.

3-2.3 HVAC Air-Handlin~ Svstem Types . Various types an d a r r a n g e m e n t s o f air-hand'ling systems are c o m m o n l y used in different types of buildings. Some types are more readilv adantible for smoke control avnl icat ions t han others. The following are examples of typical a i r -handl ing systems.

$-2.3.1 Individual Floor Ids i~yrf l~ / l l~ . Individual a i r -handl ing uni ts serving_ one floor or par t of a floor are a c o m m o n design approach. These HVAC uni ts m igh t or m igh t n o t have separate r e t u r n / e x h a u s t fans. Where these fans are n o t senarate, a m e a n s fgl" providing relief of the fire floor pressures shou ld be investigated ei ther t h rough relief d a m n e r s on the duc t system or o ther means.

595

NFPA 9 2 A - MAY 2000 ROP

Outdoor air can be supplied to each air-handling unit via one of the following:

(1) Exterior louvers and dampers

(2) A common duct system sized to handle the reouired quantities of air

(3) A common duct systerrl having a variable-speed supply fan

( ~ ) Individual variable-speed supply fans.

Air-handling units can be used for smoke control if sufficient outside air and exhaust air capability is available.

3-2.3.2 Central Systems. Some buildings utilize centralized HVAC equipment in main mechanical areas that serve multiple floors within the building. I-[VAC systems of this type might require fire and smoke shaft dampering to provide exhaust of the fire floor and pressurization of the adjacent floors with outside air. Because these central fans can be of large capacity, care must be taken in designing systems to include a means of avoiding excessive pressures within the duct system to prevent rupture, collapse, or other damage. Means need to be provided to control pressures within exits and corridors that could inhibit doors from being opened or closed.

3-2.3.3 Fan/Coil Units and Water Source Heat Pump Units. Fan/coil and water source heat pump types of air-handling units are often located around the perimeter of a building floor to condition the perimeter zones. They may also be nrovided throughout the entire floor area to nrovide the total air- conditioning system. Because the (an/coil and water source heat pump units are comparatively small in outside air capacity and are typically difficult to reconfigure for smoke-control purposes, they ca.': bc cxc!udz~ f;om are g~nerally not s~itable for performing smoke-control functions. If these units have outside air intake provisions, such units within the smoke zone should be shut down when the zone is to be negatively pressurized.

The fan/coil and water source heat pump units are typic..~. ~ . in combination with larger central HVAC equipment o~kiividu..~: interior zone air-handling units. The zone smoke-cor~?..::~ ~,..?: functionality should be provided by the larger central /Sr i ~ . : ~ ..... zone air-handling units. A~¢~:-:-:-. :::~ ~::"

3-2.3.4 Induction ~ S t_~.Lg.m~, l n d u c f i o n - ~ " a i r - ~ g l L I r i t ~ ' ~ i l

located around the perimeter of a b u i l d i n g , : ~ : p r i m ~ t,~. -';~I.:'::~" condition the perimeter zone of older mul '~ f i s~ . . . t~c tu red~ central HVAC system supplies high-pressure h e ~ . . o r coo.l~d air to each perimeter induction unit. Room air is t h ' ~ d u ~ a into the induction unit, mixed with the primary air from ~..~'entral HVAC system, and discharged into the room. ~i.:--v

Induction units within the smoke zone should be shut down or should have the primary air closed off on initiation of smoke control iO smoke zones.

3-2.3.5 Dual Duct and Multizone Systems. HVAC units used in dual duct and multizone systems have cooling and heating coils in them, each in a separate compartment or deck within the unit.

Dual duct systems have separate hot and cold ducts connected between the decks and mixing boxes that mix the air supplied to the space served. For high-pressure systems, the mixing boxes also reduce the system pressure.

Multizone systems mix heated and cooled air at the unit and supply the mixture through low-pressure ducts to each space.

Smoke control should- can be achieved by supplying maximum air to areas adjacent to the smoke zone. This should be accomplished using the cold deck because it is usually sized to handle larger air quantifies. For the smoke zone, supply fans should be shut off.

3-2.3.6 Variable Air Volume (VAV) .Sys.tems. A- Variable air volume (VAV) systems are either individual floor systenls (see 3-2.3.1 ) or centralized multi-floor systems (see 3-2.3.2l that are orovided with terminal devices th~it typically s u p p l 2 i e ~ cooling only. Individual areas served by the system usually have

.^__:_~ _At. . . . . . . . _..: . . . . . other sources of heating (e.g., baseboard or cabinet heaters).

VAV systems vary the quantity of cold air supplied to the occupied space based on actual space demands. Some VAV systems bypass supply air to the return air inlet of the fan, reducing supply air volumes and resultant pressure to avoid fan or ductwork damage. In the smoke-control mode, such bypasses must he closed. For smoke control, the speed of the VAV system ~ fan (s) should be increased and VAV terminal unit controls should be configured to open the terminals in the non-smoke zone to supply maximum volume of outside air to pressurize spaces if sufficient air is available. Bypass dampers 9n svstems usin~ this method need to be closed. It is possible to achieve smoke control with the VAV system supplying minimal air, but care must be taken to ensure that adequate pressure is developed in the space.

3-2.3.7 Fan-Powered Terminal t~.ILS&~al~. S~mz -.-..~'=5!c ̂ -= ";~!'-:mc i ; . _ . ^ m [ ~ . . . . . t ~ . . . . . . ~ ' ~ . ^ 1 7 . ^ 1 ~ 1 . . " . . . . . . I : - . i ~ A a7 . . . . . . . . . . A

co=::ant ".'ol'amc t c~ 'n^2 u~"~. Thczz tc:~,Jnal "'='~ czn=i=t of a

~.'~ he~fing cc'~ "~= =.c2nWA: the d¢~'rz~ =p=z¢ tzmpz~turz. In

.~ .3 .8 :Mixed Systems. Combinations of the examples described ~.'. 3-J2.3.1 through 3-2.3.7 are sometimes used, especially for l~.'i[iffing areas being altered for use other than originally intended. ~ - e must be exercised in the application of different ~ L~-e-s-~ variable volume systems-~ terminal units to

their effect on zoned smoke control. Designs must be based on the capability of system configurations to achieve positive or negadve pressures as needed for smoke control.

3-2.4 Ventilation Systems. In certain instances, specialized systems with no outside air are used for primary cooling and heating. These systems include serf-contained air conditioners, radiant panel systems, and computer room units. Because these systems provid¢ no outside air. thev are not suitable for smoke contrgl anDlication.

Because building codes require ventilation for all occupied locations, a separate system for providing outside air is needed. :~-t,AsThe system supplying outside air can be used for smoke control although the quantity of air provided might not be adequate for full pressurization.

3-2.5 Special-Use Systems. Laboratories, animal facilities, hospital facilities, and other unusual occupancies sometimes use once- through outdoor air systems to avoid contamination and could have special filtration and pressurization requirements. These special-use systems can be suitable for a smoke-control application. Care needs to be exercised to avoid contamination of bacteria-free areas, experiments, processes, and similar areas.

3-3 Smoke Dampers. Smoke dampers used to protect openings in smoke barriers or used as safety-related dampers in engineered smoke-control systems should be classified and labeled in accordance with UL 555S, Standard for Safety Leakage Rated Dampers for Use in Smoke Control Systems.

Dampers in smoke-control systems need to be evaluated for their ability to operate under anticipated conditions of system operation.

3-4 Controls.

596

NFPA 92A - - MAY 2000 ROP

3-4.1 Coordination. The control system should fully coordinate smoke-control system functions among the fire ~rc.tccfi':¢ ; ' :gna"zg alarm system sprinkler system fire fighters ' smoke-control system and any other related systems with HVAC and other building smoke-control equipment .

3-4.2 HVAC System ControLs.

3-4.2.1 Operat ing controls of the HVAC system should be designed or modif ied to accommodate the smoke-control mode, which must have the highest priority over all o ther control modes.

3-4.2.2* Various types of control systems are commonly used for HVAC systems. These control systems utilize pneumatic, electric, electronic, and programmable logic-based control units. All of these control systems can be adapted to provide the necessary logic and control sequences to configure HVAC systems for smoke control. Programmable electronic logic-based (i.e., microprocessor-based) control units, which control and moni tor HVAC systems as well as provide o ther building control and moni tor ing functions, are readily applicable for providing the necessary logic and control sequences for an HVAC system's smoke-control mode of ~peration.

3-4.3 Smoke-Control System Activation and Deactivation. Smoke- control system activation is the initiation of the operational mode of a smoke-control system. Deactivation is the cessation of the operational mode of the smoke-control system. Smoke-control systems normally should be activated automatically;, however, under certain circumstances, manual activation can be appropriate . Unde r ei ther automatic or manual activation, the smoke-control system should be capable of manual override.

Based on the design arm in tended performance of the smoke- control system, considerat ion should be given to the position (i.e., open or closed) of smoke dampers on loss of power and on shutdown of the fan systems that the dampers serve.

3-4.3.1 Automatic actigttion (or deactivation) includes all means whereby a specific fire detect ion device or combination of devices causes activation of one or more smoke-control systems w i t h ~ . : : : . manual intervention. For purposes of automatic act ivat io~fr~ '%il i detect ion devices include automatic devices such as sm~:.¢ ~#" detectors, waterflow switches, and heat detectors. @";:::~.~&..-'-.{"~t

cg '~ 3-4.3.2_* Manual activation (or deactivation) c o v e ~ x ~ . m e a n ~., whereby an authorized person activates one o r . ~ / ~ : ~ = ~ e - systems by means of controls provided for tls..~:-'."~urpose.'~ :'*'~'i:~ purposes of manual activation, the Iocatio~)...i...~te c o n t r o t ~ ~ ; " at a controlled device, at a local control p~ht~i~:: ,he builc~j~'s main control center, or at the fire ~ " " ~ ::: -::.:::"

s h o ~ ~ uired c o m m a n d station. The specific location(s) iql' fire al ~a l l stations by fire authority having.jurisdiction. Manual ..

generally should no t be used to activate smoke-con~4 .systems, ~thcr "hat: ~w2~'.vcr prczzur'z~fi~n z)'ztcm~ which reomre information on the location of the fire to onerate, because of the likelihood of a p e r s o n signaling an alarm from a station outside the smok-~ zone of fire origin.

3-4.$.$* Response TLme. Smoke-control system activation should be initiated immediately after receipt of an appropriate automatic or manual activation command. Smoke-control systems should activate individual componen t s (e.g., dampers , fans) in the sequence necessary to prevent physical damage to the fans, dampers, ducts, and o ther equipment . The total response time for individual components to achieve their desired state or operational ,node should not exceed the following time periods:

(1) Fan operation at the desired state: 60 seconds (2) Complet ion of damper travel: 75 seconds

3-4.3.4 Fire Fighters Smoke-Control Station (FSCS).

3-4.3.4.1 A fire fighters ' smoke-control station (FSCS~ should be t)rovided for all smoke-control systems. The firc fighter~' ~mc.kc . . . . . . ' . . . . : . . . . . ' ~ - ^ rcqu!rcd, ~ C S should provide mc.n!tc.-ng complete status indication and manual control . . . . '-:':'-" ovcr of all smoke-control systems and equipment . Status indicators and controls should be Iot, icaliv and clearly arranged and labeled to convey dae in tended svstem objectives to fire fi~hters who may be unfamiliar with the system. Ot)e-rator controls sl~ould be nrovided for each smoke-control zone. each niece of eou inment

capable of activation for smoke control, or a combinat ion of these ant)roaches. Diat, rams and ~ranhic renresentat ions of the system should be used: l~owever, they might no t be necessary where acceptable to the a~thority h~vingiurisdiction.

3-~,$,4,2 The layoot labeling, and location of the FSCS should be reviewed and at)t)roved bv the fire det)ar tment or fire official t)rior to installation.

3-4.3.4.$ The FSCS should have the highest priority control over all smoke-control systems and equipment . Where manual controls for control of smoke-control systems are also provided at o ther building locations, the control mode selected f rom the FSC~ should prevail. FSCS control should override or bypass other building controls su~:h as Hand-Off-Auto and Star t /Stop switches located on fan motor controllers, freeze detect ion devices, and duct smoke detectors. ESC, S control should no t ..... :a^ ^. ,. . . . . . take precedence over fire sunt)ression, electrical or nersonnel t)rotection devices, and ccn2.r-~!: "~tcndcd to prqtcc-t vGzLnzt

- - - : . . . . . ~ ^ - - A . . . . . ' ] ~ I . ^ . _ : - - - 1 . . A . . . . . . . . . . . • - - - - . - - - : ^ - -

only to W

~.(2 ) The

capability need not bypass Hand-Off-Auto kcated on motor controllers of ~ o l system fans, where the following

motor c m ~ o l l e r s are located in mechanical or ~ e n t rooms, or o ther areas generally accessible i'ed personnel .

:. such a motor controller switch to turn a fan on

~ 4 y cause a n ".r~u~!z =nnunz'~.ficn off-normal e building's main control center durin~ normal

ac'd;ntcd or =rc capab!c ~f acfi'.nfi~n for ~m~kc cvntr~' ~hou!d he

3-4.3.4.4 ~ status indication (ON and OFF) should be provided for ~n -~-.nd cff z~nva~ cf cach "~n~;~du~ dedicated smoke- contrg] System fan and all nondedicated fans having a capacity in excess of 2000 f tS /min (57 mS/min) and used for smoke control. ON status should be sensed by a pressure difference, an Mrflow switch. ~ o r somt~ 9ther proof of airflow. | nd i rec t indication of fan status is no t positive proof of airflow. Additional indications such as damper position can be provided where warranted b~' the complexity of the system. Status indication need no t be provided for individual fans that are included in zoned control and indications.

3-4.4 Controls for Stair Pressurization Systems. The criteria for activation of stair pressurization systems should be as follows:

$-4.4.1 Automatic Activation. Operat ion of any zone of the building prc.tccfive ~ignallng fire Marm system should cause all stair pressurization fans to start. In l imited instances, it can be desirable to pressurize only some ztalrt~;-'erz ~ due to particular building configurations and conditions. A smoke detector should be provided in the air supply to the pressurized ~%r to -we*s~ i~e l l . On detect ion of smoke, the supply fan(s) should be stopped.

3-4.4.2 Manual Activation. A manual override switch should be provided at the FSCS to restart the ~ s t a i ~ e l l pressurization fan(s) after shutdown from the smoke detector, if it m dete rmined that a lesser hazard exists f rom smoke enter ing the stairwell via the fan than smoke migrating into the stairwell f rom adjoining snaces.

597

N F P A 9 2 A ~ M A Y 2 0 0 0 R O P

3-4.5 Controls for Zoned Smoke-Control Systems.

3-4.5.1 The criteria for activation of zoned smoke-control systems should be as follows:

(a) Automatic Activation. An automatic smoke detect ion system can be used to automatically activate a zoned smoke-control system. The smoke detect ion system can be of l imited coverage having spacing greater than 900 ft ~ (84 m 2 ) per detector, provaded that the smoke detectors are so located as to detect smoke before it leaves the smoke zone. The location of smoke detectors and the zoning of the detectors needs to be carefully analyzed to achieve a smoke detect ion system that will reliably indicate the correct smoke zone.

Automatic actuation of a zoned smoke-control system, which is designed to exhaust the fire area and supply air to o ther areas, should be given careful consideration before being under taken because of the possibility of activation of a detector outside the zone of fire origin.

A waterflow switch or heat detector serving the smoke zone can be used to activate the zoned smoke-control system where piping and wiring of such devices coincide with the smoke-control zone.

automatic control accordin[g to building occupancy schedules, energy managemen t strategtes, or o ther nonemergency purposes, such automatic control should be p reempted or overr idden by manual activation or deactivation of the smoke-control equipment . Manual controls provided specifically for this purpose should be clearly marked as to the zone and funct ion served. Manual controls that are shared for both smoke-control functions and o ther building control purposes, as in a building's main control center, should fully cover the smoke-control functionality in the control center operational documentat ion.

3-4.5.3 Sequence. Separate smoke-control systems should be activated in a specific overall sequence to ensure maximum benefi t and minimize any damage or undesirable effects on ducts or equipment .

3-4.5.4_* Schedule. Each different smoke-control system configuration should be fully def ined in a schedule format that includes, but is no t l imited to, the following parameters:

(1) Fire zone in which a smoke-control system automatically activates.

(2) Type of signal that activates a smoke-control system, such as sprinkler waterflow qf: smoke detector.

(b) Manual Activation. Manual activation and deactivation ~W~r"~ic4" control o f the stair pressurization systems should be provided at the (3) Smoke zo e maximum mechanical exhaust to the FSCS as well as at the building's control center. In addition, the outside is i m p ~ ' ~ d d and no supply air is provided. FSC, S should have the capability to override the automatic ~ . . a n shutdown of a stair pressurization fan upon smoke detection, in (4) P o ~ i ~ . ~ a . o k e - c o n t r o l zone(s) where maximum air accordance with the j u d g m e n t of the fire incident commander , supply i . S . - ~ ~ ~ . ~ . ~ - e x h a u s t to the outside is provided. .,::: .... . . . . + , . ~ : .

Zoned smoke-control systems should no t be activated ~ o m (~(~....ON as r e q ~ d to implement the smoke-control manual fire alarm boxes connec ted to the building ~ s y ~ . u.,~...e-speed fans should he fur ther no ted as FAST or gign=l'ng s}-gtcn.= fire alarm svstem. There is no assurance that the ~ % ~ , k ~ . . . M ~ to ensure that the in tended control configuration manual fire alarm box is located in the smoke zone. These fire is a c h i e v @ ~ t ' alarm boxes can be used to cause doors in smoke barrier walls to .:~;.-~.~ "~-~%-~-+::'~'~" ~-..'~:':i~..-,. "~':'~" close prior to smoke control system activation. ,.;..-ii~:::@3-~....Fan ( s ) ~ f F as required to implement the smoke-control

Key-operated manual switches located within a smoke zone that "~"'-"~- ~ ' . ¢ - ' ~ 2 are clearly marked to identify their function can be used to ¢:..'.'.-:'i:'.."~';.'i~:: ~ 7 ) ~ p e r ( s ) OPEN where maximum airflow must be manually activate the zone 's smoke-control system. Where,a ' ~ '~h ieved . is provided, zoned smoke-control systems should be c a p ~ ! of ~i~!: ~!i-"::i::i~ ~':" bemg manually activated from the FSCS by switches cl. . .~'~ . . . . . . . . . '" .... a r ~ . "~::(-.~ Damper(s) CLOSED where no airflow should take place. to identify the zone and function. In addition, where/ .he IS . . . . ,n ~':~:":'-~':':"..~-.-~:: g ~:9.-'::.'::-:: is provided with a main control center, zoned sm.9.,.g.~...-..~....o..ntr, ::::::~:::.. "" (9) Auxiliary functions may be required to achieve the smoke- systems should also be capable of being to be ~ ~ t i ~i::.:-:':'::~...~:::: control system configuration or may be desirable in addit ion to from the building's main control center. ..:#; .... ~:~'-"i! smoke control. Changes or override of normal operat ion static

..::~--~::. "~il:.-'!.:: pressure control set points should also be indicated if applicable. Extreme care should be exercised when ~ J ~ [ : ~ a manu~'~ only

activation to ensure that suitably trained persori'~.]!t.are avai~ble 24 (10) Damper position at fan failure. hours a day, 7 days a week. If this cannot be g u a ~ e d , : . ~ automatic system with manual backup should be us~:~-:!;? 3-4.5.5* Automatic Response to Multiple Signals. In the event ..:::::::.,

5-4.5.23_ Sequence of Control and Priorities. The a£~omatic and manual activation (or deactivation) of zoned smoke-control systems should be subject to the following sequences of control and priorities.

(a) Automatic Activation. Automatic activation of systems and equipment for zoned smoke control should have the highest priority over all o ther sources of automatic control within the building. Where equ ipmen t used for smoke control is also used for normal building operation, control of this equ ipment should be p reempted or overr idden as required for smoke control, q'his equ ipment includes air supply / re turn fans and dampers subject to automatic control according to building occupancy schedules, energy management , or o ther purposes. The following controls should not be automatically overridden:

(1) Static pressure high limits

(2) Duct smoke detectors on supply air systems

(b) Manual Activation and Deactivation. Manual activation or deactivation of zoned smoke-control systems and equ ipmen t should have priority over automatic activation of smoke-control systems and equipment , as well as over all o ther sources of automatic control within the building. Where equ ipment used for zoned smoke control is subject to automatic activauon in response to an alarm from an automatic fire detector of a ~ ~ system, or where such equ ipment is subject to

that signals are received from more than one smoke zone, the system should continue automatic operat ion in the mode de te rmined by the first signal received. However. systems designed for ooerat ion of multinle zones using only heat-activated detect ion ~ v i c e s can exnand the control stratetw to accommodate additional zones, un to the limits of the mechanical system design.

~-4.fi*_ Control System SuFcr-Afc.: Verif icat ion and Instrumentation. Every ded lea ted smoke control system should have means of ensuring it will operate if ~ ~ . The means and freouencv will vary according to the complexity and importance of the system. Supc~A=!~n ~e'Acc: c"__ ~. inc!udc the

,_, End tc end =upcp.==:=n cf ~-he -:-.r:ng, zq'-':Fmzat, .~A A°.~.°

(b) ~- prc=cncc of ...... :-- . . . . . t- . . . . . . t, pc,=;'cr "2c,:-~-.~trcz-na of all circuit

~,t'~ Pc='~'.'e co:'~..'.~m.=dan of f=n ---cd;~don ~7 race.':= ~f duc= ~r=zzu=c, _-:'x' . . . . . . .... , or =.q='-=lcz=t G===c=s "2".=t ===~o::.~ tc !=~ of

598

N F P A 92A ~ MAY 2000 R O P

1 ~" . . . . . . . . . . . . 1 " . . . . . . . . ~ I : " . . . . . . . . . . . . . . t " . . . . . I . . . . . . . . . . . . . . J

d a m p e r " t~c..;

'~ Other " x ~ / ~e',qce~ or ~;ca~3 Z7, appFc.p~at.c.

3-5 Energy Management . Energy m a n a g e m e n t systems, particularly those tha t cycle supply, re turn, and exhaus t fans for energy conservat ion, shou ld be overr idden when their control or opera t ion is in conflict with a smoke-control mode . Because smoke control is an abnormal but critical m o d e of operat ion, it shou ld take priority over all energy m a n a g e m e n t and other n o n e m e r g e n c y control modes .

3-6 Materials.

3-6.1 Materials used for systems providing smoke control shou ld conform to NFPA 90A, Standard for the Installation of Air- Conditioning and Ventilating Systems, and o ther applicable NFPA d o c u m e n t s .

3-6.2 Duct materials should be selected and ducts should be des igned to convey smoke, withstand addit ional pressure (both positive a n d negative) by the supply a n d exhaus t fans when opera t ing in a smoke-control mode , and main ta in their s tructural integrity du r ing the period for which the system shou ld operate.

3-6.3 Special high t empera tu re ~fi~g~ fcr =:..c!'-c ex.hau:t fan= ;':i!! . . . . . . . . ,,.. i. . . . . . . . . . . E o u i n m e n t including, bu t n o t l imited to. . . . . . . . . u = ~ = l . . . . . . . . . ] _ _

fans. du#t~, and b~lance dampe r s shou ld be suitable for thei r i n t ended use and the pr0b~b[e t emnera tu res to which they migh t be exposed.

3-7 Electric Services Installation.

3-7.1 All electrical installations shou ld mee t the r equ i rements of NFPA 70, National Elecbical Code*.

3-7.2 Normal electrical power serving a i r -condi t ioning s y s ~ . ~ general ly have sufficient reliability for nonded ica t ed zone~.~knok'~i,~ control systems. :.~il~i ":?":~ ..".-x ~{*:

.::.".':i~ ~..[~.~#....:. 3-7.3 Whe the r or not ~ ~ L b . £ power ~ sh6i$~..:~..~e considered for dodic-,ated smoke-control s y s t e m s . ~ . r coi~.~_. systems. ,:..4." . . . . . . . :":'~-~!.'~ .,.::::::::.: ~':~::"<" :'.".:J:.':~:':'.~

4-2 Design Eouations. The eouat ions tha t can be used for analysis of pressurized-stairwells a n d elevator smoke control are based on idealizations concern in~ similar bui ld ing leakage f rom floor to f loor and no leakas~e t h r o u g h floors. These euuat ions are provided in ASHRAE/SFPE.-Desio'n of Smoke Manacement S~stems.

4-3* Computer Network Model. A c o m n u t e r network mode l provides a means to calculate the airflows a n d oressure differences t h r o u g h o u t a bu i ld ing in which a smoke-control system is onerat ing. In a network o rogram, a bui ld ing is r eb resen ted by a network o f soaces o r nodes, each at a snecifie oressure a n d temnera ture . Air flows t h rough leakage oa ths f rom regions of h igh oressure to re~ions of low nressure . These leakage oa ths are doors a n d windows t~hat can be o o e n e d or closed. Leak-a~e can also occur t h rough Partitions. floors, and exter ior walls ~ d roofs. See ASH[~,E/SI?PI~. Des/p'n of Smoke Management S~stems. for a discussion of the m e t h o d used to combine mul t io le leakage paths into o[1¢ equiv'4]fnt path. The airflow th rough a leakage pa th is a funct ion of the nressure difference across the leakage oath.

In network models , air f rom outs ide the bui ld ing can be in t roduced by a nressur izat ion system into anv bui ld ing soace, an d the bui ld ing space can be exhaus ted to the outside. T-his allows s imula t ion of stairwell nressurizat ion, elevator shaft nressurization. zoned smoke control and any o ther woe of smoke-control svstem. The pressures " throughout the bui ldin~ and steady flow rates t h r o u e h all the flow na ths are obta ined by solvim} the airflow network, inc luding the driving forces such as wind. the

(4) Sm

)nd~/~Jons with low bui ld ing leakage condi t ions with low bui ld ing leakage mdi t ions with h igh bui ld ing leakage condi t ions with h igh bui ld ing leakage

workmanshin , for examnle , how well a door is fitted or how well w¢'0,ther s t r iooing is installed. Typical leakage areas of const ruct ion cracks in walls and floors of commercia l bu~d ings are listed in

4-6 Friction Losses in Shafts. Pressure losses due to friction of air f lgwing in stairwells are ~imilar to those of air flowing in ducts. Friction loss da ta has been deve loned bv T a m u r a and Shaw (1976~ for Qpen and closed stair t read with various levels o f occunan t

Chapter 45 Testing

4~-1 Introduction.

4 ~1.1_* Absence of a consensus a g r e e m e n t for a test ing p rocedure a n d acceptance cr i ten~ historically has created n u m e r o u s p rob lems at the t ime of system acceptance, inc luding delays in obta in ing a certificate of occupancy.

It is r e c o m m e n d e d tha t the bui ld ing owner a n d bui ld ing des igner share thei r objectives a n d design criteria for smoke control with the authori ty having jur isdic t ion at the p l ann ing stage of the project. The design criteria shou ld include a p rocedure for acceptance testing.

Cont rac t d o c u m e n t s shou ld inc lude operat ional a n d acceptance testing procedures so tha t all parties - - des igners , installers, o w n e r s and authorityLe_s, having jur isdic t ion - - have a clear unde r s t and ing of the system objectives a n d the test ing procedure .

599

N F P A 92A ~ MAY 2000 R O P

Table 4-5 Tvaical Leakmm Areas for Walls and Commercial Buildin~,s

Constrg(tion Element

F l o o r s o f

Exterior Buildin~ Walls ~ 0.50 x 10 ~ (includes construction cracks. ~ cracks around windows and Loose ~ doors) Very Loose ~

Stairwell Walls ~ (includes construction cracks, vAx.ed.Ag.C. but not cracks around Loose ~ windows aod doors)

Elevator Shaft Walls (includes construction cracks. but n9~ cracks and gaps around doors)

Floors (includes construction cracks ~ and gaps around penetrations) v~kWd:Ag~

~For a wall. the area ratio is the area of the leaka~,e through the wail ¢[ivideO by the total wall area- For a floor, the area ratio is the area of the leakage through the floor divided bv the total area of the floor. hValues based on measurements of Tamura and Shaw (1976). Tamura and Wilson (1966). and Shaw. Reardon. and Cheung

~-Values based on measurements of Tamura and Wilson (1966) and Tamura and Shaw (1976). a-Values extranolated from average floor ti~hmess based on ~ . range of tight_hess of other construction elements. ,.x-~ ~-Val-ues based on measurements of Tamura and Shaw (1978). ':i

measurement of nressure differences and door ooening forces under the design conditions agreed on with the authority having iurisdiction.

4_5-1.2" This chapter provides recommendations for the testing of smoke-control systems. Each system should be tested against its specific design criteria- The test procedures described herein have been divided into the following three categories:

(1) Component systems testing

(2) Acceptance testing

(3) Periodic testing and maintenance

4 ~ 2 Operational Testing.

4 ~2.1 General. The intent of operational testing is to establish that the final installation complies with the specified design, is functioning properly, and is ready for acceptance tesdng. Responsibility for testing should be clearly defined prior to operational testing.

4~-2.2 Prior to testing, the party responsible for this testing should verify completeness of building construction, including the following architectural features:

(1) Shaft integrity

(2) Firestopping

(3) Doors/closers

(4) Glazing

(5) Partitions and ceilings

4_5-2.3 The operational testing of each individual system component should be performed as it is completed during construction. These operational tests normally will be performed by various trades before interconnection is made to integrate the overall smoke-control system. It should be certified in writing that each individual system component 's installation is complete and the component is functional. Each component test should be individually documented, including such items as speed, voltage, and amperage.

4_5-2.4 Because smo~.~..~control systems are usually an integral part of building o p e r a # ~ e m s , testing should include the following subsystems to t h ~ t e n ' t ' h ' ~ a t they affect the operation of the smoke- control s yst e rrg. '~:"~.

.-~'..".'i~i .... "":~"-" .~. : ~ . . ~ , . (1) F i r e , a r m system (See NFPA 72, Naturnal

~ ' . ' . ~ e r ~ . . . ~ nage m e ~ system -::::. ,:,~-..-:-~:..~.~....:~:.:,:.:. ,.~'.-:':'~?:. (3) B ' ~ g management system

c<~,(4) HVA

( , ~ ~ equipment

~i~}i6i T~'~peratu re control system %,, :!~:" Power sources

(8) Standby power

(9) Automatic suppression systems

(10) Automatic operating doors and closers

(11) Dedicated smoke-control systems

(12) Nondedicated smoke-control systems

(13) Emergency elevator operation

4~-$ Acceptance Testing.

4~-3.1 General. The intent of acceptance testing is to demonstrate that the final integrated system installation complies with the specified design and is functioning properly. One or more of the following should be present to grant acceptance:

(1) Authority having jurisdiction

(2) Owner

(3) Designer

All documentation from operational testing should be available for inspection.

4~3.2 Test Equipment. Equipment for acceptance testing should be provided as follows:

(1) Calibrated instruments to read pressure difference [differential pressure gauges, inclined water manometers, or electronic manometer (instrument ranges 0-0.25 in. w.g. (0-62.5 Pa) and 0-0.50 in. w.g. (0-125 Pa) with 50 ft (15.2 m) of tubing)]

(2) Spring scale (fi:hc...-m:.n'= :c~-)-c.),

(3) Anemometer

6OO

N F P A 9 2 A - - M A Y 2 0 0 0 R O P

(4) Flow-me~uring hood (optional)

(5) Door wedges

( t ~ ) Signs indicating that a test of the smoke-control system is in progress and that doors must not be opened (or dosed)

( ~ ) Walkie-talkie radios to coordinate equipment operation and data recording

4~-~.$ Testing Procedures. The acceptance testing should include the following procedures, a n d such procedures should meet ~rovisions of the AssociateH Air Balance Council (AABC} and the National Environmental ]~Izncin~ Bureau (N'EI~B~.

4~-$.$.1 Prior to beginning acceptance testing, all building equipment should be placed in the normal operating mode, including equipment that is not used to implement smoke control, such as toilet exhaust, elevator shaft vents, elevator machine room fans, and similar systems.

4~-3.$.2 Wind speed, direction, a n d outside temperature should be recorded ee-~hKiIIg each test day.

4~-$.3.$ If standby power has been provided for the operation of the smoke-control system, the acceptance testing should be conducted while on both normal and standby power. Disconnect the normal building power at the main sei'vice ~disconnect to .

simulate true operating conditions in this mode.

4~-3.3.4 The acceptance testing should include demonstrating that the correct outputs are produced for a given input for each control sequence specified. Consideration should be given to the following control sequet~ces, so tha t the complete smoke-control sequence is demonstrated:

(1) Normal mode

(2) Automatic smoke-control mode for first alarm

(3) Manual override of normal and a u t o m ~ L k

modes

(4) Retom tO normal f

4 ~,3~.~ It is acceptable to perform acceptance ~ fire v ' - ~ . . . . . . . . . . . . . . . . . . . . . . . . ~ - ' - - o alarm system in conjunction w i d e . s m o k e - control system. One or more device circuits on the ~re s i g ~ J ~ ~a rm system can initiate a single input signal to the smoke-control system. Therefore, consideration should be given to establishing the appropriate number of !nitiating devices and initiating device circuits to be operated to demonm'ate the smoke- control system operation.

4~-S.$.6"_ Much can be accomplished to demonstrate smoke- control system operation without resorting to demomtra t iom that use smoke or products that simulate smoke. Where the authority havingjudsdlc t ion requires such demonstrations, they should be . based on the objective ~itet.ia- of inhibitin~ smoke from mit, r-afinf ~rosm smoke ~ane boundaries m other areas, Test cJriteria h ~ e d on the svstem's abifitV to remove smoke from an area are not annrondat~ for z o n ~ ~tmoke=c_ontrol s y k e s . ~nr~ theJ~ ~wt~n~ al"e d~ i tmed for containment, not removal, of smoke. ' "

4~-3.4 ~ a i ~ e ~ S m ~ e H Pressurization Symems. This section anolles where Stairwell nre~urlzation is the only smoke control system in the building. -Where stairwell nre~udzat ion is used in combinati0n with zoned smoke control.-refer to 5-$.8.

4~-3.4.1 With all building HVAC systems in normal operation, measure and record thep remure difference across each

door while the door is closed. After recording the pressure difference across the door, measure the force necessary to open each door, using a spring-type scale. Establish a consistent

~rocedure for recording data throughout the entire test, such that e ~ ~ r w e l l s ide of the doors will s l a y s be considered

as the re~erence point [0 ~/-i~-*iil,..R~, (0 ~ ) ] and the floor side of the doors will always have the pressure difference value (positive if higher than the ~tsit.towe~ stairwell and negative when less than the s m i i . t e w ~ i l ~ , K ) . Since the s ts i t~ew~ stairwell pressurization system is in tended to produce a positive pressure within the etaiemw~ stalrw~l, all negative pressure value~ recorded on the floor side of the doors are indicative of a potential airflow ~ a . t h r , Jaa i l3 tcK~. ime the floor.

4~-&4.2 Verify the proper activation of the s t a l ~ v ~ s ta i~el l

~ ressurization system(s) in response to all means of activation, oth automatic and manual, as specified in the contract

documents. Where automatic activation is required in response to alarm signals received from the building's p t ~ e t e e t i ~ ~ f l r e alarm system, each separate alarm signal should be initiated to ensure that proper automatic activation occurs.

4~-3.4.$ With the s t a i f t ew~s ta i~e l l pressurization system activated, measure and record the pressure difference across each ~ ~ door with all ~ doors dosed. ~ t h e exterior door would normally be onen dur t r~ evacuation, it should

system grtth, the ~ a i ~ * ~ g a i l 3 S ~ preuurization

e t

"4~s.4.~;* With the smim~,~semwell pressurization system activated, open the m i ~ a a L ~ f _ d ~ m . -:--~ . . . . . . ~ ~ . . . . . . used in the swtem desi_gn, e a e - ~ 4 i m e r and measure and record the pressure difference across each remaining dosed

the pressure difference across each closed door, measure the force necessary to open each door, using a sprlng-type scale. Use the same procedure established in 4 ~ $ . 4 . 1 t o record data throughout the entire test. The local code and contract documents ' requirements should be followed regarding the number and Ioc~tion of all doors that need to be opened for this test.

~k~.4.6- All nressure di~erences ~md door o n e n i ~ forces should be documented. The results should d e m o m t r ~ e ~ the a~tem i i funcfionln~ ~ronerlv. No oresmre d i ~ e r e n ~ ~ o u l d be-lem than the minimum d ~ i ~ n r m r e diffm'encm in Table ~-2.1 or the nressures mecified-in the desifn documents. Door onen ln f forces should not exceed that allowed bY the bnildin~ code. -Any nortion of the system not workin~ properly should b e renalred and

t " •

5-3.4.7 Pre~urlzed stairwell ves'dbuies should be t reat~l as a zone in a zoned sm0ke-control s~tem.

4~-~.5 Zoned Smoke-Control System.

4~-&5.1 Verify the exact location of each smoke-control zone and the door openings in the r - - . . . . . . ~ of each zone. If the

N F P A 92A - - MAY 2000 R O P

plans do not specifically identify these zones and doors, the fire Frctcc'.2:'c z 'E~!:=g ~ a r m system in those zones might have to be activated so that any doors magnetically held open will close and identify the zone boundaries.

4~-3.5.2 Measure and record the pressure difference across all smoke-control zones that divide a building floor. The measurements should be made while the HVAC systems serving the floor's smoke zones are operating in their normal (nonsmoke- control) mode. The measurements should be made while all smoke harder doors that separate the floor zones are dosed. One measurement should be made across each smoke barrier door or set of doors, and the data should clearly indicate the higher and lower pressure sides of the doors.

4~-3.5.3 Verify the proper activation of each zoned smoke-control system in response to all means of activation, both automatic and manual, as specified in the contract documents. Where automatic activation is required in response to alarm signals received from the building's prctecr ' :e =!g~: .~g ~ system, each separate alarm signal should be initiated to ensure that p roper automatic activation of the correct zoned smoke-control system occurs. Verify and record the p roper operat ion of all fans, dampers, and related equipment as outlined by the schedule(s) referenced in 3-4.5.4 for each separate zoned smoke-control system.

et 4~-3.5.4 Simulate a fire alarm inout to activate t t ~ all zoned smoke-control systems that is, are-appropriate for each separate smoke-control zone. Measure and record the pressure difference across all smoke barrier_s door~ that separate the smoke zone from adjacent zones. The measurements should be made while all 5- smoke barrier doors that separate the smoke zone f rom the other RI zones are fully closed. One measurement should be made across se each smoke barrier d, oof- or set of ~mc, kc ~arrAcr doors, and the tr data should clearly indicate the higher and lower pressure sides of the doors ~r b~-r'~erg. Doors that have a tendency to open slightly ,..:~.:.::.. :.. due to the pressure difference should have one pressure ":'~!~iii:~ measurement made while held closed and another made while not iii'i, held closed. ":iiii~ i

4~-3.5.5 Continue to simulate fire alarm innuts to actlvate~i~i..>.~. ~: the zoned smoke-control systems ~ ~ ~

make pressure difference measurements as described ~"~...5...:4.~~~2~..-.i- .~'*;~ Ensure that after testing a smoke zone's s m o k e - c o n t r o l " s y s t : ~ - . . - ~ the svstems are properly deactivated and the HV..A..~.:~ems "~ ~.,',-~..-..':-..'~..'..~ involved are re turned to their normal operatan~'~' i ' fr~.i~:or t~% ;' activating ano ther zone's smoke-control s y s t ~ Also e ~ t h a ~ all controls necessary to prevent excessive ~ r e d i f f e r e ' ~ . ~ , ~ " " functional so as to prevent damage to dud~s ~ i ~ . ¢ l a t e d buil~ing

equipment. "%iiliiiii)ii~i::." .J'::"

functioning orooerlv. No nressure difference should be less than the min imum design oressure differences in Table 2-2.1 or the oressures snecified-in-the design documents. Door opening forces should not exceed that allowed bv the building code. Any oordon of the system not working properly should be renaired anti"

5-3.6 Elevator Smoke-Contro l Systems.

5-3.6.1 Hoistwav Pressurization Systems. This section aDolies where elevator hoistwav pressurization is the only smoke-control system in the building. ~There elevator hoistwav nressurizadon is used in combination with zoned smoke control, refer to section 5-

5-3.6.1.1 Verffv the nroner activation of the elevator oressurization system(s) in resnonse to all means of activation, botl~ autorqgti~; and manual, as snecified in the contract documents. Where automatic activation is reouired in response to alarm signals received f rom the building's fire alarm svstem, each seoarate alBrffl signal should be ini t ia tedto ensure that p roper autom-~t,i~ activation occurs•

5-3.6.1.2 With the elevator nressurization system activated, measure and record the oressure difference across each elevator door with

all elevator doors closed. If the elevator door on the recall floor would normally be ooen dur ing system t)ressurization, it should be onen during testing. The I-IVA~C system should be off unless the rlgrmal mode is to-leave the HVAC system on during smoke- control onerations.

5-3.6.1.3 Establish a consistent procedure for recording data th roughout the entire test. such ' that the shaft side of the doors will always be considered as the reference ooint l0 in. w.g. (0 Pa)l and the floor side of the doors will always l~ave the nressure difference Valq~ (oositive if higher than the shaft and net, ative when less than

~3.6.1.4 Since the hoistwav pressurization system is intended to or0duce a nositive oressure ~ t h i n the hoistwav, all ne~cative pressure va)ues recorded on the floor side of the doors are irlOicative of a ootential airflow from the shaft to the floor•

~-$,{i.1.5 If the elevator oressurization system has been desit, ned to operate during elevator movemenL the tests should be repeated under these conditions.

difference should be less than the min imum design nressure differences in Table 2-2.1 or the pressures soecified in the design documents. Elevator lobby door-ooening forces should not exceed that allowed by the building code. Any oort ion of the svstem not working nronerlv should be reoaired and retested.

5-3.7 Area of Refu~e; An area of refuge should be treated as a v

zone in a zoned smoke-control system. The tests outlined in 5-3.5 should be conducted.

5-3.8 Combinat ion o f Smoke-Contro l Systems.

5-3.8.1" Stairwell and Zoned Smoke-Control System. The stairwell oressurization system should be considered as one zone in a zonfgl smoke-control system. The tests outlined in 5-~.5 should be conducted• In addition, the tests oudined in 5-8.4.5 through 5- 3.4.5 should be conducted. All tests should be conducted with both systems operat ing in resnonse to a simulated fire alarm inpqt,

5-3.8.9 Area o f Refuge and Zoned Smoke-Contro l System. An area of refuge should be treated as a separate zone in a zoned smoke-control system. The tests outlined in 5-3.5 should be conducted.

5-$.8.$ Elevator Pressurization and Zoned Smoke-Contro l System. The elevator nressurization system should be considered as one zone in a zoned smoke-control system. Each elevator lobby in an enclosed elevator lobby oressurization system should be considered as one zone in a zoned smoke-control system. The tests oqtliIle~ iB 5-3.5 should be conducted. In addition• the tests outlined in ~i- .g.6.1 or 5-.g.6.~ or both should be conducted.

602

N F P A 92A - - MAY 2000 R O P

/I ~ II 1 ~ . . . . I "-tiN^ ++.^+ __^+k^..l . . . . . .I" .... I....11 . . . . : k ^ . l .I~^..I.,I

s)~tcrn..'z pc~cr r r~ncc . O ~ c r test .-nz',.~c.4,z b--vc ~zzn uzc~ h~:tc-c^2!) • in inz'^,~c= :':here "&c a ' - '~cr '~ ' ha: 'ngj ' - -zdaic 'den

: . . . . . . . + : . . . . ~ ^ : . . . . ~ . . . . . . 4 7 . . . . . . . . . . . I + I . . ^ : . . ~ I I A : ~ . . . .

mcahaJ af tca~ag : :m;":;: ma:=agcmcnt -;.:tom ": quea~cna~!e.

4 L$.2 v . . . . ~,+ v . . . . . p,^. ^r ^.~ . . . . . . . . . ~^~- .-.--. ,,- . . . . , -^^- +

~^~* Ct:em!za! zmc!:c-test~, x ~ z

I V , X * W " . . . . . . . + ^ - + + ~ A

[ ~ \ ~ k D ^ - - I -'~:.--^ + ^ . + .

4_5-3.g~ Testing Documentation. On complet ion of acceptance testing, a copy of all operational testing documenta t ion should be provided to the owner. This documenta t ion should be available for reference for periodic testing and maintenance.

4_5-3.810 Owner 's Manuals and Instruction. Information should

4[~-4A Special arrangements might have to be made for the in t roduct ion of large quantit ies of outside air into occupied areas or computer centers when outside tempera ture and humidity condit ions are extreme. Since smoke-control systems override limit controls, such as freezestats, tests should be conducted when outside air condit ions will no t cause damage to equipment and systems.

Chapter~6 Referenced Publications

6-1 The following documents or port ions the reof are referenced within this r e c o m m e n d e d practice and should be considered as part of its recommendat ions . The edition indicated for each referenced document is the current edition as of the date of the NFPA issuance of this r e c o m m e n d e d practice. Some of these documents might also be referenced in this r e c o m m e n d e d practice for specific informational purposes and, therefore, are also listed in Appendix B.

r ~ - l . l NFPA Publications. National Fire Protect ion Association, 1 Batterymarch Park, P.O. Box 9101, Quincy, MA 02269-9101.

NFPA 70, National Electrical Code*, 1999 edition. be provided to the owner that defines the operation and mmntenance of the system. Basic instruction on the operat ion of NFPA 72, N a t i o n a ~ A l a r m Code, 1996 edition. the system should be piovided to the owner 's representatives. ....-.--.-~.•..-..:..-... Since the owner can assume beneficial use of the smoke-control NFPA 90A, S ~ r d f o ~ : ~ h e Installation of Air-Conditioning and system on complet ion of acceptance testing, this basic instruction Ventilating $ 3 ~ . ~ . 0 . 9 edition. should be completed pr ior to acceptance testing. ~

NFPA 9 2 ~ . ~ d e ke Mana~rement S~stems m Malls. Atria. 4_5-3.011 Partial Occupancy. Acceptance testing should be and La~8"::~:~]::~0()O e ~ ' : " per formed as a single step when obtaining a certificate of ..:#;:'~ %, ~":" occupancy. However, if the building is to be completed or N.....~.:'~.:..dO~!~ife Safet~ ~"de* 2000 edition. occupied in stages, multiple acceptance tests can be conducted in ::" "~':-"iiii~!~i.~. -ff~@~..-'..• - - ' order to obtain temporary certificates of occupancy. NFPJ[:~?..:Guidefor Smoke and Heat Venting, 1998 edition.

4_5-3.4012 Modifications. All operational and acceptance testang ~T'..'~i~i~::.-,NFPA 2001":.'::~"~dard on Clean A~rent Fire Extinguishing Systems. should be per formed on the applicable par t of the system whenever ~li~':!:j'~edition,~y" the system is changed or modified. Documentat ion should be "~.~• :~:~::#:'~:ii~ii.::':~::.,.:#" updated to reflect these changes or modificauons .......... .-. ...... : . . • " " ......... "%::~'nwke~overnent and Control in High-Rise Buildings, 1994 edition.

4_5-4 Periodic Testing. +.:-~iii ;::: "%i ~ . .~ ' 2 Other Publications. -..i'-'~ii#:'::.. .-:ig..:.::,~.,~. % .#ii::"

4_5-4.1 During the life of the budding, maintenance l§"ef f~-~+ ' : ' : : : ? , :~b#~~l .2 .1 ASHRAE Publication. American Society of Heating, ensure that the smoke-control system will perform its intenct'~:i. "::'.'.'.:.'.::.~:" • . . . . . . . . . . . . . . . . ,<.:.. ..~,,.,̂ ~ :. Refrigeraung_ and Air C o n d m o n m g Engineers Inc., 1791 Tulhe funcUon under fire condmons . Vroper m m n t e n ~ - ~ i : f l a e s ~ ... Circle N/17 h.tl~nt~ CA ~()g9Q-99,f)P~ should, as a minimum, include the periodic t # n g o"+'--¥~f.~i, :':"~i~fi~?!';':;:: ' . . . . . . . . . . . . . . . . . . . . . . . equ ipment such as initiating devices, fans, ~ p e r s , c o / : i ~ , ~ " ~qrao~P/S~~E Oesi~',, oeSmoke Manw, ement ~stems 1999. doors, and windows. Tae equ ipmen t s h o ~ : : . m a i n t a i n ' ~ i i i n . . . . . . . . . . . . . .

er 's r e c o m m e ~ . o n s . ( ~ N F P tg of Air -Condi t ioni~d V ~ a t i n g

• ".~:~'.:.i.%.. .#..'.-

b~:Yperformed

accordance with the manufacturer ' s recomm( NFPA 90A, Standard for the lru+tallation Systems, for suggested maintenance practices.)

4 ~4.2 This section describes the tests that should on a periodic basis to de te rmine that the installed systems continue to operate in accordance with the approved design. control system or the zone boundar ies have been modified since the last test. accentance testing should be conducted on the port ion modified.

4~-4.3 The system should be tested in accordance with the following schedule by persons who are thoroughly knowledgeable in the operation, testing, and maintenance of the smoke-control systems. The results of the tests should be documen ted in the operations and maintenance log and made available for inspection. Any port ion of the system not funct ioning in accordance with the original desima should be repaired immediately and the system

4~4.3.1 Dedicated Systems, A t L ~ t Semiannually. Operate the smoke-control system for each control sequence in the current design criteria and observe the operation of the correct outputs for each given input. Tests should also be conducted under standby power, if applicable.

4_5-4.3.2 Nondedica ted Systems, A t L ~ t Annually. Opera te the smoke-control system for each control sequence in the current design criteria and observe the operat ion of the correct output for each given input. Tests should also be conducted under standby power, if applicable.

6-1.2.2 SFPE Publication. Society of Fire Protection Engineers. 7315 Wisconsin Avenue. Suite 1225W. Bethesda. MD 20814.

Handbook of Fire Protection En~neering. 1995.

~;~1.2.83 UL Publications. Underwriters Laboratories Inc., 333 Pfingsten Road, Northbrook, IL 60062.

UL 555, Standard for Safety Fire Dampers, 1999.

UL 555S, Standard for Safe~ 3 Leakage Rated Dampers for Use in Smoke Control Systems, 1999.

6-I.3 Additional Publications.

Achakii. G.Y. and Tamura. G.T.. Pressure Dron Characteristics of Typ. i c~ Stairshafts in High-Rise Buildings. ASI~RAE Transactions. American Society of Heating. Refrigerating and Air Condit ioning Engineers. Atlanta, GA~ Volume 94. Partl . 1988. on. 1223-1236.

Shaw• C.Y.. Reardon. I.T. and Cheung. M.S.. Changes in Air Leakage Levels of Six Canadian Office Buildings. ASHRAE lournai. American Society of Heating, Refrigerating ancl Air Condit ioning Engineers. Adanta. GA, 1993.

Tamura, G.T, and Shaw, C.Y., Studies on Exterior Wall Air TiEhtne~ and Air Infiltration of Tall Buildings• ASHRAE Transactions. American Society of Heating• Refrigeratin~ and Air Condit ioning Engineers• Atlanta• CA. Volume 82. Part 1~ 1976. riD+

_ - o

603

N F P A 92A - - MAY 2000 R O P

Tamura. G.T. and Shaw. C.Y.. Air Leakage Data for the Design fo Elevator and Stair Shaft Pressurization Svst-ems. ASHRAE Transactions. American Societv of Heating. Refrigerating and Air Condit ioning Engineers. Atlanta. GA. Volume 82[Par t 2. 1976. pp.

Tamura. G.T. and Shaw. C.Y.. Exnerimental Studies of Mechanical V e n t i n e f o r Smoke Control in Tall Office Buildinws. ASHRAE Transactions. American Societv of Heating. Refrigerating and Air Condi t ioning EnL, ineers. Atlanta. GA. Volume 86. #art 1. 1978. nn. 54-71.

Tamura. G.T. and Wilson. A.G.. Pressure Differences for a Nine- Story Building as a Resutlt of ( ~ i m n e v Effect and Ventilation SYstem Onerat ion. ASHRAE Transactions. American Society of Heating. Refrigerating and Air Condit ioning Entfineers. Atlanta. GA. Volume 72. Part 1. 1966. nn. 180-189.

Appendix A Explanatory Material

Appendix A is not apart of the recommendations of this NFPA document but is included for informational purposes only. This appendix contains explanatory material, numbered to correspond with the applicable text paragraphs.

A-I-4 Approved. The National Fire Protect ion Association does no t approve, inspect, or certify any installations, procedures, equipment , or materials; nor does it approve or evaluate testing laboratories. In de te rmin ing the acceptability of installations, procedures, equipment , or materials, the authority having jurisdict ion may base acceptance on compliance with NFPA or o ther appropr ia te standards. In the absence of such standards, said authority may require evidence of proper installation, procedure , or use. The authority having jurisdict ion may also refer to the listings or labeling practices of an organization that is conce rned with product evaluations and is thus in a posit ion to de te rmine compliance with appropria te standards for the current product ion of listed items. ~ . ~

A-l-4 Authority Having Jurisdiction. The phrase " a u t h g , h a v i ~ jurisdict ion" is used in NFPA documents in a b r o a d ~ " , jurisdict ions and approval agencies vary, as do__their~ t e ~ ~ ~ responsibilities. Where public safety is primary, tt}.~.~ thol " ~ having jurisdict ion may be a federal, state, loc,~l~ ~ ~" depa r tmen t or individual such as a fire chief; #i% • ~ a tire prevent ion bureau, labor d e p a r t m e n ~ . . h e : ~ t " building official; electrical inspector; or o m e ~ , f i r start ~ry authority. For insurance purposes, an msuran~ir~. , :ctio ! depar tment , rating bureau, or o ther insurance : d i ~ ( q :~. representative may be the authority havingjurisdi( ?..~.~ many circumstances, the proper ty owner or his or her desi ti~ted agent assumes the role of the authority having jurisdiction; : government installations, the commandin~ officer or depar tmenta l official may be the authority having jurisdiction.

A-l-~i,3 Airflow can be used to limit smoke migration when doors in smoke-control barriers are ooen. The desima velocity through an onen door should be sufficient to limit smoke back_flow dur ing building evacuation. It should take into considerat ion the same variables as used in the selection of design nressure differences. Design informat ion is orovided in A S H I ~ / S F P E . Desiun of Smoke Management S~stems.

While airflow can be used to inhibit smoke movement through a space, the flow rates needed to nrevent smoke backflow are so large_ that there is concern about the amount of combust ion air that is sunnlied to the fire. When airflow is used to manage smoke movement, the flow of air th rough the onenin~ in to the smoke zone must be of sufficient velocity to ore-vent smoke from leaving th~,l; zgne through such ooenings. The air velocity necessary to inhibit smoke movement t h rough lartee onenings results in air ouantities which are sufficient to sunoort-fire growth to approximately 10 t imes the size comoared to fire growth without this addit ional airflow. More informat ion on fire ~rowth can be found in SFPE Fire Protection Handbook.

A-1-6.3 One source of data is ASHRAE Handbook of Fundamm~tals. Chapter 26. Climatic Desitna Information. It is suggested that the 99,6 percent heat ing dry bulb (DB) temnera ture and the 0.4 per¢¢tat cooling DB-temperature be used as the winter and summer 4¢siwn condition, resnectivelv. It is also suggested that the 1 percen t extreme wind velocity be used as the design condition. Where available, more site-soecific data should be consulted.

A-l-7 The oer formance obiective of automatic sorinlders installed in accordance with NFPA 1-3. Standard for the ln;tallation of Sprinkler S~stems. is to orovide fire control, which is def ined as follows; Limiting the size of a fire bv distribution of water so as to decrease the hea-t release rate and 0re-wet adjacent combustibles. while controll ing ceiling gas temneratures_ to avoid structural damage. A limited number of investiwatiom have been under taken in which full-scale fire tests were conduc ted in which the snrinlder system was chal lenged but nrovided the exnected level of 0erformance (Mad~zvkowsl~i. 1992 and Lougheed et al.. 1994L These investigations indicate that. for a fire control situation, the heat release rate is limited but smoke can cont inue to be produced. However. the temnera ture of the smoke is reduced and the oressure differences nrovi-ded in this d o c u m e n t for smoke

~2-2 .1 the "" of" ~

obtain

or wind.

, ~ should be des igned to maintain differences under likely conditions

min imum design pressure differences le red spaces are values that will not be :s o f ho t gases. The me thod used to -2.1 for nonspr inklered spaces is are difference due to buoyancy of ho t Ilowing equation:

1 1 A P = 7 .64 h

To Tr

A P = pressure difference due to buoyancy of ho t gases (in. w.g.) 7", = absolute tempera ture of surroundings in (°R) T F = absolute tempera ture of ho t gases in (°R) h = distance above neutral plane (ft)

The neutral plane is a horizontal plane between the fire space and a sur rounding space at which the pressure difference between the fire space and the sur rounding space is zero. For Table 2-2.1, h was conservatively selected at two-thirds of the floor to ceiling height, the tempera ture of the surroundings was selected at 70°F (20°C), the temperature of the ho t ~ases was selected at 1700°F (927°C), and a safety factor o f 0.03 m. w.g. (7.5 Pa)was used.

For example, calculate the mi n i mum design pressure difference for a ceiling height of 12 ft as follows:

7", = 70 + 460 = 530 °R Tp = 1700 + 460 = 2160 °R h = (12) 2/5 = 8 ft

From the above equation, A P = 0.087 in. w.g . Adding the safety factor and rounding off, the min imum design pressure difference is 0.12 in. w.g.

A-2-2.2 The forces on a door in a smoke control system are illustrated in Fitrure A-2-2.2. The force reouired to onen a door in a smoke control system is

where:

F = F ~ + 5 . Z ( W A ) A P

2 ( W - d)

F = total door onening force fib) F = force to overcome the door closer and other friction flb~

6O4

N F P A 9 2 A ~ M A Y 2 0 0 0 R O P

W = door width fit) A = door area (sq fO AP = nressure difference across the door fin. w.g.I d = distance f rom the doorknob to the knob s l deo f the door fit)

F Low-pressu re I side , Knob/

- ~ i ~ / / l l / l l l l l l / l l / l l l l A ~ . : ; _ : : : : . N . . : : ~ : . : . : . ~ : : : . $ N y . ~ : . : . . . - ~ x

t a ~ p ~ High-pressure

Hinge Z A side

F i g u r e A - 2 - 2 . 2 F o r c e s o n a D o o r i n a S m o k e C o n t r o l S y s t e m .

A-2-3.7 During the time that occuvants of the smoke zone ar¢

A-3-4.2.2 The control system should be designed as simply as possible to attain the required functionality. Complex controls, if not properly designed and tested, can have a low level of reliability and can be difficult to maintain.

Controls for nonsmoke control imrOoses. Manoal con~'ols exclusively for other building control ournoses, such as Hand-Off- Auto switches located on a thermostat, are not considered to I~e manual controls in the context of smoke control. Manual activation and deactivation for smoke control nurnoses should override manual controls for other ournoses.

Manual troll stations• Generally. stairwell pressurization systems can be activated f rom a manual null station, vrovided the resvonse is common for all zones• Other-systems that-resnond identically for all zone alarms can also be activated from a mammal pull station. An active-tracking stairwell pressurizatioa ~ystem that nrovides control based o n t h e pressure measured at the fire floor should not be activated f rom a manual null station,

.4.-3-4.3.3 Activation of the smoke-control system should occur immediately after receint of the activation command, In order to

exiting the area. the conditions in the smoke zone are still nrevent damage to etminment, it might be necessary ~o delay tenna[ale. Although onening the stairwell door on the fire floor activation of c e r t a i m " ~ v m e n t unti l o ther euuinment has achieved during this time may release some smoke into the stairwell, it will a vrereuuisite s t a~ : t i . e f~e lav starting a fan until its assodated no t create untennable conditions in the stairwell. Once conditions damuer-is n a r ~ . ' ~ f u l l v ot~enL Tl~e times ~iven for comnonent,~ in the smoke zone become untennable, it is unlikely that the door to achieve ~ d e s l ~ s t a t e - a r e measured f rom the time each

" ~" ":÷" • "~Et"."~. to the fire floor would be onened bv occunants of that floor. Fgr comnonea~-'.,_~ z~- vate~.}-.%~ .,, this reason, it is normally not required to clesign for an open A - ":''::if+ ...... ~ . . . . "~'z'~*~" stairwell door on the fire floor. Doors blocked oven in violation of , ~ . 4 ~ m p l e of a E..~ Fighters' Smoke-Control Station.

C aU for a fire Iighters smoke control stauon should annlicable codes are beyond the capability of the-system. ~ ~ ' - "

The imnortance of the.. exterior stairwell door can be exulained 19Y "a" L o ~ : " Access - " ~C,S . . . . . . . . • - • • .- . . ~:~.'~ t ) m a r ~ . . . ~ . a n e r~ snoum De mcatea m considecauon of the conservauon of mass of the pressunzauon air, ~ . ~ . ; ~ m i w to l : ~ r fire fighters' systems as can be provided within This air comes f rom the outside and must eventually flow back to % ~'~¢~.~[.in.~:~I.eans should be provided to ensure only authorized the outside. For an open interior door. the rest of the building 9n ~ . a c ~ ' ~ . ~ " FSCS. Where acceptable to the authority having that floor acts as flow resistance to the air flowing out the ooCn-,.:>.~. ~ i : i s d i c ~ " h , the FSCS should be provided within a specific location doorway. When the exterior door is open. there-is no o t h ~ ~ room, s ep~a ted from pub!ic areas by a suitably marked and

~, c~ ooor wnere locatecl in a separate room, the room resistance, and the flow can be 10 to 30 times more thart+."~rou~h~ " . . . ~ . • . . . . . . . . a n o n e n i n t e r i n r d n n r ~x{~. . :~}. . - ~ . ~ . ~ \ l ~ t a o n , size, access means anta omer pnysica~ uesign - - ° ~ - " . . . . . . . . . . . . . . . . . . . " ':~::::~::~ -'::."':~>~:-q~- .:~; ~i '~" . . . . '.:$..'$~:~,.:.-~- "~:~.-..~.'~onslderataons should be acceptable to the authority haxan~: .~...~.~. ,:g$~?,.,. . . - -

%':'::':~"-'::- " '=+:"" j u r l s d l c t l o n . A-2-4.1(5) Rule 211.3a, Phase I Emergency R e c a l ~ t i o n ~ . ~ ASME/ANSI A17.1, Safety Code for Elevators a n ~ . . ~ g e q u + ~ :': tb~ Physical Arran,,ement Th,~ ~SO-q sh,,,,~a h~ a ~ ; ~ , ~ a , , that elevator doors open and remain open . a ~ r the e l e ~ are .~6-'." granhicallv depict ~ ' ~ ' - : - ~ " . . the ~h" slcal build: . . . . an . . . . . . ['1 ~ l " ~-~ y I l l ; t i l l ¢711 I ¢ I I L , recalled. This results in large openings i n . , ~ elevator ' ~ t w a ~ . - - , - , . . . . . "7 "~ . . . g .

• trreatlv incroag~ the a l r f l . w r e n n ~ d ~ n r e . ~ n r i ¢ ~ n smote-control systems ann equipment, anct m e areas 9.LRIe This can o . . . . ~ . . . . . . . . . . . . . . . . . . . . . . . eta r~.~:i . . . . . . . ~ . . . . . . NFPA 80, Standard [or $ire Doors and Fire W / n d o ~ r m i t s ~ o s i n g ~ served by the equipment. Following is a summary of the of elevator doors aider a . . . . nredetermined time w h e n < ~ i r . g ~ t ~'" by, the status indicators and smoke-control, capability applicable to the authority having jurisdiction. Local requirements ~ r a t i o n of FSCS smoke-control graphic(s) . elevator doors should be determined and incorporat .~-into the system design. Status indicators should be provided for all smoke-control

equipment by pilot lamp-type indicators as follows: A-2-4.3 The following references discuss research concerning elevator use during fire situations: Klote and Braun (1996); I~lote ( 1 9 9 5 ) : Klote. Levin and Groner (1995): KIote• Levin and Gror~er (1994): KIote (1993): Klote. Deal. Donoghue. Levin and 'Grone r (1992): and Klote. Alvord, Levin and Groncr (1992).

A - 2 - 5 . 2 . $ Design guidarlce on dilution temoerature is provided in ASHRAE/SFPE. /)es/gn of Smoke Management S3stems,

A-2-6 Methods of desimi for areas of refuge are presented in the ASHRAE Transact ions-raper Desien of Smoke Control S~stems for Areas of Refu~,e (Klote 1993).

(1) Smoke-control fans and other critical operating equipment in the operating state: Green

(2) Smoke-control equipment and other critical equipment that may have two or more states or positions, such as dampers: Green (i.e., OPEN), Yellow (i.e., CLOSED)

The position of each piece of equipment should be indicated by lamps and appropriate legends. Intermediate positions (i.e., modulating dampers that may not be fully open or fully closed) can be indicated by not illuminating either of their pilot lamps.

(3) Smoke-control system or equipment faults: A m b e r / O r a n g e

605

N F P A 9 2 A - - M A Y 2 0 0 0 R O P

Use-o~ The posit ions of mult i -posi t ion control switches shou ld no t be used to indicate the status of a control led device z.~.eul~ : a t be--used~ in lieu of pilot l amp- type status indicators as descr ibed in A-3-4.3.4(1) t h r o u g h (3) above.

Provision for test ing the pilot lamps on the FSC,S smoke-control panel(s) by m e a n s o f one or more "LAMP TEST" m o m e n t a r y push bu t tons or o the r self-restoring m e a n s shou ld be included.

(c) Smoke-Control Capability. T h e FSCS shou ld provide control capability over all smoke-control system e q u i p m e n t or zones within the b u i l d i n ~

Wherever nractical, it is r e c o m m e n d e d that control be provided by zone. ra ther t han by individual equ ipmen t . This app roach will aid fire f ighters in readily u n d e r s t a n d i n g the ooerat ion of the system, and will he ln to avoid p r o b l e m s causeci bv manua l ly activating e o u i o m e n t in the wrong seouence or neglec t ing tO control a critic~al c o m n o n e n t . Control by zone shou ld be a c c o m n l i s h e d with PRESSURIZE-AUTO'EXHAUST control: control over each zone tha t can be control led as a, ~ingle entity, Control o f this tvoe relies on system p r o g r a m m i n g to proper ly seuuence all dex;ices in the zone to n roduce the des i red effect. In systems util izing c o m m o n suoolv a n d / o r re turn ducts, inclus ion of an ISOLATE mode is desireable. To enable use of the system I;9 f lush smoke ou t of a zone after the fire has been ext inguished, a PURGE (eoual sunoiv and exhaus t ) m o d e may also be desireable.

Where control over individual nieces of e o u i n m e n t is d e e m e d necessary, the following control on t ions shou ld be provide~;

(1) ON-AUTO-OFF control over each individual piece of opera t ing smoke-contro l e q u i p m e n t tha t can also be control led f rom o ther sources within the building. Control led c o m p o n e n t s inc lude all stairway pressur izat ion fans; smoke exhaus t fans; HVAC supply, re turn, a n d exhaus t fans in excess of 2000 ftSmin (57 m / m i n ) ; elevator shaft fans; a t r i um supply and exhaus t fans; a n d any o ther opera t ing e q u i p m e n t used or i n t ended for smoke-co ~ol purposes . ~ :'-".~

(2) ON-OFF or OPEN-CLOSE control over all s m o g . :*" .. :!. and o ther critical e q u i p m e n t associated with a fire or ~ , , ~ , ~ ~':g"" emergency a n d tha t can only be control led f rom the FSCS

~ : . : . : : : : : : : : : - : . ~ . . ~:.-:.

(3) OPEN-AUTO-CLOSE control over all i ~ ,,~.~:...::, relat ing to smoke control and that are also ~ t r o l l e d f r~ r#;" sources within the building. *:" "::~'~::" %!i"

(4) HVAC termina l units, such as VAV m i x i n f f ~ . e s th~}~are all located within a n d serve one des igna ted s m o k e - c o ~ . ~ t $ e , can be control led collectively in lieu o f individually. ~ t n i t coil face bypass dampe r s tha t are so a r r anged as n o t to r e ~ J c t overall airflow within the system can be exempted .

Addit ional controls m i gh t be requi red by the author i ty having ju r i sd ic t ion .

(d) Control Action and Priorities. T h e FSCS control act ion shou ld be as follows:

(1) ON-OFF, OPEN-CLOSE. These control act ions shou ld have the h ighes t priority of any control po in t within the building. Once issued f rom the FSCS, no au tomat ic or m a n u a l control f rom any o ther control po in t within the bui ld ing shou ld contradic t the FSCS control action.

Where au tomat ic m e a n s are provided to in te r rup t no rma l n o n e m e r g e n c y e q u i p m e n t opera t ion or p roduce a specific result to safeguard the bui ld ing or e q u i p m e n t (e.g., duc t freezestats, duc t smoke detectors, h igh- t empera tu re cutouts, t empera tu re -ac tua ted linkage, a n d similar devices), such m e a n s shou ld be capable of be ing overr idden or reset to levels no t exceeding levels of i m m i n e n t system failure, by the FSC, S control action, and the last control act ion as indicated by each FSCS switch posit ion shou ld prevail.

Control act ions issued f rom the FSCS shou ld not 0vgrrid¢ or bypass devices and controls i n t ended to orotec t against electrical overloads, nrovide for ne r sonne l safetv, and n reven t maior SYSt¢~ damage . These devices inc lude overcur rent orotect ion-devices and

electrical d i sconnec t switches, high-l imit static nressure switches. and combina t ion f i r e / smoke d a m o e r s beyond their dem'adat ion t emnera tu re classifications mee t inb UL 555. Standard for Safet~ Fire Damt~ers. or UL 555S. Standard for Safet~ LeakatTe Rated Dam~e~s for Use in Smoke Control S~stems.

(2) AUTO. Only the A U T O posi t ion of each 3-position FSCS control shou ld allow automat ic or m a n u a l control action f rom other control points within the building. The A U T O posit ion shou ld be the normal , n o n e m e r g e n c y , bui ld ing control position. W h e n an FSCS control is in the A U T O position, the actual status of the device (on, off, open , closed) shou ld con t inue to be indicated by the status indicator(s) .

(3) FSCS Response Time. For purposes of smoke control, the FSCS response t ime shou ld be the same as for au tomat ic or manua l smoke-control act ion ini t iated f rom any o ther bui ld ing control point. (See 3-4.3.3.)

FSCS pilot lamp indicat ion of the actual s tatus of each piece of e q u i p m e n t shou ld n o t exceed 15 seconds after opera t ion of the respective feedback device.

(e) Graphic Depiction. T h e locat ion of all- smoke-contro l systems and e q m p m e n t ~ bui ld ing sh_ould be indica ted by symbols within the overall ~ ~ " " g ranh ic panel .

smoke

~ ~ . ~ j o r ducts , fans. a n d d am p e r s ~lat a re par t ' :~4.he smoke-control system.

for ~ shou ld be shown on the

)anel and, where appropr ia te , shou ld be d to thei r respective ducts, with a clear indicat ion l e ~

,n o f airflow. In e i ther case. the bu i ld ing areas ~erv¢~

system, a n d can be omi t ted where the i r inc lus ion would hint~gl" u n d e r s t a n d i n g of the system, such as on an already densely ooDulated oanel . DamDer nosi t ion indicat ion can also be omit ted where no seDarate control over d a m n e r r)osition is nrovided.

A-3-4.5.2 Manua l controls used exclusively for o ther bqildin~ control ournoses , such as Hand-Off-Auto switches located oq a thermosta t . -are no t cons idered to be m a n u a l controls in the context of smoke control. Manual activation and deactivation for smoke control nu rnoses shou ld override m a n u a l controls for o ther

A-3-4.5.4 Examples of auxiliary func t ions tha t can be useful. I~ t are no t reouired, are the o n e n i n g or closing of te rminal boxes while nressur iz ing or exhaus t ing a smoke zone. These func t ign s are cons idered auxiliary if the c/esired state is achieved without these funct ions. These func t ions can. however, he ld to achieve the desired state more readily.

A-3-4.5.5 Dur ing a fire. it is likely tha t e n o u g h smoke to activate a smoke detector may travel to o ther zones, and subseauen t lv ~ u ~ e alarm inputs for o ther zones. Systems activated bv smoke detectors shou ld con t inue to operate accord ing to the first a larm inpu~ received, ra ther t han divert ing controls to r e spond to any s u b s e o u e n t a la rm innut ( s ) .

Systems initiated bv heat-activated devices, and des igned with su~:ficient canacitv to exhaus t mul t io le zones, can ex o an d the n u m b e r of zones be ing exhaus ted to include the original zor~e and subseouen t addit ional zones, uo to the limit of the mechanica l system's ability to main ta in the design oressure difference.

v _

Exceeding the design canacitv will likely result in the system failing y

to adeouate lv exhaus t the fire zone so as to achieve the desired

606

N F P A 92A - - MAY 2000 R O P

pressure differences. If the n u m b e r o f zones tha t may be exhaus ted while still ma in ta in ing the design nressure is no t known, this nurqber shou ld be ass~tlled to be one.

A-3-4.6 Verification devices can inc lude the following:

(1) End-to-end verification of the wiring, eou inmen t , and devices in a m a n n e r that includes nrovision for oositive conf i rma t ion of activation, periodic testing, and manua l -over r ide ooerat ion

(~) The preserlc¢ of ,?perating power downst ream of all circuit d i sconnec t s

(3) Positive conf i rmat ion of fan activation bv m e a n s of duc t .°ressure- airflow, or equivalent sensors that r e snond to loss o f onera t ing power, probleg0s in the nower or control circuit wiring. airflow restrictions, and failure of the belt, shaf t counl ing, or moto r

(4) Positive conf i rmat ion of d a m n e r onera t ion bv contact. proximity, or euuivalent sensors that r e snond to loss of onera t ing power or c o m p r e ~ g d air: p rob lems in the nower, control circuit. or oneuma t i c lines: and failure of the d a m n e r actuator, linkage, or d a m n e r itself

(~) Periodic acceotance test ing in accordance with Chan te r 5

(6) O t h e r devices or m e a n s as annronr ia te

It¢r0s (1) t h r o u g h (6) describe mul t io le m e t h o d s tha t can be usgd, e i ther slnglv or it~ combinat ion , to verify that all oort ions of the controls a n d eouip lBent are onerat ional . For examnle . convent ional (electrical) sunervls ion can be used to ver{fv the integrity of the conductors f rom a fire a larm system control uni t to the relay contact within $ feet o f the control system i n n u t (See NFPA 72. National Fire Alarm Code °. Section 3-9). a n d end- to-end

off-normal condi t ions Qn its resnective segment .

position) is achieved. True end- to-end verification, therefore. reouires a compar i son 9f the desired onera t ion to the actual end resu l t .

An open control circuit, failure o f a fan belt. d i sconnect ion of a shaft coupling, blockage o f an air filter, failure of a motor , or o ther abnormal condi t ion tha t could n reven t n r o n e r onerat ion, is no t e~;pected to resul t in an off-non;hal indicat ion wi~en the control led device is no t activated, s ince the measu red resul t at tha t t ime matches the exnec ted result. If a condi t ion tha t nrevents n rone r onerat ion nersists dur in~ the nex t a t t emn ted activation o f t h e - devjt;e, an off-normal indicat ion s h o u l d b e disnlaved.

Ao4-3 Over the nast th ree decades, several network c o m n u t e r mode l s have been written to calculate steady state airflow-and oressures t h r o u g h o u t a building. At one t ime. ASCOS (See ASHRAE/SFPE. Desio'n of Smoke Mana~,ement S~stems) was the most wjd¢ly used mode l for ,smoke control ~nalvsis.. and it has been validated against field data f rom flow exner imen t s at an eight-story tower in C h a m n s Sur Marne. France (Klote and Bodart 1985).

Wray and Yuill (1993) ¢valuated several flow a lgor i thms to f ind the mos t annronr ia t e one for analvsis o f smoke control systems. T h e best al~oritlam f rom this s t u d y b a s e d on comnuta t iona l sneed a n d use of comou te r m e m o r y has been incornora ted in the CONTAM

c o m n u t e r mode l develoned bv Wal ton (1997). CONTAM is a sitmi'ficant i m n r o v e m e n t over ASCOS with resoect to bo th numer ics

Network c o m n u t e r mode l s shou ld be used for the design of smoke-control systems in comnlex bui ldings for which tlae algebraic equat ions are no t applicable or are imnractical to use. This inc ludes the analysis of stairwell nressur izat ion systems with onen doors, combina t ion smoke-control systems a n d smoke- control systems in asymmetr ic buildings.

A-4-5 Leakage areas for exterior bui ld ing walls have tvoicallv been based on the m e a s u r e m e n t s of T a m u r a an-d Shaw (1975) an d Tamur~ and Wilson (1966). Recently. several bui ldings used in the p r e v i o ~ studies were retested after they were retrofi t ted for energy efficiency (Shaw. Reardon and C h e u n g 199$). The values for leakage areas of exter ior bui ld ing walls were based on this new

A-5-1,1 Door open ing forces inc lude frictional forces, the forces o roduced bv the door hardware, a n d the forces n r o d u c e d by the smoke-control system. ' In cases where fritional f~rces are excessive.

A-4 ~-1.2 Whil test ing of b u ~ smoke z ~ ; ' ~ test ing

,art o f the formal test ing procedure , the ~..~letermine the a m o u n t of leakage between ~ u e in developing the initial system. The s r : ~ p . . ~ n use existing airflow m easu r in g

t h ~ m s . This sect ion describes the of ~ i e t y of systems and test ing m e t h o d s le te~mining the leakage of enclosures. comes f rom a variety of sources, such as the

cons t ruc t ion where leakage paths can be fo rm ed surface and the floor slab

~ralt par t i t ions where gaps in the drywall beh ind cover can fo rm leakage pa ths

Electric switches and oudets in drywall part i t ions that fo rm ,~e pa ths t h r o u g h the part i t ions

(4) Installation of doors with undercu ts , la tching mechan isms , and o ther gaps fo rming leakage pa ths

(5) Interface of drywall part i t ions at f lu ted metal deck requi r ing seals in the flute

(6) Electric outlets in f loor slabs within the space or above the space and providing leakage to o ther floors of the bui ld ing

(7) Duct pene t ra t ions t h rough walls where there can be leakage a round the duc t beh ind angles tha t hold fire dampe r s in place

(8) Per imeter induct ion systems tha t of ten have gaps a r o u n d ducts t h r o u g h floor slabs tha t are h idden beh ind air distr ibution enc losures

(9) Pipe, condui t , and wireway pene t ra t ions t h rough walls a n d floors requi r ing listed th rough-pene t ra t ion seals

Building. HVAC Systems Suitable for Enclosure Tightness Testing. Many bui ld ing HVAC systems can be used to measu re the leakage t h r o u g h enclosures. These systems typically contain a central fan that can draw large quant i t ies of outside air into the bui ld ing for pressurizing. Because all of these systems conta in openings , ductwork, a n d somet imes fans to re tu rn the air f rom the enclosure to the central air handler , it is impor tan t tha t these systems be shu t off du r ing the test. The use of smoke dampe r s at the points where the ducts leave the enclosure will give m o r e assurance tha t leakage f rom the space t h r o u g h this source will be minimized.

(a) Single-Floor VAV Systems. Many m o d e r n office bui ldings are provided with a separate air hand le r on each floor of the bui lding to supply condi t ioned air to the space. These systems are a r ranged as variable vo lume systems, whereby the thermosta ts vary the a m o u n t of air delivered to the space ra ther t han the t empera tu re of tha t air. This requires a variable f requency control ler on the fan

607

N F P A 92A - - MAY 2000 R O P

that r esponds to pressure in the duc t system. As the variable vo lume control device is closed, the pressure bui lds u p in the duc t a n d the fan speed is slowed in response to tha t pressure. Normally these systems conta in a i r -measur ing devices in the supply a n d re turn ducts tha t are used to synchronize the re tu rn fan opera t ion with the supply fan, so a cons tan t quant i ty of outs ide air can be in t roduced into the space to main ta in indoor air quality. These airflow m e a s u r i n g devices can be used to measu re the airflow in t roduced into the space, and the speed of the fan can be adjusted to control the pressure across the enc losure barriers.

(b) Centxal Fan VAV Systems. Central fan VAV systems are a variation of the single-floor VAV system. A single fan will supply 10 or more floors, each of which will have a n u m b e r o f variable volume boxes. As in the case of the single-floor system, the fan will r e spond to a pressure sensor in the duct. The re will be a flow- measu r ing stat ion at the fan that is used to track the re turn fan with the supply fan in order to main ta in cons tan t outside air, as in the case o f the single-floor VAV system. Generally, these systems will also be provided with a motor -opera ted shut-off d a m p e r at each floor as the system can be economical ly used to supply only a por t ion of the floors when o ther floors are vacant.

These systems can be used for test ing of spaces by c o m m a n d i n g that all of the supply damper s to the floors be closed except on the floor being tested. In this manne r , the airflow onto the f loor can be measu red as the pressure across the barriers is adjusted.

The leakage characterist ics o f the ma i n duc t system as well as those of the d a m p e r s tha t are to be shu t mus t be known so the correct ions for duc t a n d d a m p e r leakage in the system of the floor u n d e r test can be d e t e r m i n e d ahead o f t ime. This can be accompl i shed by shu t t ing all t he d a m p e r s on the system, pressur iz ing the duc t system to various pressures us ing the supply fans, a n d measu r ing the airflow at the a i r -measur ing s tat ion in the supply duct.

O n e variation of mult i f loor VAV system that can be e n c o u n t e r e d will have a i r -measur ing stat ions on each floor of the building. This is done to verify tha t a part icular t e n a n t is no t creat ing so m u c h load on the floor tha t more airflow is used than is des igned I~t~..->,, the system. W h e n this is encoun te red , the airflow can b ~ ~ ) ar'~i "~ directly on the floor so tha t ad jus tments for main duc t 1 unnecessary. ~i::":m":' ~ > : i { i

(c) Cons tan t Volume Mult izone Systems. C o n s ~ volun mul t izone systems mix ho t and cold air at a c e r ~ ~ ( m~"~'~i~ii~ ~ uni t a n d has a separate duc t system tha t goes . .~ ' t to v ~ ~ / ~ pac Typically, they are no t provided with a i r - m ~ i n g s t a t i o ' ~ nat would have to be retrofi t ted to the ducts d ~ V O ~ air to ff'~ i spaces. T h e spaces need to coincide with the ~ s u r e s b~hag tested. The re is also, typically, no m e a n s of v a r y i ~ . ~ e f i ~ ' t o each space. Varying the flow requires the addi t ion ~ i ~ . ~ : r manua l or motor ized dampe r s in the duc t system tha~i~e adjusted to achieve the test pressure or pressures. :~"

(d) Cons tan t Volume Termina l Rehea t System. Cons tan t vo lume terminal r ehea t systems are the mos t difficult to use for test ing for enclosure t ightness. Typically, these systems contain central fans that deliver air to a duc t system at a set t empera ture . T h e duc t system is dis t r ibuted t h r o u g h o u t the bu i ld ing and r ehea t ed coils are placed at various locations to t e m p e r the air to main ta in space condit ions. The re are typically no m e a s u r i n g stat ions or any au tomat ic dampe r s in the system. In order to use this system for testing, it is first necessary to retrofit it with a i r -measur ing stat ions a n d dampe r s to coincide with the enclosures be ing tested.

Building HVAC Systems Not Suitable for Enclosure Tightness Testing. The re are a n u m b e r of HVAC systems that have little or no value m tes t ing the t ightness of an enclosure because they in t roduce a l imited a m o u n t o f airflow into the space or are a r r anged so tha t the re are mult iple duc t en t rances into the space. Therefore , mak ing airflow m e a s u r e m e n t in such systems is impractical. T h e s u m m a r y of these systems is as follows:

(a) Uni tary Hea t P u m p / F a n Coil Systems. Uni tary hea t m p / f a n coil systems come in a n u m b e r o f configurat ions. ese systems are similar in the fact tha t the space ts provided with

a n u m b e r o f separa te units , each of which has l imited airflow capacity. Outs ide air to the space is in t roduced in one o f the following th ree manners :

(1) Uni ts are located on the pe r ime te r with a separate outside air duc t for each unit . This typically has a small pene t ra t ion t h r o u g h the outside wall of the bu i ld ing with no ductwork a t tached. The a m o u n t of outs ide air in t roduced is so small and the capacity of the systems to pressurize the space is so l imited that the systems canno t be used for tes t ing the integrity of the space. In these instances, the units will be de t r imenta l to the operat ion of any system in the space des igned to pressurize it unless each outside air duc t is fitted with a t ight.closing au tomat ic damper .

(2) Uni ts are located only on the pe r ime te r a n d outside air is in t roduced t h r o u g h a separate duc t system. In this instance, the uni ts are used in conjunct ion with an inter ior duc t system. The outside air duc t for the pe r imete r is of l imited capacity a n d should be fi t ted with t ight.closing au tomat ic d a m p e r s to main ta in the integrity o f the enclosure. Tes t ing of the space shou ld be done t h rough the inter ior duc t system.

(3) Units are dis t r ibuted t h r o u g h o u t both the per imeter a n d interior. In this instance, outs ide air is in t roduced into the space t h rough a separate duc t system that distr ibutes t h r o u g h o u t the entire floor area. This duc t system is sized to handle the m i n i m u m outside air quanti t ies n e e d e d in the space a n d mig h t or migh t no t have sufficient flow to provide pressure in the space. Wh e th e r or no t this system can bc..used for the pressure test ing m u s t be

a case ...~..]~e basis. It will be necessary to provide the a i r - ~ u r i ~ : . s t a d o n s and possibly shut-off damper s if

dec ided on a case .t system with i r - ~ the system se~.. .~: ."~tipl

.3-.':-~: "-'.~..'.'.~ ~'....

(b) P e r i . ~ I n ~ a r e

~tems~are arrang are typi~ Thes~...~ p e ~ t e r tt.'.B.der the win d u ~ i ~ a . c ~ . ~ , air dish s ~ ¥ ' ~ ' O ~ : ' i n. ~ (1~. floor to ~ V ~ b u t i o n sy~ vertical r i s d C f i : - ~ x t e n&

.~...n Systems. Pe r imete r induc t ion systems ~ . ~ . . the per imeter of the bu i ld ing only. g ' ~ . . ~ t h a te rminal un i t a long the

w i n q ~ , each of which is provided with a air dis t r ibut ion system. T h e ducts typically are

(129 cm ~) per unit] a n d e i ther penet ra te the ~ b u B o n system on the f loor below or co n n ec t to a ~ x t e n d s u p t h r o u g h the bui ld ing a n d supplies

:o six u ~ e r floor. "These systems do n o t l en d themselves to • :~0f. s p i e s because of the mul t ip le duc t connec t ions on each ~ i ~ u c t connec t ions shou ld be provided with t ight.closing a a ~ dampe r s so pressurizat ion o f the space will be possible.

re is general ly an interior system provided, which is one of the previously described, that can be used for the test ing an d

A-5-$.3.6 The test m e t h o d s descr ibed in Chan te r 5 shou ld Drovide an adeouate means to evaluate the smoke-control system's oer formance . O the r test m e t h o d s have been used historically in instances where the author i ty hav ing iur i sd ic t ion requires addi t ional testing. These test m e t h o d s have l imited value iB evaluat ing certain system oef fo rmance , a n d their validiw as m e t h o d s o f t e s t i n ¢ a smoke.control system is ouest ionable . Examples o f o ther test m e t h o d s tha t have been used are as follows:

(1) Chemical smoke tests

(2) Tracer gas tests

(3) Real f ire tests

A A ~ ~ oz^x ,~L__ : . . , e _ _ , . _ ,r . . . . Chemical smoke tests have achieved a degree of popular i ty out o f p ropor t ion to the l imited in format ion they are capable of providing. The m o s t c o m m o n sources of chemical smoke are the commercia l ly available "smoke candle" ( somet imes called a smoke bomb) a n d t h e smoke gene ra to r apparatus . In this test, the smoke candle is usually placed in a meta l conta iner and ignited. T h e metal con ta iner is for protect ion f rom hea t damage after ignit ion; it does n o t inhibi t observation of the m o v e m e n t o f the chemical smoke. Care needs to be exercised du r ing observations, because inhala t ion o f chemical smoke can cause nausea .

This type o f test ing is less realistic t han real fire test ing because chemical smoke is cold a n d lacks the buoyancy of smoke f rom a f l aming fire. Such buoyancy forces can be sufficiently large to overpower a smoke-control system tha t was no t des igned to wi ths tand them. Smoke f rom a spr inklered fire has little buoyancy, and so it may be expected tha t such smoke m o v e m e n t is similar to the m o v e m e n t o f u n h e a t e d chemical smoke. This has n o t yet been confwmed by test data. Chemical smoke test ing can identify

6O8

NFPA 92A - - MAY 2000 ROP

leakage paths, and such tests are s imple and inexpensive to perform.

The quest ion arises as to what informat ion can be obtained from a cold chemical smoke test. If a smoke-control system does not achieve a h igh e n o u g h level of pressurization, the pressures due to hot, buoyant smoke could overcome that system. The ability to control cold chemical smoke provides no assurance of the ability to control hot smoke in the event of a real fire.

Chemical smoke is also used to evaluate the effectiveness of so- called smoke "purging" systems. Even though such systems are not smoke-control systems, they are closely related and so they are briefly addressed here. For example, consider a system that has six air changes per hour when in the smoke purge mode . Some testing officials have mistaken this to mean that the air is completely changed every 10 minutes, and so 10 minutes after the smoke candle is out, all the smoke should be gone from the space. Of course, this is no t what happens. In a purging system, the air entering the space mixes to some extent v~th the air and smoke in the space, ff the purging system is part of the HVAC system, it has been des igned to promote a rather complete degree of mixing. If the concentrat ion of smoke is close to un i form within the space, then the method of analysis for purging presented in Section 2.3 of ASHRAE/SFPE, Design of Smoke Manatg.ement Systems, is appropriate. Based on such perfect m~xing, after 10 minutes, $7 percent of t h e o r i g i n a l smoke remains in the space.

? . ~- ~ . ~ . 2 ~ ) T . - - c c : C = : T ~ : = . B c c = u : : o f uhe man)" p c z z , h , ! , d c : ~ f . . . . . . . a .~.^ , : - - : . ^ . : . . . . . ~: . . . . . . . "~ ' ~ ^ ' . . . . . . . . . . . . " - - ~ : h c u i d hc conducted ~nd c;~'2uated :;~th : h, gh level of prafc='an=1

In "..heat t c : = , : t u n : r a n t tic'.-" ~ t e c f = t r a c e r - ~ , : r c ! c = c ~ in t h e k . . l l A : - - - - i ~ - - ~ l . . . . . . . I I ^ . ~ ^ A ~ k . . . . . ~ . ~ . . ~ + ~ - L - . . . : I . I : . . . . .41 - - . .

~ - - I . - - - - A + ^ A - - . - - . - - : . . . . . . . : L . I - - - - - . ~ ^ ^ £ . - - ~ 1 . . . . . . . . . . ~ : - - . L ^

at*~'ib'dtc: r c z u l t ~n tczta t h a t do n e t , n t c r f c r c - ' . "~ "~e horn ' . ̂..2

. . . . . . . ] . . . . . " . . . . . . . . . . . . . l . . . . . .

. . . . . . . . . . • * ~ = ,)t.~., o^_,______.._ _ _ _ _ ~ ' - - ~ ^ _ - I t i c ~ a n ~ e ~ : a t c : n c n t t a :a}.__-'-^-

2F;E__--U_227;22~-..'FU~Y:;~_ . . . . . . . . . . . . . . . . . . . . . ~ . . . . . . .

. . . . . . . . O ' ~ . . . . . . . l . . . . . . . . . . . . , " ~ l " . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . ! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ! . . . . . .

. . . . . . I A . . . . . I .

h=zdened" tc. prc:'cnt :tr-acrar-" d^_-n=gc. The fire load ccnz!ztcd cf

. . . . . . . . . . . . . . 1 " . . . . . . . l - " 1 ' ' 1 " . . . . . . . . I ~ . . . . . ~ . . . . . . . 7 . , l i .

. . . . . I . . . . . . # ? [ C t O ' l ~ ' l A . ' % r t o p I g . . . . . . . ¢ ) ~ - - : - - . . + . . . . - . I ~ r ~ A . . . . ~ I | . . . .

A ^ . . . = t ~ ^ A - - k . . . . . . . ~ - - ~ I : . ~ : . .~. large ---' c . - ,:,._ . u . . . . - ........ -: . ~ . -

. . . . . . . . . . . . . . . . . . . . ~ V ~ . . . . . . . . . .

(2 to. !9 - -s , - - : -x ~. . . . . h c : = : : cd . . . . . / . . . . . . / . . . . .

f o r =~ l ,h rad~r . ~ f ~ c g: . : c ~ = : : n . a : a ~ r a ~ h . C== : h r a m = . t ~ p ~ :

~;,7, ,~2"~. ;.~..T~, ";.~"3. L';~'.'5~7.2y_".~.21 ""~Z.'.2,.~\'2"'=2~Z.'=~_F"~,~ -

. . . . . . . . . . . . . . . . . . . . . . . . ~ - ' " ~'--" .^ . . .~2.2TK :FL2U.23Y2 . . . . . . . . . . - " . . I . ^ 1 . ~ . : ^ ~ ^ £ . 1 . . _ I ^ ~ . . . . ~ ^ I - . .

. . m._ a . . . . . . . . . . . . . t h :_:. : ~ = = , e . . . . . . . . . . 6 . . . . ~ u ~

~ n d SF~ m , x t u r e c a = b.: h e a t e d , h u t c a u t i o n a h c u ! d ~e e x e r t , t e d k . . . . . . . at h igh . . . . . . . . . . . . . . e~= M_ a . . . . . . . . . : . . . . . . . : . ce .m. .panenm. ? = ::-xh c h e m i c = ! a n . e k e , ~ . . . . . . . . . . . . . . . . . . . . . . . . . . .

' (a~ For the differential pressure test. the open doors should inc lude tho~e for vdaich the highest pressure -d i f f cr~ce was measured ill the tests with all doors closed (see 5-3.4,flL When measured with the stairwell as the reference, as described in 5- 8.4.1. thes~ ~loors will have the greatest negative vmlu¢~,

(b~ When systems are des igned for o n e n stairwell doors and total bui lding cwcuat ion , the n u m b e r o f o n e n doors sh0uld inc lude the exterior stairwell door.

(c) Because the oressure in the stairwell must be ~reater than the vressure in the occuv ied areas, it is n o t necessary t o r e v e a t the door o p e n i n g force tests with open doors. On en in ~ any door would decrease the nressure in the stairwell and thereby decrease the door o n e n i n ~ force o n the remain in~ doors.

A-5-$.8.1 When testin~ the combinat ion o f zoned smoke-control systems and stairwell pressurization systems, the tests aDDlicable to each stand-alone system should be conducted. Differential pressure t¢¢t~ are specif ied in both 5-3.4 and 5-3.5. When the two systems are used in combinat ion , the stairwell should be treated as a zone in a zoned smoke-control system. The m i n i m u m deskm pressures snecif ied in Table 2-2.1 annlv only to the differential pressure t ¢ ~ specified in 5-$.5. - -

Differential pressure tests conducted as directed in 5-$.4.~ are used to determine the doors that should he o n en ed durin~ the tests soedf i ed in 5-$.4.4 and 5-8.4.5. It is no t ext)ected that these values will comnlv with the m i n i m u m d e s i ~ nressures sneeif ied in Table 2-2.1. except at the fire floor.

In lieu o f soecific direct ion in the local code or contract documents , choose the doors to be o o e n e d as follows in order to produce the most severe conditions:

609

N F P A 9 2 A - - MAY 2000 R O P

(a) For the differential pressure test. t he open doors should inc lude those for which tl~e h ighes t pressure -difference was measu red in the tests with all doors c losed (see 5-3.4.3L excluding the door on the fire floor (see A-2-3.7) for rationale. W h e n m e a s u r e d with the stairwell as the reference, as descr ibed in 5-

-3.4.1. these doors will have the ~reatest negative values.

(b) W h e n systems are des igned for open stairwell doors and total bui l~lag gvacuation, the n u m b e r of ooen doors shou ld inc lude the exter ior stairwell door .

(c) For the doo r -open ing force test. the ooen doors shou ld include anv doors (UP to tlTte soecified n u m b e r ) f o u n d in the tests with all doors closed (see 5-3.4,3). to have Dressure in the occunied area greater than the oressure in the stairwell. O n e n i n g these doors adds oressure to t he stairwell, the rebv inc reas imi door- 9 p f n i n g forces on the r ema in ing doors. W h e n m e a s u r e d with the stairwell as the reference, as descr ibed in 5-3.4.1. these doors will have the greatest oositive values. If no doors mee t this criteria, it is [IQt necessary, to reoea t the door -ooen ing force tests with open

B - 1 . 3 A d d i t i o n a l P u b l i c a t i o n s .

Klote• I.H.• Design of Smoke Control Systems for Areas of Refuge. ASHP-,AE Transact ions . Amer ican Society o f Heat ing. Refi'igeratir]g and Air Cond i t ion ing Engineers . Atlanta. GA. Vol. 99. p a r t g . 199-3. PP. 793-807.

Klote• I.H.. A Me thod for Calculat ion of Elevator Evacuation Time. lourna l of Fire Protect ion Engineer ing . Volume 5. 1993• DD. 83-96.

Klote. I.H.• Design of Smoke Control Svstems for Elevator Fire Evacuati-on Inc luding Wind Effects. 2 nd S~mDosium on Elevators. Fire and Accessibilltv. Balt imore. ASME. New York. 1995 DD. 59-77.

Klote. I.H.• Alvord. D.M.. E.A,. Levin. B.M. a n d Groner . N.E.. Feasibili-tv and Desima Considera t ions of Emergency Evacuat ion by Elevators. NISTIR 4-870. National Instute of Sta~nda~ds a n d Technolo~¢. Gai thersbur~, MD. 1992.

Klote. I.H• and Bodart. X.. Validation of Network Models for Smoke C:ontrol Analysis. ASHRAE Transact ions . Amer ican Society

dq0~ , since open ing anv door would decrease the Dressure in the of Heat ing. Refr igerat ing and Air Condi t ion ing Ent, lneers . Atlanta. ~tairwell a n d thereby decrease the door -open ing force on the GA.. Volume 91. P a ~ : ~ 1985. PP. 1134-1145~" - r ema in ing doors. ....~ ":'% - -

Klote• I.H. a z ~ : ~ a u n • E.. Water Leakage of Elevator Doors with Append ix B R e fe r enced Pubficat ions Aoolicat ion ~ i ~ u i i : ~ Fire Suppress ion . NISTIR 5925. National

B-1 The following d o c u m e n t s or por t ions t he reo f are re fe renced |~s-tute o f : : ~ d s ~ f f e c h n o l o ~ v . Gai thersburg. MD. 1996• ..:~:~-:~&., x::- x:~..~#~;:~..:~:- -- within this r e c o m m e n d e d practice for informat ional purposes only Klot~!s'l,l-i~ Deal. S.P.. D ~ o g h u e . E.A.• Levin. B.M. an d Groner . and are thus n o t cons idered par t of its r e c o m m e n d a u o n s . The edit ion indicated here for each reference is the cu r ren t edi t ion as N.F: . - :~e E ~ a t i o n by '~']evat-ors. Elevator World• 1993. DD. 66-75. of the date of the NFPA issuance of this r e c o m m e n d e d practice. I~lot~{~: : '~ :~Mn. B.1VI. a n d Groner . N.E.. Feasibility of-Fire

E W q U ~ f i ~ w Elevators at FAA Control Towers. NISTIR 5445. B-I.1 NFPA P u b l i c a t i o n s . National Fire Protect ion Association, 1 .'~<~v.,. National l t ~ . , ~ f Standards a n d Technology. Gaithersburg• MD. Bat terymarch Park, P.O. Box 9101, Quincy MA 02269-9101. " ~ . ~ t M _ ~{-:~ -" ' q~ . ' ~ . .::::~

":i.~:. " : '- ':: ~ - . - ' . : : ' : : ' . ~ : .~ . . ~ . 4 ":~

NFPA 13• Standard for the Installation of Sprinkler Systems. 1999 % J ~ ' ~ - " : " ~ Levin. B.M. a n d Groner . N•E•. Emergency Elevator ~'" n d • ~.:#~}~.~:, "~{'acuatfi~n Systems. 2 Sympos ium on Elevators• ~ r e a n d .::::.+. x<.:.:::'::'6 ~.: . . . - , - -

.~.:}" :'>.':?-':: ~O~s lb l l l tv . Balt imore. ASME. NewYork. 1995 DO. 131-150. NFPA 80, Standard for Fire Doors and Fire Windows, 199.%~.~ition.}'.:-.-'Y .. "~,:.:,%-: . . . . .

.~q:~$.':$:::.. .<~: a..'.q<. " ~ .-'::::~

. . . . . . . . . . ~';':: ':':"~{~}}~::.. ~ , . i } : " - "~" '~ . , ougheed . G.D.. Mawhinnev. I.R. a n d O'Neill. I. 1994. Full- ll-t.Z O t l a e r r u o u c a t t o n s ~: . - : ! : :" " ~ . ~ : _ ~ . . . . ~ - , ~ , ~ , '%':'::.::'.%. " "" scale _riFe l e s t s a n d t he Deve lopment o i Ueslm3 t~nterla ior

B-I 2 1 ASHRAE Publ icat ions Amer ican S o c i . ~ ~ n g i ' : : ~ 6 ~ } ~ ........ Sorinkler Protect ion of Mobile-Shelving Units. Fire TechnoloE¢• Ref'rigerating, a n d Air Condi t ion ing Enginee~: : ]nc . , i~~~.~.u] l ie" :~ ~'~ Volume 30. D. 98-133. - - Circle, N.E., Adanta , GA 30329-2305. ~:-~':"i~}i::.. '%~: :# . . . .

"."¢" ":~L:'}'..'~i::i~:, :!~..:i~! Madrzvkowski. D. and Vettori. R. 1992. A Sprinkler Fire ASHRAE/SFPE, Design of Smoke Management ~ , 1 9 9 ~ i" Suppress ion Algori thm. J. o f Fire Prot. Engr] 4: 151-164.

ASHRAE, Handbook of Fundamentals, 1997. ":~)}~# "i:''r Walton. G.N•. CONTAM96 User Manual. NISTIR-6056. National

B-1.2.2 ASME Publication. Amer ican Society of M~"-"hanical Engineers , 345 East 47th Street, NewYork, NY 10017•

ASME/ANSI A17.1, Safety Code for Elevators and Escalators, 1993.

B-1.2.3 SFPE Publication. Society of Fire Protect ion Engineers , 7315 Wisconsin Avenue, Suite 1225W, Bethesda, MD 20814.

SFPE Fire Protection Handbook, 1995.

B-1.2.4 UL Publication. Underwri ters Laborator ies Inc., 333 Pfingsten Road, Nor thbrook, IL 60062.

UL 33, Standard for Heat-Responsive Links for Fire Protection Servic~ 1982 (Rev. 1984).

Inst i tute of Standards and Technoluv• Gaithersburg. MD, 1997,

Wrav. C.P. a n d Yuill. G.K.• An Evaluation o f Algorithm~ for Analvzing Smoke Control Systems• ASHRAE Transactions• Americar~ Society of Hearing. Refr igerat ing and Air Condi t ion ing Engineers . Atlanta. GA. Volume 99 fPa r t 1~ 1993. on. 160-174. -

P. !.°-.2 "J.S. C : ' : : ~ - m c r . : P ' :S] 'c=: 'c=. 'J.S. Cc.;'crnm..c:.t P - n ' - n g

UL 555, Standard for Safety Fire Dampers, 1999.

UL 555S, Standard for Safe 0 Leakage Rated Dampers for Use in Smoke Control Systems, 1999.

610

N F P A 92B ~ M A Y 2 0 0 0 R O P

NFPA 92B

(Log #CP1) 92B- 1 - (Entire Document): Accept SUBMITrER: Technical Committee on Smoke Management Systems

[ RECOMMENDATION: The Technical Committee on Smoke I Management Systems proposes a complete revision of the 1996 | edition of NFPA 92B, Guide for Smoke Management Systems in I Malls, Atria, and Large Areas, as shown in the draft at the end of I this report.

SUBSTANTIATION: The Technical Committee has conducted a complete review of this guide and updated it to reflect the best current information and data related to management of smoke in large spaces such as atria and malls. A number of changes were made while significant portions of the text were left as in the previous edition. Since the entire document is affected by the changes, however, the Technical Committee believes the entire document should be open for public comment.

The proposed revisions are designed to reflect the best current state-of-the-art and the understanding of application that has come though use of plume dynamic smoke management as covered in this guide. The principle changes include:

a. Added and updated definitions covering Communicating Spaces, First Indication of Smoke, Plugholing, Smoke Layer, and Verification.

b. Additional data o.q the impact of sprinkler operation on smoke management, including information developed by recent tests.

c. An extensive discussion of the basic principles and design methodologies and limitations. This revision addresses important limitations related to creating and maintaining a smoke layer. It also addresses the potential and limitations of using gravity vents in atria, malls or similar facilities.

d. Extensive additional information has been provided on estimating the heat release of potential fires.

e. Coverage of the response of smoke and heat detectors, including sprinklers, has been revised to reference NFPA 72, National Fire Alarm Code, as the prime means of estimating activation of these devtces.

f. Coverage on system verification. The Technical Committee wishes to recognize the effort of the

task group that developed the proposed revisions and presented them to the Committee for consideration and adoption. The members of that taskgroup include:

Harold E. Nelson, Hughes Associates, Inc. James A. Milke, University of Maryland John H. Klote, John H. Klote, Inc. Gary D. Lougheed, National Research Council of Canada William Brooks, Eichlea Engineers Michael Ferriera, Hughes Associates, inc. Douglas H. Evans, Clark County Building Department, NV Jvames Quintiere, University of Maryland

ic Dubrowski, Code Consultants, Inc. Stacy Neidhart, Marriott International Richard Roby, Combustion Science and Engineering Doug Carpenter, Combustion Science and Engineering Win irwin, North An~erican Insulation Manufacturers Assn. Amy McGarry, Code Consultants, Inc Gregory R. Miller, Code Consultants, Inc Michael E. Dillon, Dillon Consulting Engineers, Inc. Craig Beyler, Hughes Associates, Inc.

COMMITTEE ACTION: Accept. NUMBER OF COMMITTEE MEMBERS ELIGIBLE TO VOTE: 24 VOTE ON COMMITTEE ACTION:

AFFIRMATIVE: 21 NOT RETURNED: 3 Carrafa, Chapman, Pihut

COMMENT ON AFFIRMATIVE: MILKE: Figure 1-4: Where is it? Secdon 1-5.7: Replace "Pauls" and "Nelson and MacLennan" with

reference numbers and include these two as references in Appendix G. References:

Pants, J., "Movement of People," SFPE Handbook of Fire Protection Engineering, 2nd Edition, NFPA, 1995.

Nelson and MacLenr~an, "Emergency Movement," SFPE Handbook of Fire Protection Engineering, 2nd Edition, NFPA, 1995.

Section 1-6.3: There's rio change on the last line. Section 2-3.1: Move reference to Appendix G, replace the

reference in this section with a number for the reference. Section 2-3.3<1: On page 15, line number 4, replace "1998" with

reference number for:

Milke, J.&, and Klote, J.H., "Smoke Management in Large Spaces in Buildings," Building Code Commission, 1998.

Section 2-3.3e: Replace "delta-t" with ~AT." Table 2-4.1: Replace "Eqn. (9)" with ~Eqn. (3)." Replace "Eqn.

(10)" with "Eqn. (4)." Replace "Eqn. (13-16)" with ~F.qn. (8, 9, 10, 15)" (in all 3lines).

Section 2-4.1.1, 2nd paragraph, 1st line: Replace "indications" with "indication."

Section 2-4.1.2, 1st paragraph, 9th line and 2nd paragraph, 1st line: Replace "Equations (7), (8), (9) and (22)" with "Equations (8), (9), (10) and (15)."

Section 3-2.2.2: In equation (1) "B" should be "n." Section 3-2.6: Entire section was omitted from ballot. Section 3-3.1: Move reference to Appendix G, replace "[Schifiliti

and Pucci]" with reference number. Figure 3-4: Where is it? Section 3-4.3: In line 5, insert "and" between "present" and "any."

In line 8, insert "a" between "at" and "level." Section 3-5: In line 3, "fire" should be "fires." Section 3-6.1: (b) and (c) do not relate to "first indication of

smoke," but to position of smoke layer interface. As such, these 2 sections should be included elsewhere, i.e., in beginning of Section 3-7.

Section 3-8.5.2: Unnumbered equation, presented first, should be deleted.

Section 3-9: Replace "[CIBSE, 1995]" with reference number. Section 3-9: Under equation 21, replace upper case "B" with

lower case "13." In Appendix A, reference numbers need to be inserted where

references are noted, such as in Section 1-4 [Cooper et al 1982], etc. and references gathered together and appended to list in Appendix G. See the following sections: A-1-4, A-1-5.4.1, A-1-5.4.2, A-3-2.2.1, A-3-8.1.2, A-3-9, 13-5.1, B-6.1, B-6.3(a), (b) and (c).

NELSON: 1. Page 92B-16, third line in paragraph following Figure 2-3.3: Revise (delta-t) to (AT). Lower case t is normally used to indicate time while upper case T normally represents temperature.

2. Page 9213-17, Table 2-4.1: In the third column of this table revise the equation numbers. (9) becomes (3); (10) becomes (4); and (13-16) becomes (7, 8, 9 and 22).

3. Page 9213-30, Insert following 3-2.5.3: The insert starting with k~ appears to be a hold over from deleted material and the entire ..~q line'should be deleted.

4. Page 92B-43, extra equation preceding equation (18): There are two versions of the same equation. Delete the one with 0.0067 as the constant.

RICHARDSON: 1. The terms "natural" and "gravity" ventilator are both used interchangeable. Suggest correct to "natural" throughout.

2. Appendix A refers to BR 258. You should be aware that this will be superseded by a new document from BRE in June 1999, you may wish to amend the reference.

SCHUMANN: Page 15: Graphs need dries and figure numbers. Page 38: Add an * to 3-8 ~7.1.2. Page 42:A-3-7.2.1 (from the 1995 edition) must be revised to

A-3-8.2.1 to match the new text. Page 58: In A-3-4 there is a reference to Figure A-3.4. There is no

Figure A-~4 in the new text. I do not know what Figure A-3.4 should be.

TURNBULL: The majority of the document has been substantially improved by this revision. However, I have concerns that the changes to Paragraphs 4-4.5 and A-4-4.5 have significandy reduced the integrity of dedicated systems.

The 1996 edition of this document recommended supervision (now called verification) for all dedicated equipment using methods that automatically verify proper operation each time the equipment is activated. The recommended methods did not rely on manual intervention. Chapter 4 (now Chal~ter 5) contained additional recommendations for Periodic Testing of dedicated smoke control systems on a semi-annual basis.

The proposed 2000 edition of this document removed the distinction between dedicated and non-dedicated equipment in Paragraph 4-4.5, meaning that this paragraph now applies to all smoke control equipment. At the same time, a new Item (e) "Periodic acceptance testing in accordance with Chapter 5," was added to the list of verification methods in Paragraph A-4-4.5. This change suggests that equipment should be verified byautomatic means, such as those described in Items (a) - (d) OR through semi-annual manual testing. In other words, a non-binding agreement to test the system twice per year removes the need for automatic verification. This seems entirely inappropriate, since system reliability and readiness is a desired factor.

611

N F P A 92B m MAY 2000 R O P

Furthermore, combining the discussions of verification methods for dedicated and non-dedicated equipment may result in the recommendations of this document becoming unclear. Many non- dedicated components are operated daily for purposes such as comfort control, and therefore failures of these components are generally noticed quickly. Equipment operated in this manner would not normally need the type of verification discussed in Paragraph A-4-4.5, Items (a) - (d), to provide assurance that the equipment will operate when activated for smoke control. In contrast, dedicated equipment and some non-dedicated components are infrequently or never operated during normal building conditions. In these cases, automatic verification methods, such as those described in Items (a) - (d) would be appropriate. In both cases, the periodic testing described in Chapter 5 should be performed.

To remedy the situation described above, I recommend the following changes t6 the proposed document:

1) Restore the word "dedicated" to Paragraph 4-4.5 so that it reads:

"4-4.5* Control System Verification and Instrumentation. Every system should have means of ensuring it will operate if

activated. The means and frequency will vary according to the complexity and importance of the system."

2) Delete Item (e) in Paragraph A-4-4.5.

(Log #2) 92B- 2 - (1-4 Smoke Layer Interface): Accept in Principle SUBMITIT_~ William Brooks, Eichlea Engineers Inc. RECOMMENDATION: Revise the current definition of Smoke L~oer Interface. Retain all existing text, and add the following:

r the purpose of determining the position of the smoke layer interface in experiments or CFD simulations, the smoke layer interface shall be assumed to be height where the temperature has increased to 20% of the ceiling temperature. SUBSTANTIATION: The current language defines the smoke layer interface height in a qualitative manner. This is not acceptable as we move toward CFD modeling of fire and smoke behavior. By placing a benchmark in NFPA 92B, experimenters will have a single definition of smoke layer interface to use in assessing test data. COMMITrI~E ACTION: Accept in Principle.

See the Committee Action on Proposal 92B-1 (Log #CP1). COMMHq'EE STATEMENT: The submitter's proposal, as expanded by the committee, is incorporated into the proposed Appendix A-l-4 Smoke Layer Interface (see the draft at the end of this report). NUMBER OF COMMITTEE MEMBERS ELIGIBLE TO VOTE: 24 VOTE ON COMMITrEE ACTION:

AFFIRMATIVE: 21 NOT RETURNED: 3 Carrafa, Chapman, Pihut

VOTE ON COMMITTEE ACTION: AFFIRMATIVE: 20 NEGATIVE: 1 NOT RETURNED: 3 Carrafa, Chapman, Pihut

EXPLANATION OF NEGATIVE: MILLER= I agree with the proponent for the reasons stated.

(Log #1) 92B- 4 - (3-5.4): Accept in Principle SUBMIIq'ER: William Brooks, Eichlea Engineers Inc. RECOMMENDATION: Replace the existing text with the followinl~:

3-3.4 (~eilingJet Temperature. Ceiling jet temperature can be approximated by using the following correlations which depend on the radial distance from the center of the fire plume.

3-3.4.1 For r /H less than or equal to 0.18: Ceiling Jet Temperature = Ambient Temperature + (16.9" (Q^(2/B)) ) /H^(5/$)

5-3.4.2 For r / H greater than 0.18: Ceiling Jet Temperature = Ambient Temperature + (5 .38*(Q/ r )^ (2 /3 ) ) ) /H where Q = Total heat release rate r = radial distance from center of fire plume to selected point H = vertical distance from fire to selected point

All values in SI units. SUBSTANTIATION: The present correlation is of little practical value in design problems, and may not characterize time/temperature variations from fire test data.

For example, in an atrium 60 feet tall, sprinklers spaced at 15 feet on center will be at r / H = 0.18. In the same atrium, smoke detectors spaced at 900 sq ft will be at r / H = 0.35. In both cases the existing correlation will underpredict actual expected temperatures due to the 0.6 r / H assumption.

This proposal utilizes the correlations now incorporated into DETACT for prediction of ceiling je t temperatures.

The Society of Fire Protection Engineers is conducting a number of experiments, measuring ceiling jet temperatures. The suggested correlations can be adjusted during the ROP. COMMITTEE ACTION: Accept in Principle.

See the Committee Action on Proposal 92B-1 (Log #CP1). COMMITTEE STATEMENT: Rather than specify the equations used to determine ceiling je t temperature for detector actuation, the user is referred to NFPA 72, National Fire Alarm Code, in the proposed Section $-3 (see the draft at the end of this report). NFPA 72, in turn, references use of the DETACT models. NUMBER OF COMMITTEE MEMBERS ELIGIBLE TO VOTE: 24 VOTE ON COMMITTEE ACTION:

AFFIRMATIVE: 21 NOT RETURNED: 3 Carrafa, Chapman, Pihut

(Log #4) 92B- 3 - (1-4 Smoke Layer Interface): Reject SUBMITTER: William Brooks, Eichlea Engineers Inc. RECOMMENDATION: Revise the definition of Smoke Layer Interface as follows:

Delete "which can be several feet thick" from the second sentence. SUBSTANTIATION: The calculation methods implied by NFPA 92B produce significant differences between the "first indication of smoke" and the "smoke layer interface." These differences are more than "several feet," and the present language can lead the user to feel that the differences are not that substantial.

In fact, these differences can be substantial. For example, using an atrium size of 100,000 sqft, a fire size of 2000 Btu/sec, and an atrium height of 100 feet, the calculation methods predict a "first indication of smoke" at 40 feet and a "smoke layer interface" at 55 feet after approximately 950 seconds.

The term "several feet" does not belong in the definition when its value depends on a number of variables which could cause it to exceed "several feet." COMMITTEE ACTION: Reject. COMMITTEE STATEMENT: The committee believes the current definition, as modified in the proposed draft, is adequate. The language, which is proposed to be deleted, does not hurt the definition. NUMBER OF COMMITTEE MEMBERS ELIGIBLE TO VOTE: 24

(Log #3) 92B- 5 - (Appendix E, Example 3): Accept SUBMITrER: William Brooks, Eichlea Engineers Inc.

I RECOMMENDATION: Delete Example Problem $. SUBSTANTIATION: This example illustrates the weaknesses of the current methodology. The "apparent inconsistencies" should alert the committee to the fact that there may be serious errors in the formulation of the correlations used to develop the "answers." COMMITTEE ACTION: Accept. NUMBER OF COMMITrF.E MEMBERS ELIGIBLE TO VOTE: 24

• VOTE ON COMMITrEE ACTION: AFFIRMATIVE: 21 NOT RETURNED: 3 Carrafa, Chapman, Pihut

(Log #5) 92B- 6 - (Appendix E, Example 4): Accept SUBMITrER: William Brooks, Eichlea Engineers Inc. RECOMMENDATION: Revise this example to be consistent with previous definitions. In all cases, use the term "smoke layer interface height" rather than "smoke" or "smoke layer depth...". SUBSTANTIATION: As the definition and illustration on page indicate, there is a substantial difference between the appearance of smoke and the position of the smoke layer interface. If Figure 1-4 approximates real conditions, the use of a design smoke layer interface height at only 5 to 10 feet above the highest walking surface

612

N F P A 92B - - MAY 2000 R O P

would provide very little safety benefiL By using only the term "smoke layer interfac& in this example, users would not make the mistaken assumption that the conditions below the smoke layer interface height are safe, or that there is no smoke present. COMMITTEE ACTION: Accept. NUMBER OF COMMITTEE MEMBERS ELIGIBLE TO VOTE: 24 VOTE ON COMMITTEE ACTION:

AFFIRMATIVE: 21 NOT RETURNED: 3 Carrafa, Chapman, Pihut

NFPA 92B

Guide for Smoke Management Systems in Malls, Atria, and Large Areas

-I-99g 2000 Edition

Chapter 1 General Information

1-1 Objective. The objective of this guide is to provide owners, designers, code authorities, and fire departments with a method for managing smoke in large-volume, noncompartmented spaces. This guide documents the following:

(1) The problem of smoke movement in indoor spaces (2) Basic physics of smoke movement in indoor spaces (3) Methods of smoke management (4) Data and technology (5) Building equipment and controls (6) Test and maintenance methods

1-2" Scope. This guide provides methodologies for estimating the location of smoke within a large-volume space due to a fire either in the large-volume space or an adjacent space. These methodologies comprise the technical basis for assisting in the design, installation, testing, operation, and maintenance of new and retrofitted smoke management systems installed in bui] ~s having large-volume spaces for the management of smok...~ the space where the fire exists or between spaces not s~.'~ "~_. smoke barriers. Buildings within the scope of this g ~ those with atria, covered malls, and similar large-vo[fim~ ~r (See NFPA 92A, Recommended Practice for Smoke-Con..t.r.ol Syst mechanical smoke control between f i r e - c o m p a r t n ~ ~ .~.:.....:. separated by smoke barriers and NFPA 204-M, (¢~,~ d~,=~,~..A~. Ventin~ for gravi~ venting.) This guide is~:~..intended::~ p o warehouses, manufacturing facilities, ov:~'~ifft~!milar s[ ''~,~ 'his guide does not provide methodologies to ass~i~.~.e effeci ~f smoke exposure on people, property, or missi(~%~.ntin~ y.

The

means, It is. in some circumstances, possible to remove smoke by gravity venting. The capacity ofgTavity vents to move smoke through a vent is a function of both the depth and temperature of the hot layer, Procedures for determining the capabilities of gravity venting are contained in NFPA 204. Guide_for Smoke and Heat Venting. That document, rather than this. should be used to the extetlt that gTavity venting is considered, In general, gravity venting and mechanical venting should not be used in combination for the same space without comprehensive modeling of the situation to erasure that the tyravity vents will not lose efficiency, or even be reversed by the mechalaical venting.

1-3 Purpose.

1-3.1 The purpose of this document is to provide guidance in implementing smoke management systems to accomplish one or more of the following:

(1) Maintain a tenable environment in the means of egress from large-volume building spaces during the time required for evacuation.

(2) Control and reduce the migration of smoke between the fire area and adjacent spaces.

(3) Provide conditions within and outside the fire zone that will assist emergency response personnel in conducting search and rescue operations and in locating and controling the fire.

(4) Contribute to the protection of life and reduction of property loss.

(5) Aid in post-fire smoke removal.

1-3.2 Specific design objectives can be established in other codes and standards or by the authority having jurisdiction.

1-4 Def'mitions. For the purposes of this guide the following terms have the meanings given in this chapter.

Atrium. A large-volume space created by a floor opening or series of floor openings connecung two or more stories that is covered at the top of the series of openings and is used for purposes other than an enclosed stairway; elevator hoistway; escalator opening; or utility shaft used for plumbing, electrical, air-conditioning, or communications facilities.

Ceiling Jet. Aflow of smoke under the ceiling, extending radially from the point of fire plume impingement on the ceiling. Normally, the temperature of the ceiling jet will be greater than the adjacent smoke layer.

Communicating Space. A g~paces within a building that ~ h a s an open pathway to.a large-volume space such that smoke from a fire either in the l ~ u n i c a t i n g space " -v can move , ""~ "'" " " " ~ f i ~ l ~ ' ~ t 4 n ~ ~ Communicating spaces can open directly ~ ~ ' - . ~ s p a c e or can connect through open

activated, such as during smoke control, testing, or manual 9verride qperations. Failure or cessation of such positive confirmation results in an off-normal indication.

First Indication of Smoke.* The boundary between the transition lone and the smokefree air. as depicted in Figure 1-4. Equations 0 3 and 40 4 are used to predict the height of this boundary, for smoke fillipg with n0 mechanical exhaust; operating.

c layer interlace 0 • ~ o n s 1 4 , 15,

121) UJ

. . . . . . . . dication of smoke (Equations 9 and 10)

Figure 1-4 Smoke layer interface.

Guide. A document that is advisory or informative in nature and that contains only nonmandatory provisions. A guide may contain

613

NFPA 92B - - MAY 2000 ROP

manda to ry s ta tements such as when a guide can be used, bu t the d o c u m e n t as a whole is no t suitable for adop t ion into law.

Large-Volume Space. An u n c o m p a r t m e n t e d space, general ly two or more stories in height , within which smoke f rom a fire ei ther in the space or in a c o m m u n i c a t i n g space can move a n d accumula te without restriction. Atria and covered malls are examples of large- volume spaces.

Pluvholing. The condi t ion where air f rom below the interface is oulled th rough a relatively shallow smoke la~'er t ~ ¢ 1~o a h igh exhaus t rate at that noint .

Separated Spaces. Spaces within a bui ld ing tha t are isolated f rom large-volume spaces by smoke barriers tha t do n o t rely on airflow to restrict the m o v e m e n t of smoke.

Smoke . The a i rborne solid and liquid part iculates a n d gases evolved when a material unde rgoes pyrolysis or combus t ion , toge ther with the quant i ty o f air tha t is en t ra ined or otherwise mixed into the mass.

Smoke Barrier. A c on t i nuous m e m b r a n e , ei ther vertical or horizontal , such as a wail, floor, or ceil ing assembly, t ha t is des igned a n d cons t ruc ted to restrict the m o v e m e n t of smoke. A smoke barrier m igh t or m i g h t no t have a fire resistance rating. Such barriers m igh t have nro tec ted onenin~s. SmS!zC bz~--c~ cma

Smoke Damper. A device tha t meets the r eau i r emen t s of UL 555S, Standard for Sakt~ Leakage Rated Darnt~ers for Use in Srtlok,¢ Control S~stems. ciesi~n-ecl to resis t the nassa~e of-air or smoke. A c o m b i n a t i o n f i r e and smoke darnner shou ld m e e t the requirement,~ of UL 555. Standard for Safet~ Fire DaraOers. and UL 555S,

I ^ . 1 . . . . E I . ~ A T~t . . . . . . 47 ' ^_ TT .~ : ~ ~ - - ^1 . ^ / ~ ^ - -+ . ^1 ¢ ~ . _ + . . . . . A : .

A ^ . : ~ - - ~ A + . . . . : . ~ +1~ . . . . . . . . . g" ~ : . . . . . . I , ^ A . ~ - - I . . : - - - ~ : ^ - - g '~ . . ^

~ A . - - ~ 1 . - - A . . . . . : . . . . + 1 ~ - . - - - ^ - - ~1~ . . . . . . : . . . . . + . ~ . f ' TTT K K K

Smoke Layer. The accumula t ed th ickness of smokd;.:~;t physical or the rmal barrier. T he smoke laver is no t ~ "

smokefree air. "::'~!i!i:~ii~.: ''~:'" ~'~!~'~

Smoke Layer Interface. The theoretical b o u n d a i ~ e e n a smoke layer and smokef ree air, as depic ted in F igu l~ 14 . In practice, the smoke layer interface is an effective b / ~ n d a r y within a t ransi t ion buffer zone, which can be several feet thick. Below this effective boundary, the smoke densi ty in the t ransi t ion zone decreases to zero._This he igh t is used in the appl icat ion of Eouat ions 8. 9. 10. and 15.

Smoke M a n a g e m e n t System. An eng inee red system that includes all me thods tha t can be used singly or in combina t ion to modify smoke movement .

Stack Effect. The vertical airflow within bui ldings caused by the tempera ture -c rea ted densi ty dif ferences between the bui ld ing inter ior and exterior or between two inter ior spaces.

Tenab le Envi ronment . An env i ronmen t in which smoke and hea t are l imited or otherwise restricted to main ta in the impact on occupants to a level that is no t life threa tening .

Transi t ion Zone.* The laver between the smoke laver interface and the first indicat ion of smoke in which the smoke laver t emnera tu re decreases to amb i en t .

be achieved by the smoke laye of the smoke 1~

1-5 Design Principles.

1-5.1 Fire in Larg~Volume Spaces, Malls, and Atria.

1-5.1.1 Smoke p roduced f rom a f i r e in a large, open space is a s sumed to be buoyant, rising in a p lume above the fire and striking the ceiling or stratifying due to t empera tu re inversion. After the smoke e i ther strikes the ceiling or stratifies, the space can be expected to begin to fill with smoke, with the smoke layer interface descending. The descent rate of the smoke layer interface d e p e n d s on the rate a t which smoke is suppl ied to the smoke layer f rom the plume. Such smoke filling is r ep resen ted by a two-zone model in which the re is a distinct interface between the bo t tom of the smoke layer and the ambien t air. For e n g i n e e r i n g p u r p o s e s , the smoke supply rate f rom the p l u m e can be e s t i m a t e d t o be the air e n t r a i n m e n t rate into the p lume below the smoke layer interface. Sprinklers can reduce the hea t release rate an d the air e n t r a i n m e n t rate into the plume.

1-5.1.2 As a resul t of the zone mode l approach~ the mode l assumes un i fo rm proper t ies (smoke concen t ra t ion a n d t empera tu re ) f rom the poin t of interface t h rough the ceiling and horizontal ly t h r o u g h o u t the entire smoke layer.

1-5.1.3 An equi l ibr ium posi t ion for the smoke layer interface can exh.~usting the same rate of smoke as is suppl ied to

:note l a y e r . . . . ~ : . . s m o k e exhaus t can delay the rate of descen t

1-5.1.4 adjacenl atr i u m~i

supplie~. rate of a

.tt"-~x.'~oke layer has descended to the level of i e d ~ e s , p reven t ion of smoke migra t ion f rom the Lo t h ~ . ~ i ~ q ~ a t spaces can he accompl i shed by ; or o p ~ airflow. NFPA 92A, Recommended ,e-contra~..#~Systems, provides gu idance on the use of smoke migrat ion. Opposed airflow can be used to a igradon into open adjacent spaces, with air rithin the adjacent space. T h e r equ i r ed volumetric ]ed to achieve the necessary velocity can be

:l.~?l~..-'~ order for the smoke exhaus t fans to be effective, makeup ~ r mff.~ be provided. Makeup air shou ld be provided a t a low velod{y. For effective smoke m a n a g e m e n t , the m a k e u p airflow

's t be sufficiently diffused so as no t to affect the flame, smoke Jme, or smoke interface. The supply points for the m ak eu p air

shou ld be located benea th the smoke interface. T h e rate of m a k e u p airflow shou ld no t exceed the exhaus t rate su ch tha t the a t r ium or mall achieves a positive pressure relative to adjacent spaces. If air en ters the smoke layer above the interface, it mus t be accoun ted for in the exhaus t calculations.

1-5.2 Fires in Communicating Spaces. Fires in c o m m u n i c a t i n g .~Phaces can p roduce buoyant gases tha t spill into the large space.

e design for this case is ana logous to the design for a fire in the large space. However, the des ign mus t consider the difference in e n t r a i n m e n t behavior between a free p l u m e and a spill p lume. If c o m m u n i c a t i n g open spaces are p ro tec ted by au tomat ic sprinklers, the calculations set for th in this guide migh t show that no addit ional vent ing is required. Alternatively, whe the r c o m m u n i c a t i n g spaces are spr inklered or not, smoke can be prevented f rom spill ing into the large space if the communica t ing space is exhaus ted at a rate to cause a sufficient inflow velocity across the interface to the large space.

1-5.3 Detection. Effective design of smoke m a n a g e m e n t systems requires early de tec t ion of the smoke condit ion.

1-5.4 Fire Suppression Systems.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . t " . . . . . . . . . . . . . . . . . . . . . . . . . . . ]

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . b .................. l ~

1-5.4.~L~ Automat ic suppress ion systems are des igned to limit the mass bu rn ing rate of a f i r e - s ~ and will, therefore, limit smoke generat ion. By limitin~ the mass burn in~ rate o f a fire. s m o l ~ ~enera t ion will be reduced. Fires in spr ink le r fd spaces adjacent to atria and covered mall pedes t r ian areas can also be effectively l imited to cauzc .-- 'nimal r educe the effect on a t r ium spaces or covered mall pedes t r ian areas and thus increase the viability of a smoke manageme(l~ system.

614

N F P A 92B - - MAY 2000 R O P

1-5.4.82" Ac ' - ;~dea of :~.~.:k!er: ne:r : fire ~."l! ~."~u=e ccvl iag of "~ . . . . . . . . . . . . . . " ^, rc~ulfing in a lo~: of buo)~'~c)'. The likelihood of sprinkler activation is d e p e n d e n t on--the on many factors includin~ heat release rate of the fire and the ceiling height. Thus, for reddest fire sizes, sprinkler operat ion is most likely to occur in a reasonable t ime in spaces with lower ceiling heights, such as B ft (2.4 m) to 25 ft (7.6 m). Activation of sprinklers near a fire will cause smoke to cool. result~n~ in reduced buoyancy. This reduced buoyancy can cause smoke to descend and visibility to be reduced. The equations in Chapter .B that illustrate smoke filling [ ( ~ ) and (~-~)] and product ion [(4-¢8), ( 4~ . ) , (4-7_J_0.), and (-24-_1,5.)] do not apply where a loss o~ buoyancy due to sprinkler operation has occurred.

1-5.4.3" Snrinkler activation in spaces adiacent to an atrium will result in coolin~ of the smoke. For low heat release rate fires, the tempera ture of the smoke ]eavin~ the comnar tment is near ambient , and the smoke will be dispersed over the heiffht of the opening. For hiffher heat release rate fires, the smoke tempera ture will be above ambien t a n d t h e smoke enterin~ the atrium is bu o__~o~_~.

1-5.5 Operat ing Conditions. The smoke management system componen t s should be capable of cont inuous use at the maximum temperatures expected, nsing the calculations contained in this guide.

1-5.6 Tenability Analysis. Where the design is based on maintaining tenability of a por t ion of space, one of two approaches can be pursued. First, the design might d e p e n d on preventing the deve lopment of a smoke layer in that por t ion of the space. Second, the design might be based on comparing the qualities of a smoke layer to hazard threshold values. Such a demonst ra t ion requires that the effects of smoke on people be evaluated. Tenability factors that can be considered include, but are no t limited to. the followina:

(a) Heat exnosure (b) Smoke toxicity

(c) ~ ~<"~:::':*':~"~"'~a~ili i Tenability analvsis s u c h aa c--M'aut ' :on is outside t h ~ e oL~is , ,

guide. However, o ther references are available that '~re:~:,.:...::.:"~ analytical methods for tenability analyses [34]. ~ ..... "''~:'*-~'?:-2~:..

to

Ti~

1.

1-6 Design Parameters.

1-6.1 General. Design criteria should include an unders tanding with the authority having jurisdiction of the expected performance of the system and the acceptance test procedures.

1-6.2 Leakage Area. Design criteria and acceptance testing of smoke managemen t systems should be based on the following considerations with reference to the smoke zone and communica t ing zones:

(1) Small openings in smoke barriers, such as construction joints, cracks, closed door gaps, and similar clearances, should he addressed in terms of maintaining an adequate pressure difference across the smoke barrier, with the positive pressure outside of the smoke zone (see NFPA 92A, Recommended Practice for Smoke-Control S y a ~ s ).

(2) Large openings in smoke barriers, such as open doors and other sizable openings, can be addressed in terms of maintaining an adequate air velocity through the openings) with the airflow direction into the zone of fire origin.

1-6.3" Weather Data. The temperature differences between the exterior and interior of the building cause stack effect and de te rmine the stack effect 's direction and magnitude. The stack

effect must be considered in selection of exhaust fans. The effect of temperature and wind velocity varies with building height, configuration, leakage, and openings in wall and floor construct ion.

1-6.4 Pressure Differences. The max imum and mi n i mum allowable pressure differences across the boundar ies of smoke control zones should be considered (see NFPA 92A, Recommended Practice for Smoke-Control Systems). The maximum door opening forces should not exceed the requirements of NFPA 101% Life Safety Code ® , or local codes and regulations. The min imum pressure difference should be such that there will be no significant smoke leakage dur ing building evacuation. The performance of the system is affected by the forces of wind, stack effect, and buoyancy of ho t smoke at the time of fire.

1-6.5 The design objectives contained in Chapter 1 can be met by a variet~ of methodologies. Some of those methods are fur ther explained in Chapter 2.

Chapter 2 Design Considerations

\ E ' : . ^ : . . . . . . I . . . . . . . . . : . . . . . . . . . C . ~ 1 . ^ (a, . . . . . n I . . ~ . . . . . . . . . . r . . . . . . . . . . . . . . . . . . . . .

2. Rc.'r..::'c :meEt from k".e !=gc vc!ume space a: a rctc : 'Jffic 'cn:

V '

n

* h ~ I . . . . . . . . . . . . . .

2. Prc: 'cnt ;mokc from er.tcNr.g Xk.c !m-gc, -vv . . . . . . . . . . . . v ~ ~)' . . . . . . "~ ~ r f l c-'.'. ~ t ' t " . . . .

2 !.2.2 ~.~=:=F,c=c:t c f S-----e!:c -- Ce==:=='==d=g Space.

~ Fire c - - - aa t c= "n l=:'~e ":c!'-mc :-=co.

_¢'... ':T::'=2":.Td.*.':FZ:22"2Y':2Z~:o':,::':~ZT;2: 2Z 2;;'." ~...:2 2 " e - "

2=:2:oo~,: Z "~.Y:..~.v:L'::'.T::F'-7~Z::L~'L=:" ~r~7:22?.2 ..:::~-2~:~.

n o t be co,~,~!c te ! 7 c f f c c d ' : c "Z ~h.c go'-:rce o f ~-.e .~;c :.3 ~ : c c d 7

. . . . . . . . . . . . . . . . . . . . . . . . . . . . ~, : p : cc : :v. xh.e ".=FFer pot t le= of t.he I . . . . . . . I . . . . . . . . .

2. E: '~au:t the !argc vc!ume :pace co that i t i : at a n c ~ d : ' e

g. U:c ~-::2.ow ^~ "~::c"..=:ed in tD.~: ace ' -morn or ha r : ' . e : ~z d i scu=ed in N ~ A 9~ s. or bo+~.

615

N F P A 92B - - MAY 2000 R O P

. . . . . . . . + ~C . - - ^ I . . . . . : . I - , : ~ + I * . : . . . . . . : ^ . 3 : . . . . . . . .4 : ~ ~ . T ~ ' O A I + X O A

the use cf phTsic=2 ~ar -c rs ta limit =mckc mc- 'emcnt cr methadc *-w limit smoke v~'~a"- ' :^- , such az c=ntrcll!ng t.hc fucl ~.r us!rig autom=tiz fire suppre==!cn.

2 A ~ = e :.= ++he w...=rg= Ve!'.=---= SF+-.=e.

O A 1 ~ ' - - ^ I . . . . . . . . . . . + . . - - ~ - - - . c ^ - 1 . . . . . . . I . . . . . . . . . . . . .

!ar-e~, -'al'-mc s-=cc~, cr tc !ir:-:t "&c m..c'.:nt cf smc!:e f ram . ~ v ^ _ + . 1 : . . . . . ~ - ^ . . . . . . . : . - 3 - - * 1 - . ^ I . . . . . . . I . . . . . . . . . ~ I P I I ~ _ C - - l l ^ . . a - - - -

2 !.$ gad= Caaddc~-t':caa. The =c:cc+~an af :'=r'ca= 4csign

p . . . . . . 0 . . . . . . . . . . . . . . . . . r . . . . 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ + r " +

. . . . . . . . . . . . . . . ~ ) - - p . . . . .

., +, ,=,_ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

communica~eg space. (d) A=caz cf refuge, cP..hcr tcmpara= 3' c r indefinite.

. . . . . . . . . 0 . . . . . . . . . . . . . . I . . . . , . . . . . . . . . . . . r l ~ . . . . . . . , + "

• " P 1 - _ ~ . . . . . . + I - . _ A I ^ + ^ ~ . _ ~ I - ^ - I . . . I ~ + + : . k ^ C ^ . ^ + I - . . . . . 1 . ^ I . . . . 1 ~a, . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ . . . . j . . . . . . . . . . . . . . . . . . . . . . . . .

management s) : tcm ":s pra-a.dcd t= ms!st s~fc c;'acaa*dcn, ace=par:: rcac'dcn ~mc tc "..he cmcrgcnc 7 and e-ac-.:a4cn "4me =hc'a!d hc ca,n=!dered.

q ' l ~ ^ L r t r A O . , - , . . . . . a ~ - . I + . ^ I . . . . . . . I . . . . . . . . . . . A ( b ) . . . . . . . . . . . , . . . . . . . . . . . . . r ~ . . . . . + , . . . . . . . . . . V . . . . . . .

• . + + • . + . ~ cammun:canng space= .-=.us. ~c =sopped :f := apera~cn -:'au!d

I . \ 4 ~ - - ^ 1 . - - . I - . . . . I . . I L . . . . . . . . . A I : ~ + I - . ~ I . . . . . . 1 . . . . . . . . . . l b . . . . .

"&c dedrcd smcke !a:,'cr interface.

~ . , ~ . . . . + I + : . . . . . . * : ^ + I + I . . ~ + . I + ^ ~ - I . . . . . . : . . . . . . I . , : ~ I ^ . - - - . . 4 . I - . ^

:..~2.~:;" 2, . . ,~ +~2":=L--_'+'2.'+L,".'= ~L~"~.E . . . . +,T-, ---5 ~ . . . . . . . . . . . . . . . . . . . . . . . L " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

m l A q A . . + - - - - . + t + A . * ; . . - - + : ^ - - T I ~ . . . . ~ . . . . . . + . ' ~ _ ^ ~ + + I ~ ^ I . . . . . . I . . . .

. ~ . ~ ^ . I - . ~ . . 1 A I - -+ . ~ . : A . . . A ~ . . . I . . + : - - - - + 1 . . . + , . , - - . - - 4 : A^+.~+^. + ^ I ~ .

2 2 1~==~-_ L-:mR=t'c==.

2 2 .1 Smoke Aae=m:!ar:a= D~pt~. It !a nct a =ea!!='dz.72! 7 achic;=blc dc=!gn abjcc'd:'e ta prevent =ccumu!=:dan af smckc . . a + l , ~ : ~ + i + . . . . . . . . . . ~ . " . . . . ++ ~ 1 . . . . . . . I . . . . . . . . . . . . .,.4 . . . . . + . . . . . . . . . . . . 1 . 1 ~ . . 1 ~ . . . . . . . . + - - . - - ~ . . . . . . . . . o p . . . . . . . . . . . . . . .

0 c l ~1 T X : . . . . . + . * ~ - - ^ I ~ ~ ' - - ^ a . - - I ' ~ _ + I . l - - . - - - c - - - - ~ . . . . + : ~ ^ C

=,7;2~2*.-~2"+-7.Z2.+-2:~2--7-Trz -. "++"-L-'J".~" .CY.2.7^.'+-~ 7Z^ , . . . . . . . . . . . . . . . . . . ~ " r . . . . . . . . I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ p p , . + . . l - + . ' ^ - - I . . ~ I ~ . . . + I ~ ^ A ~ . ~ ' - - A I - - - - + I - .

2 2 D c a ~ a = F c a : " = c = .

~ 1 . . . ^ . . 1 . : 1 1 ~ ^ A I ^ + ^ . ~ + ~ ^ A I

- - A . . . . . . . + ' . . . . . . . . 4 : - - + - . + . c . . . . . . ~ + I ~ ~ + I . . . . . . . + - - - . X . . . . . 1 . . . .

i~kh *..heir pe~er. ' r~nce Luring nerma! . . . . . . : . . . . . as+: . . . . . affected 57 en'Aronmcnt=2 factors, aver "&e life of the a;.=tc..'q, -~d • c~r ab!!K~/ta ;.~h=tand ~hc s~csscs endured d a ~ n g a fire. Typi~ll7, s'.:cE a c c m p e n e n t re : 'ew "s cenducted in *2".e e':a!ua*don

+u~c!ent enough to ensure re!.:abi!!t 7 of the cemponen~. ?Ass, the impact cf xhc func4anal dependence cf "..he ce.mponcn~= an one ~ + h . . . . . . . + ' I~ . . . . A : I . . . . . . . . ' - - - - a I ~ . . + I ~ . . . . I . . ~ + . ' ~ - - ^C : - - A I : . J . 3 . . ~ I

. . . . . . . +I++ + + + ~ I + . . + + - - + . + l t ~ k t l ~ . . . . I . , . ; . : . . . . A + A A I + c ~ . . . v . . . . . . . . A . . . . . : . . . . . . . . . . . . . . . . . . - - . -+- . . . . . . . . . . . . . . . . o, f rcquznt ma!ntenm-:cc =rid tes'dng ==c nccdcd tc =.==c= the =}~tzm

+ 1 . . . . . . . . . . . t - " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I . . . . . . . I . . . . . . . I

2 ~.2 Pc='c~'c Taa:':=.g. Pc -ad ic te=+dng is c+~aen*a~ ta cnaurc that t.t,. - 4

. . . . . . . . . . . . . . . . l "~ . . . . . . . . . . 1 ̀ + . . . . . . . . . . . 0 r . . . . . . . . . . . . . . .

_ ^ . + ~ . . . . . . . 4 ~ . - + . . . . I ~ . . 1 ~ 1 I . . ~ A ^ . . ' - - - - ^ A + . . . . .++ + - . + . ' - - - -

. . . . . . . . . . . ! + P " . . . . . . " a - - " 1 " " . . . . . . . . . . . . . . . . . . . . . . . . . r . . . . . . . . . . . . . . .

sygt~,. . ~ccat ; ;c .c.ccc~ f o r pc;~c;-m..~.'~cc vcri~cat~c~+ ,,"~c.~z'z;c.,-'~,e.-~ : ~ ^ . ^ ~ A : f - ~ . . . . . l + : + : - ~ I ^ . : - - I . . I ^ + I . + . ^ + . . . I + _ . . . . . . . : I , ~ I ~

par'dall 7 prc;~dcd as . . . . . u, . . . . . :+^--

2 *-.2.2 Narrra!!7, a!! ==tams'de dolce'dan dc; 'cca :;-'.-hln ~k¢ large . . ^ I . . . . . . . . . . . +..3 . . . . . . . . ' ^ ^ + : . . . . . . . . . I , . ^ . . I + 3 ^ ~ + : . ~ + ^ + I , . ^

=meke m ~ z g e m e n t ~=tem. Detect¢~ f=r :Feci~ F.urp::=:, :uc~

+ . . . . . . . . . . . . . . I + ~ . . . . . . . . . . . . . I . . . . . . . . . . . . . . . . . . . . . . . . . . . .

- - + - - . . ~ u v ~ * . + ~ + . + . . . . . . u v . * + . . ~ . . ~ + . ~ . . . . . . . . + . . - - . . + ~ . . . . I . . . . . "

A'-t~ma+2c a-.etec~en de:'ce= :h='-'!d ne t ~e :~nnecte~ a~rec+J}' te the :mo!-c management :}:tern " ; '~c : : : f a r ' he r caneern for ~ e . . . . . ~ " I . . . . . . . . . . " . . . . 7 . . . . . . . . . . . . ~ ' ' + 1 . . . . . . . . . . . . . . . I . . . . . . . .

m l A ~ ~ ~ ^ + + . . . . . . I . ^ J ^ + . ^ + . . . . . . . 1 + 4 I . . . . . . . . 3 . . . . . . . I . . . .

+ + + I t . . . . , c I ~ . ~ . . . . I . . . . . . . . . + . . . . , ~ A + A + h ~ + l . h + e l = , l + + + + . . . . .

. . . . . . : ~ . I ^ C . . . . . , : ' . : . . . . ..-I ~ ^ - : * : ^ - - - - 1 I * . - - . - - , - 3 . . . . . . : . . I ^ . ^ + : ^ - - ^C

+ ~ ^ ~ _ ^ + . ^ C . + - - + : ~ - ^ + : . . . . . 3 ~ : . . . . . . . + . . . . . . . .+3 I - . . . . . ~ . . - - I - - - . . 3

mechanical ¢ . . . . .

+ I - . = k ^ - - - I , ~ - - : . ^ - - + - I I . . . . . ' _ _ + I + . . . . . . . . + - I . , . I + ~ . . + + - - + ~ ^ _ . . . . . . . . . . . . . . . . . . . . • .T . . . . . . . . . " - - - - - - + L "+" . . . . . . . . " ' - - . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . + . . . . + . . . . . . . . . . . . . + . . - - . . u l b . . +

. . . . . . . . . . . . . . . . . . . . . . . . . . I t .+ ' . + . 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

+ ^ + I ~ ^ A . I . . . : - - 1 - - 1 ~ + . . . . . . 1 . . . . . . . . . . . +

2 ! . 2 . 5 Autama*dc : p - n k I c r ":rater .qcw shauld alas usual! 7 Ee used + ~ + + ~ . . . . . . . . . . . . . . - " ~ ' " + I ~ . . . . . . . . . ) ' + + + " ' ° + . . . . . . . - t " . . . . . . . . . . . . . + ~ , , ~ I . I + + + . . + + + m h + + ~ + . , - I . . r ~ + h + ' h + + ~ I . + + , ' 4 + + + ~ + ' 3 ~ + . - - + + ~ . + ~ + k S

,~ZT..\.;.~T'++2 - ;2 72~;.~"+"S."2 .~_'+-'+.+'_'+".'Z ,̂".-..+'.~7+.--22~'_"+ . . . . . . " .Z~-5":". '2 ,gr2Z". ,+~E'~ ~'._~,_2Z'; rZ~ ,2 Z2"..'..'..TZ~ Z.".Z-..'.. . . . . .

. . . . . . . . . . . Jl=" . . . . . . . . . . . . . . . . . . . . . ..' . . . . . . . . . . . . . . . . . . . . .

~Eh.'.L~d^72EL"~2"~ ".:[[2"..'~Z::'I- yT..~2,;['TL'2Z:7.,~'. '"2:r..2~2~ ";;-.$.'2"~.:2~.'P~+'~I~,~7;'27"S~22,~:72,7. r3U:2=2"7_.?.292"~'22"..-22"3.W" g.'d7-Z?2::.2Z +^T,-,X~.2Z.2 aZ~.\E2,'J',..'2"~7.~g 7.'..'-Z'~,J2L-=~g'Z2T

. . . . . "+I . . . . . . . . . . . . . I " . . . . . . . . . . . . . . . . . . . . . . ! . . . . . . . . . .

6 1 6

N F P A 92B - - MAY 2 0 0 0 R O P

0 A ~ ]kA" . . . . . I A ^ + . ' = - - + . ~ - - - - A . . . . . . C . . . . . . I 1 . . - . ~ . ; ~ _ _ . . , 1

(a) ~" . . . . . . . . . . . . qua.n:ity "= d c t c r m ~ a c d . . . . . . . . . . . . . . . . . . . . . .

+ - - ^ I , ~ ~ I . . . . . . . k . . . . . . . + ^ A + ^ . ' - - + . . . . . + ~ I I . : A ^ . ~ . f ' +I++. . . . . . .

2£'~'3933 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . b " "

O A ~ I ~ _ _ . _ _ + . ' _ . ~ . J ~ . . . . . . . ' _ _ + : _ _ ~ . . . . . I ) . . . . . . +1 . . . . . I . ^

" _ . . 2 " , 2 L k ~ ' 7 7 2 " . . ~ 7~Z-,T2=;~Z,\',~3~.ZZL\-;.;U22:L."2R2"'-';;~ . . . . . . .

. I - . ^ . . . . . . . A . .1- . . . . . . . . . , , - I ^ - - , - - : . . . . + . . ^ I ^ . : ~ . ^ C - g : . . ^ ~I,,~^

: 22~ 72217,--- FJ7.1%~\~ :~ Ffi: 7:22-~" "e2~ ~ Z: :l ~';'~ E_'"'"Y'~

ca ! cu la~ .n . ~ . : ~ : ~ f := xhc m'=':=:u=n ~ : : -- : tczi ty.

2~..~ For zAjacen t =p=v.~ ~¢!=:-" t-he : m = k e !:}'=r "n'.e~ace, ~--n 7

1-. . . . . . . . . ^ . , 1 . ^ I * . ^ I . * - - * . . ^ A : - - I - . ^ - ^ 1 - . . , ^ - - . • ^ - - I . ^ _ . _ _ : _ _ + f . ^ _

2 g N r : ' : S ~ : : = S - ~ : - : ~ - : : g ~ . . ~ : V o ! - - - : EF: : : .

. l ~ | a l ~ l . . . . C . o l h . ~ . . I ~ . . . . . " . - - ~ | . . . . A A . * . . I - . . . . . . . . . . ' - - - - . ~ _ - - . . 1 . 3

The !~.~'-"cn ~f +-he er~=.".==t ~__:-chz=ge tc +-he ~u~:.a-e :h-~'-'!d ~e . . . . . . . . . . ! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2-1 Basic Considerations. The select ion of various design objectives and methods depends on the vrotect ion ~oals, such as pl-9~ecting egress paths, maintaining areas of refu~effacilitatin~ fire deoartment access, or orotectin~ Drooertv. Consideration nee~ls to be ~iven to the following:

(a/ The height, cross-sectional area. and plan area of the large volume to be protected. Height and area ~re kev e lements in the determinat ion of smoke accumulation, descent , and control.

(b l Tvoe and location of OCCUPancies within and communicat in~ with the large-volume space. The height, size. and arrangement of onenin~s between the occupancy within the communicat in~ soace and the large-volume soace are important considerations.

,~.,,-, ~ \ ~ - Z : T 2 : j v T . ~ 2 . . . . . - . ~

j . . . . . . . . . . . . . . . . j . . . . . . r

2 ~.~ ~ - c ": Cc , - -m- : ' ca t 'ag SFacca

O ~ O 1 0 " P t . . . . . K . . . . + - - . . * ^ ~ - . ^ - - . 1 . _ I . . . . . . . I . . . . . . . . . . . . A . * ^ k ^

: . 2 ; : ' 2 " : ~ 7 2 : ~ 2 k - ~ ' T Z T , , 3 : : ' _ . ' _ ' " . L - _ ~ 2~'"_"_'~, :g2~ ' : 2:72~..Z"_ - ~

: { 27 .Z ' 322~ :3 . Y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 .

cc~ 'n~= ~ f u p p : r ~ e ~ e.-2:m =_rid :h:'.: '! ~ - ~ : : c = : ' ~ : r : ~ . Th.ere is a . . . . : K I I I ~ , t K - - t t K . " . . . . I . . . . ~ 1 1 ^ - - + . . . . . . . . f l . . . .

. . . . . . . . . . • : . . . . . . . . . . A . 1 ~ - I ~ . . . . A + ~ : . . . . I . ^ - - : - - L . + ^ . - - : - - I ~ .

;:7'Z=22~'2"~Y%Z;2L" 2., ~ ' 7 . ~ -L ;5- ~-2 "L2;: f ; ih ~ . . . . . . ~ . . . . . . . . . ~ " "

" _ : ' : ~ o ~ " ZF : .~ ~.77":~,. U~Z\~. ~ . '~L .,-~.: Z ~ "L'f ~ _ " ~

. . . . . . +I~^ C . . . . . . . . C+I~ . . . . . . . . +^ - 0 1 . : . . . . . 1 - ' . . ^ - - J t ~ l ~ _ _ . ^ _ ~2 o

t ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ ................

- - . . . . - - ' - - . . . . . . . . . . . . . " I . . . . . / . . . . . . . l . . . . . . . . . . . . . . . . . .

to reach an exit or area of refu~e. (21 Maintainin~ the smoke laver interface to a t)redetermined

(31 Allowinlt fire deoartment t)ersonnel to atmroach, locate, and extintmish a i~e . - -

(41 -Limitln~ the rise of the smoke laver temt)erature and toxic ~as concentration, and reduction of visibilltv.

2-3 Deshm Limitations.

2-3.1 Smoke Accumulat ion Devth. The rate o f s m o k e laver ~ ~ l u m e s p a c e i s ojaly weaklyrelated. l~ t h ~ $ o f t h g . ~ a c e and l~e rate o f heat release o f the fire. S m o k . ¢ , . l ~ descenK_however, is stroogly rela~ed to the cross - sec t ion~ .a£g ,a .~ t h ~ l l ~ m e space i n ~ l v e d . For these reasons, careful calculations using the equat ioas and m e t h o d o l o ~ this d o c u m ent are necessary_ in any situation where the i a ~ n f i o n is to . p3.2vide smoke management througt:Lfl3e use o f a n u n e x h a n s ~ d _ volume such as a smoke coilectiou.K~ace.

The m i n i m u m d.ed~thof the s m o k e layer is de termined by_~oth the l h i c ~ s (drd~th)__Qf,_~e c e i l i n g i e t a s the rising_plume rams as it reaches the too of the svace and the d~d~thnecessarvto Dreven$ p_lugJa o l ~ g . For thes e reasons, nod_esigu s h o u l d b e b a s e d o n ~ g a smoke laver at a pQint higher than the level of,.fl3e ce i l ingi et Or the pgi nt o f eli m i n a t i o u o ~ g l ~ l i n g ~ _ ~ i c h e v e r is lower.

The thickness o ~ e cei l ing jet has been_r._e,~rted [B.e~.e~ C.(1986) . Fire Plumes and__CeilingJem. Fire SafetyJ o u m i ~ . l J . ~ l ~ _ 63-651 as in_t~e range of 10% to 20% 9f th.~_distance from the sou rce f~e ~ th e tOl~Of__~b~_laace.

617

N F P A 9 2 B - - M A Y 2 0 0 0 R O P

Plugbo l ing_ iL~e cond i t i on wbere air f rom below the i nm~ace is g g g e d throLivh a relatively s hal low_ smoke laver due to a h i ~ _ exhaus t rate atthi~£Doint. The impac t o f _ D l u g J ~ l i ~ be m a n a g e d (see Section 3-91,

$-3.2 Disrunt ion o f Smoke Laver I n t e r f a c e . Any factor that causes i n ~ r b u l e n c e i n o r i n ~ w _ i o t o the s m o k e laver o r a t t h e interface can affegU;be smoke laver. Amol~g3j3ese factors are t he following~

(a) O p e r a t i o n o f au tomat i c spr inklers above the smoke laver interface can draw the smoke belo~' the smoke laver interface.

.(.~) Stron~ air currents f rom I-[V~(~ systems or el ements o f t he stooge m a n a g e m e n t s vstem dis charged nea r the s m o k e laver interface can disrut~t th.c.Loterface so as to cause s m o k e to descend_ belo_wd~e s m o k e laver interface.

(c) A ~ cur ren ts a t over 200 f t / ra in (61 m / m in)_ s _ . L r . U ~ g _ ~ gl~.me below the interface can cause the_l~lume m b e n d ann_ ~ c r e a s ¢ the rate o~¢_0trainment air. cau~_iDg smoke to d e s c e n d belo_w_khe level calcula ted b z.th.e_.c_qua6om in this d o c u m e n t . The Ioca t io taof the fuel load. and the ootential p l u m e ~ o m s u c h fuel load. th.c.,olacement o f su~Rly_Doints, a n d the vein city at the s up/~lv p_.flints in re la t ion to the,QJ~me Iocadoo_.0eeds to be analyzed,_

(dL_U_.oward thrus ring airflows locatedbelo_w the interface having LUfficient m o m e n t u m to reach the layer can cause bo th mrbi ! lent mi~ng_t~ disruRt the i n , t r a c e and add m a s s to the smoke I aver to. cause t he layer to d e s c e n d b.~gw ~ e laver interface.

(e) Air forced._Qr in[Juced into t he ul212er layer by m e a n s o ther than t h ~ . l ~ m e will increase the mass in the u ~ e r laver caus ing th a_~J.~L~_~.d e s c e n d b e l ow thg,._desi ~ de_othunless COilll~ ens ated_ for i n t h e smoke m a n a g e m e n t s ~ t e m d e s i ~ x

2-3.3 S n e c i a l C o n s i d e r a t i o n s R e l a t e d to Natura l V e n t i n g . The canabilitv of buovant forces to move smoke t h rough a natural vf~t ~.-.::..<... • - . . - _ ......-.::- ~ ~ .:.;.:.: ~. is a f u n c u o n of both the deDth and t emnera tu re of the ho t lavgL '-.:~U:::.-':.:~'.:.:i~.:"~i The ~ravitv-induced m a s s flow t h r o u g h vents increases with " '~iii!.. '"::: increas in~ dep th a n d increasin~ t emnera tu re . T h e methodolo~,v % ..~! for assessin~r the mass flow t h r o u g h a vent is con ta ined i n _ . . ~ . ::~" ~04, Guide for Smoke and Heat Ve~tin~, ...#; .... -i~iili "%:,

_ _ ::..~....-,.-..~,~. ..p'.. ... '~:-.'~

No~a_a l l z~a tu rM and mechaui~;al vent ing a r e i n ~ ~ ' : " each o ther if thev serve t~¢ s a ~ ¢ ai r vo lume . Thex.e.is a ~ i ~ c a n t ....... p_Q~glial for a sho r t c l rcui t o f the airflow w h ~ ~ , . . , : ; ; : are reversed in flow directiol] 1;9 b e c o m e t t ~ % u r ~ ~ ~ i "~'" mechan ica l vents. AnLdes i t , n that c o n s ~ : = u c h a m ~ ve6~..~g m e t h o d s needs careful engil~eering.AIlat~sis"'~..:'.':':':~.h, ysical ( ~ l e ) m o d e l l m to ensure tha t th.e desit, rl will f u n c d ~ ! ~ i n t e .r~f~d.

Potenfi al envirolamental wind co n ~ fi o n ~ ~ nsiderat ioiL o f the impac t ofanx_gearb,Ll~ortions o f the b _ u ~ d j ~ , _ . q ~ s t r u c m r e s . ~ a t can cause d2_wn d ~ to_.Lbe eva lua~d in anx d e s i ~ d e p e n d e n t o n n a t ~ a l vents.

The mass of smoke is onlv weakly related to the rate of hea t felt'rise of the fire whereas the smoke laver temDerature nearly varies directlv with the rate of hea t release. Conseouent lv . a fire that is s i~ i f i can t lv smaller than the des ign fire wilf onlv n roduce a low t empera tu re smoke laver, with less mass flow than flaat of the design fire. However. less flow is necessary to provide vent in~ for the smaller fire. Fi~,ure 2-$.$(a) is an evaluation of the efficiency of mass flow th rough a vent where the indoor air t emnera tu re is the same as the ou tdoor t empera tu re . T h e fi~ure is fo rmula ted by keeDin~ all Darameters cons tan t ex ceDt the t empera tu re rise of the smoke [aver.

1 E 0 • 9

• i 0 . 8 0.7

g o.6 0.5

"~ 0.4

0 . 3 0

= 0.2 (/)

o.I

~ : 0 0

, , . i . . . . [ i I ' ' l . . . . i

1 O 0 200 300 400

T e m p e r a t u r e rise (°F)

Figure 2-3 .3(a) M a s s f l o w e f f i c i e n c y t h r o u g h a vent .

Al though Figure 2-3.3(a) indicates an aonreciable reduct ion in the effic[encv of a natural vent with small-fires Droducinu a m o d es t increase in smoke laver tempera ture• a small fire also Droduces less smok~, thereby reouir in~ less venting. Mjlke a n d Klote e v a l u a t ~ the effect of d i f f e r ~ : : h e a t OUtPUtS of fires on the vent area that is needed to ma i rk~"~"~kr t i cu l a r clear he igh t [1998]. This analysis indicates t h a ~ t z o u i r e d vent a rea is re(afivelv insensitive to the hea t o u t n u ~ f t h ~ i ~ e .

.::::... '::~i~., "::':@~ii-:~ .... F i ~ u d i ~ ! ~ ! i ~ ) den~ired vent areas to main ta in various

s m ~ interface laver h ~ t s for uiven fire sizes a n d ceilinu heights ~ ¢ ~ v s i s by M i ~ e and IZ~gf;¢,

":~:i:.::>....i:i:: ~ -.x~

70 -:%i%::::~ -::i:: 15 m 2.5 MW iili:: ::!'::" ~ 15 m 5 MW

%'-::~' ~""?:': .... l - - - 30 m, 2.5 MW / ----3orn, SMW /

4o 3o / /"

0 ~ ~ " t ' ~ " ~ ' , , ~ , . . ~ . . ~ I 0 6 1 2 1 8 2 4 3 0

C l e a r h e i g h t (m)

Figure 2-3 .3(b) V e n t a r e a r e q u i r e d to m a i n t a i n c l e a r he ight .

The effectiveness o f natural vents can be appreciably reduced or e l iminated where the ou tdoor air t emne ra tu r e is high. O n e scenar io of nar t icular concern involves a fire occurr in~ in a spBce with an indoor t empera tu re that is less t h a n the ou tdoor t empera tu re (i.e.. s u m m e r condi t ions with an a i r -condi t ioned a t r ium). While the smoke may be buoyant relative to the indoor air a n d rise to the ceiling, once the vent onens• ou tdoor air will en ter the bui ldin~ if the ou tdoor air t emne ra tu r e is ~reater than tha t of the smoke laver. As such. no smoke will be exhaus ted and the smoke laver interface can descend .

An examole of the l imitations of na tura l vents due to ou tdoor t empera tu re is indica ted in Figure 2-3.3(~;). In this example, an outdoor t emne ra tu r e of 100°F~($8°C) is assumed. The smoke layer t emnera tu re versus clear heit~ht is de t e rmined bv the euua t lon for

t emnera tu re rise (A-t) in the uDDer laver for a vented fire (see TaMe 3-5). for three fires with different hea t release rates. Where the smoke laver t empera tu re is less t h a n the ou tdoor tempera ture , no exhaus t is expected. As such• natural vent inu is no t a viab|¢ m e t h o d of smoke m a n a u e m e n t for a 2500 B t u / s e c (78 kW) fire where the in t ended cleat" he iuh t is L~reater t han 60 ft (18 m), Similarly. clear heiuhts ~reater thar~ 80 f t a n d 90 f t (24 m a n d 27 m)

618

N F P A 92B ~ MAY 2 0 0 0 R O P

canno t be achieved with natural vent ing for the 5000 B tu / sec and 7~00 B tu / sec (155 kW a nd 235 kW) fires.

600

5 0 0 " L L o

400

~_ s00

200

100

0 0

2500 Btu/sec t - - - - - 5000 Btu/sec

t

~ . . . . . . 7500 Btu/sec ~ ~ Outdoor temperature

~',~,,

I I I l I I I I I I I I I I I

20 40 60 80 100 120 140

Clear height (ft)

Figure 2-3:3 Limitations of natural vents due to outdoor temperature.

2-4 D e s ~ n Approaches . T h e design opt ions available for the dr;sign of ~mok~ m a n a g e m e n t d e p e n d on the space in which the smoke is to be m a n a g e d a n d the space in which smoke or i~nates . as descr ibed in 2-4.1 and 2-4.2. The des ign me thod , if any• for r f m o y i n g smoke f rom a space (mechanica l exhaus t versus natural vent ing) or corltailaing smoke to i~ space (airflow m e t h o d versus pl 'essurizafion m e t h o d ) needs to be considered.

~-4.1 Managemgl l t o f Smoke in a Large-Volume Space. A n u m b e r @:.::::., o f acceotable m e t h o d s exist for m a n a g i n g smoke from a fire "i#~ or ig inat ing in a large-volume snace. Table 2-4.1 summar izes the '?:i-':':: basic design cons idera t ions for-each of these methods , which ":i~ inc lude the following: .::~::::.-':.'.-~i:'~:?::'~::, - ... :.. ".:~.::::.:~

. . . . ,:i-i~: %" (a ) U u h z m g the large-volume space as a smoke r e ~ r an.d..-:ii~.:~::...~ ~

m o d e l i n g smoke laver descen t tO de t e rmine whetheU~h'~il ~o~:"":"-"-'-"-:~::'.~i:: laver interface r¢~ches a he igh t at which occunants are e~ ~ to '*:"-:*-~ ~llaoke before they are able to egress f rom the ~ ' :~: i :~:~. "-:?.-:!:-:.~!!-~. ¢,~

(b) Removitlg Smoke f rom the l a r g e - v o l ~ " s n a c e ' ~ r a "::!'!iii~ -~!;:':: ~ e c h a m c a l exhaus t caoao tv suff tcmnt tO.~t~nta ln the gi~ ake I ~ e r Interface at a n redef ined h e m h t m the soace~l : : .an mdefi iSte per iod of t ime. "~":$~ . . . . sff "(c~ ~emoving smoke from the ~ge -vo lume s ~ k : . ~ a llqechamcai exhaus t capao ty that slows the rate o f g ~ k e laver descen t for a Defied that allows occuoants to safelv:"~ress f rom the

(d) Providing natural vent ing sufficient to main ta in the smoke layer interface at a n redef ined he igh t in the space for an indefinite per iod of t ime.

(e) Providing llatural vent ing sufficient to slow the rate of smoke layer descen t for a period that allows occupants to .~afely e t r ess f rom the space.

Table ~[-~.1 Large Volulge S

5 o " v " e s . "

• " U . . . . .

Natural Vent ing with Cons tan t Laver Heigh t+

NOTE; Only algebraic calculat ion m e t h o d s are discussed with regard to each of the desi tm approaches listed in Table 2-4.1. Scale model ing , c o m p a r t m e n t fire mode l s (zone mode l sL or computa t iona l f luid dynamics (CFD) mode l s can be used to demons t r a t e each as out l ined elsewhere in this docume n t .

2-4,1,1 Smoke Filling Versus T i m e d _Egress Analysis. A m e t h o d for removing smoke f rom a large-volume space is n o t necessari ly needed if it can be demons t r a t ed tha t occupants are able to egress the ~pace safely before the smoke layer descends to the po in t at which the occupants are exposed to the smoke. Exposure can be in terms of p resence of smoke or tenability of the env i ro n m en t to which occupants are exposed.

A ¢otaservative es t imate o f the posit ion o f the smoke laver is the first indicat ion o f smoke, as shown in Figure 1-4. and as calculated us ing the empirically derived Equat ions (3) and (4) in Section 3-6. Equat ion (3) applies to s teady fires, and Equat ion (4) applies to uns teady fires, as de f ined in Sect ion 3-2. Equat ions (3) a n d (4) implicitly account for the t ranspor t lag associated with the m o v e m e n t of smoke f rom the fire into the uppe r layer.

Eauat ion (3) c a n n o t be combined with Equat ion (4) to calculate laver ~l¢~cent for ~ n g fires with a steady-state m a x i m u m . Each of these e q u a t i . ~ ' l s ' ~ i r i c a l l y derived a n d c a n n o t be used in c o m b i r l a t i o n ~ - ~ ' t h e other. Calculat ion of laver descen t for growing f i ~ ( w i t ~ ) L t e a d v - s t a t e m a x i m u m shou ld be accomnl i shed in a m ~ a ~ m i l a ? : ~ ' : ' t h a t descr ibed in Section 2-4.1.$. - ..:&::.::::~:~:~.-., -:~ -%-.:::.:~ .~:-

..~..;~. "-~:. ":~.~g. 24~!'2 Smoke Exhaust ~:"Achieve Constant Laver Height . A ~ . . e g r ~ . a n a l v s i s n~:~d no t be n e r f o r m e d if ' i t can be shown that the " ~ : : ~ ' ~ v interface is ma in t a ined at a he igh t so as to n o t C X O O S ~ : ~ 0 a n t s tO smoke for an indefini te he-rind of time. This is accom~i~t~edP'ff an exhaus t canacitv eoual to the volumetr ic ~ r .oduc t ion '~ ' smoke at the design laver interface he igh t is ~ , : ¢ - g t a t e d otherwise, t h e v o l u m e of smoke bei-ng in t roduced i~:to ~ m o k e laver is equal to the vo lume of smoke being r e m o ~ d by the mechanica l exhaust• In general , this m e t h o d t~ l ies strictly to steady fires, unless the oeak volumetr ic smoke

9~13~on is known for an uns teady fire-over the des ign per iod of s m o k e m a n a g e m e n t system operat ion. The volumetr ic smoke

woduct ion rate at a given laver interface he igh t can be calculated us ing Equat ions (8). (9). (10). a n d (15). The t empera tu re o f the smoke en te r ing the layer• calculated us ing the equat ions in Table 3- 5. need to be accoun ted for in calculating the smoke densi ty used

[ ~ ¢ ~ e Equat ions (~), (9). (10L an0, (15~ reference an interface he igh t co r resnond ing to the top of the t ransi t ion zone shown in Figure 1-4. a des ign interface he igh t needs to be selected tha t eilsurcs tha t occupants are no t exposed to smoke• W h e n selectirq_ this design interface he ight , the expected dep th of the t ransi t ion zone [leeds to be considered.

Exposure can be in terms of presence of smoke or tenability o f the env i ronmen t to which occupants are exposed•

race Smok~ Control Method~

st2agr_ady_y_~

~teadv

~moke Transvort

See NFPA 204

Natu I Ve ti vs. " s s" ~

Unsteadv Fire See NFPA '204 t An uns teady fire is no t an not ion for this ano roach because only a steady fire results in a cons tant laver height .

~ Icu la t io l l s

hint Necessary

~_qraur~

619

N F P A 92B - - MAY 2000 R O P

2-4.1.3" Smoke Exhaus t Versus T imed ~ Analvsls. Smoke exhaust can be used to slow the rate of smoke laver descen t for a t)eriod that allows occunants to safely e~ress f rom a snace. This annroach can be used where it is no t possible to nrovide an exiaaust canacitv sufficient to main ta in smoke at a-desima interface laver for axa indef ini te ner iod of t ime. In order to calculate the smoke laver Position over time. a t rans ient analysis needs tO be nef formed tfiat takes into accoun t the change in smoke n roduc t ion as a funct ion of the Dosition of the smoke laver in ter face as well the smoke removal t~rovided bv a mechanica l smoke exhaus t system. This ann roach is discussed in detail in A-2-4.1.3. Eouat ions (8). (9). (10). and ¢15) are used in to de t e rmine th¢ volumetr ic i nnu t of smoke into the smoke laver for a ~iven t ime steo. A specif ied auan t i tv of mechanica l smoke exhaust is then removed ~ o m the s m o k e laver over the same t ime sten. The new laver oosi t ion at the end of the t ime sten is then calculated. The t empera tu re of the smoke en te r ing the laver, calculated using the eouafions in Table 3-5. must be accounted for in calculat in~ th¢ smoke densi ty used in Eouat ion (22). Transt)ort lag associg~fd with the movement of smoke f rom the fire into the Ut)Der laver may or may no t be inc luded in this analysis. Imaoring t ramDort ]a~ f ields a more conservative resul t as the smoke is ] n s t a n t ~ e o u s l y added to the uppe r laver, resul t ing in a more ranid layer descent, The t ransnor t lag may be at)t)r eciable when cons ider ing fir¢~ irl soaces with l a rge areas. - -

Because Eouat ions (8). (9). (10L a nd (15) reference an interface he igh t co r resoond ing to the too of the t ransi t ion zone shown ill Figure 1-4. a des ign interface he igh t needs to be selected tha t ensures tha t occuoants are not exoosed to smoke. When selectjt3g this desima interface he ight , the exnec ted den th of the t ransi t ion zone needs to be considered.

~3] Providing an onnosed airflow over the face of the ooen ing to t )rohibi t smoke snread into the communica t i ng soace

(4) Promt)t lng sut)t)ression of the fire to ternainate the deve lonmen t of a hea ted smoke 9lt~me

2-4.1.6.1 Smoke exhaust can be t)rovided within the large-volume space to l imit the den th of smoke accumulat ion, or increase the t ime for smoke fil l in~ within the large-volume soace, so that the smoke laver in te r face remains above-the level ol ~ the highest ot)ening to communica t i ng spaces for the t ime necessary to achieve ~ e design objectives. This t echn ioue migh t no t be completely effective if the source of the fire is clirectlv adlacent ~9 ~ ¢ c o m m u n i c a t i n g st)ace. This m)oroach might- no t be feasible for communica t i ng spaces in the u p p e r t)ortion of the largf-vo|~pae

2-4.1.6.2 Smoke barriers can be orovided to l imit smoke spr¢~i into the communica t ing st)ace. D e n e n d i n g on the ex ten t of onen ings in the barrier~ a t)ressure d i f ferent ia l may n e e d to be anol ied-across the smoke barrier . This m e t h o d is discussed in I~bqPA 92A. Recommended Practice for Smoke-Control S~steras. A t)ressure differential can be achieved bv exhaus t ing the large-

Ext)osure can be in terms of t)resence of smoke or tenabi l i ty of the env i ronmen t to which occuoants are exoosed.

tha t snecified in 2-4.1.3 so as to slow the rate of smoke laver descen t for a ne r iod that allows occuoants to safely e~ress f rom a st)ace. This a-ooroach can be used where i t is no t oossible to nrovide mechanica l exhaust or natural vent ing of sufficient canacitv to main ta in smoke a t a des ign interface laver for an indef in i te t)eriod of t ime. In o rde r to calculate the smoke laver oosi t ion over time. a t rans ien t analysis needs to be ne r fo rmed tha t take~ il~t9 account the change in smoke t ) roduct ion as a func t ion of the nosi t lon of the smoke laver in terface as well as the smoke removal t)rovided bv natura l venting. A s imilar annroach is discussed in detail for mechanica l exhaus t (see A-2-4.1.3 ). The volumetr ic smoke removal nrovided bv natural ven t ing can be calculated us ing methods ou t l ined in NFPA 204. Guide for Smoke and Hea¢ Vgntn~.

2-4.1.6 M a n a g e m e n t o f S m o k e S n r e a d to C o m m u n i c a t i n g St)aces. M a n a g e m e n t of smoke sv read to c o m m u n i c a t i n g st)aces may be accom-olished by one o f ' t h e fol lowing m e t h o d s : - -

(1) Main ta in ing the smoke laver interface at a level h igher than t h a t of the h ighes t ooe n ing to the communica t ing SlP~,¢¢

(2) Providing a smoke barr ier to l imit smoke snread into the c o m m u n i c a t i n g snace (A nressure differential may need to be aonl ied across the smoke bar r ie r . )

2-4.2 Managemen t o f Smoke Within Communica t ing Sp~¢~8,

24.2.1 Fire in Spaces Sur round ing a I a ~ e - V o l u m e Spact h Possible confi ta l ra t ions for the relationshi-n be tween the largf- vo lume snace and the s u r r o u n d i n g soaces- include the following.

(1) Seoara ted soace (2) C o m m u n i c a t i n g snace

2-4.2.2 Fh'e in Setmrated_ Snaces. Where construct ion eeparat ing the large-volume snace f rom the su r round i ng areas is suflficientlv t ight s o t h a t the pressure differences between-the fire zone and the nonf i re zones can be contxolled, the large-volume snace can be t rea ted as one of the zones in a zoned smoke-control sy~ggrn. Zoned smoke-control svstems are descr ibed in NFPA 92A. Recontmend~ Practice for Smoke-Control S~stems.

2-4.2.3 F'we in Communica t ing Snaces. C o m m u n i c a t i n g spaces can be des igned to allow the smoke to snHI into the large-volume snace. In tills instance, the smoke svi l l ing in to the lar~e-vpltamf snaee should be hand l ed by the smoke m a n a g e m e n t system, which is n rovided to main ta in the desRm smoke laver interface height . C o m m u n i c a t i n g snaces can also-be des i tmed to nrevent the

v _

m o v e m e n t of smoke in to the large-volume snace. Such a ~f~it~n would reoui re sufficient exhaus t f rom the communica t i ng space so as to establish a m i n i m u m flow between i t and the large-volume

2-4.2.3.1 Exhaus t Through a Large-Volume Snace. For fires in unsDrinklered svaces, the exhaus t rate f rom the large-volume sDace needs to be evaluated no t only for a free n lume from a fire in the large-volume snace bu t also for a n l u m e ork, ina t ing in the

620

N F P A 92]8 - - MAY 2 0 0 0 R O P

c o m m u n l c a t i n ~ suace. T h e s m o k e m a n a g e m e n t system shou ld he able to hand le -e i the r condi t ion , b u t n o t I ~ t h s imultaneonsiv. The m e t h o d s for calculat ing the volumetr ic s m o k e oroduc t ion fo r soill v l u m e s a n d window ~ l umes a re discussed in $-8.9 a n d ~8.$. resnectiv~lv. T h e em~ations in ~-8.~ a n d $-8.g are only valid for f i r ~ in uiasnr inklered snaces as they w e r e d e r i v e d empirically ~rom test data.! (3~ce smoke e n t e r j the large-volume snace, the ~ l l } ~ 2 ~ L ~ [ ~ s m o k e c u d i n ~ back-onto u n n e r f l o o r s or imnin~in~ on overhan t6ng ceilings o f u o o e r t ] o o r s exists a n d shou ld be cousidered.-TlT~ere is a nossil~i-litv t ha t this maoke will en t e r uppe r floors o f communica t fn~ s_nacea, a n d t h f hazard this smoke mit~nt or m i g h t n o t p r e sen t t 9 these epaces s h p u | d be e ~ u a t e d .

2-4.2.3.2 Con t a inmen t o f Smoke to Communlca t in~ Snaces. C o m m u n i c a 6 n g soaces can also be deslt , n e d to orevent the m o v e m e n t o f smoke in to the large-volume rmace. Such a denims would renu i re suff icient exhaus t f rom the c o m m u n i c a t i n g spac¢ so as to e s t a~ i sh a m i n i m u m flow between it a n d the large-volume svace. T h e face velocltv across t he face a rea of the o ~ n l n g that achieves this is descr ibed in 2-4.1.6.$. and Chan te r $ ~rovic[e~ calculat ion m e t h o d s for smoke gene ra t ion in the c o m m u n i c a t l n ¢ soace. T h e exhaus t ouant i tv ne~-ssarv for this s i tuat ion can re, early exceed the canacitv of t he n o r m a l bu i ld ing HVAC s~Tl:lI~ a n d can reuu i re the instal lat ion o f a ded ica ted smo~ke n m n a o e m e n t system for - the c o m m u n i c a t i n g soace.

su r ead l r~ m areas outs ide the large-volume fa3ace. T h e following events need to occur to accomnlls-h these m~ls .

(a~ T h e fire needs to be de tec ted early t 'before the s m o k e level or rate o f descen t exceeds~the des ign oblecfivesL W h e r e t h e sm o k e m a n a g e m e n t sys tem is nrovided- to assist safe evacua t ion o ccu n an t react ion t ime t'o the en~-r~encv a n d e ~ c o a t i o n t i m e s h o u l d ! ~

(b) The HVAC ss s t em serving t he large-volume suace a n d c o m m u n i c a t l n ~ sna re s needs to" be s topped ff its 01~ecation would adversely affec~ tl~e smoke m a n a g e m e n t s ~ t e m .

(c] Smoke shou ld be r e m o v e d - f r o m the l a r t ~ v o l u m e space above the des i red smoke laver interface.

(d] Sufficient m a k e u n air shou ld be nrovided to satisfy the exhaust . It is essential flint t he m a k e u n air s u n , I v inlet and the exhaus t out le t be senara ted so tha t t he contan-~nated air is n o t drawn into the buiiclinm

2-5.2 Automat ic Activation. T h e confkrura t ion o f the l a r ~ - y o l u m e snace shou ld be cons idered in selectin~ t he .type o f de tec tor to be

to activate the smoke m a n a g e m e n t smtem. T h e size. sha~e. and he igh t o f the snace n e e d to be evaluated. ; these factont ~ - v

The n l a c e m e n t of the exhaus t o o e n i n ~ shou ld he evaluated carefu|lv. Exhaus t intake a n d d i ~ h a r t , e~oneninaa shou l d be l o c a t e d so tha t smoke m o v e m e n t will n o t in ter fere with exit~ The locat ion o f the exhaus t d ischarge to t he outs ide shou ld be located away f rom outs ide air intakes to min imize the l ikelihood o f s m o k e be in~ recircolated. Smoke barriers can also be provided between the large-volume m a c e and c o m m u n i c a t i n g suaces. Where cons t ruc t lon semr~fin~r the l a r t ~ v o l u m e st~ace f rom t he s u r r o u n d i n ~ ar~as is sufficlently t lgh t so tha t the presm,tre

Control S~stems.

cons idered in t he analysis i n d u d e the following:

(3~ Smoke temoera tu re

The de t e rmina t ion o f s m o k e toxicity usually inc ludes the anal@i~ of exoosure to ca rhon m o n o x i d e fC OL Fxru~ure to o t he r fneJ- d e o e n d e n t toxic wages can also be considered. Ten~h;~i¢/l imits for bo th smoke toxicity, a n d s m o k e tem_~u,m~r ~ nam~llv con~id~r t he t ime of e x n m u r e to the smoke.

The calculations ner ta in in~ to the de t e rmina t ion o f visibili W dis tance a re cl iscu~ed in A--~-5. An evahmtlon o f the effects o f smoke on neon le d u e to smoke toxicity a n d s m o k e t e m n e r a m r e is outs ide the scoue o f this tmide. However. as statPd in I-5.6. o ther re ferences are available tha t n r e sen t analvtlcal m e t h o d s for ten~thilltv ~m~t|vges [~41.

9.fi Smoke Manam~naent Svsmm Onera t inn .

24L! s m o k e n m n a ~ e m e n t m~tems for large-volume s_rme~ are i n t e n d e d to restrict t he smoke laver to the u n n e r oor t ion o f t h r large-volume snace or to l imit t he a m o u n t o-f-smc~ke f rom

2~;.2.2 Normally. all au tomat ic de t ec t i on devices within the lame- vo lume snare_ and c o m m u n i c a t i n g _maces s h o u l d activate the, smoke m a n ~ e m e n t sVW'rfi. Detpctors for fo,,~-ial nurDose~ such as elevator recall and doo r r e l , ' ~ P a n d for -_~poc;6c bar~rds~ liceh as sueda l f i re-ext in~ulshlng a~JemL can he excenf iom. In order to avoid unnecmsarTv oner-a~ti0n o f t he sys tem f rom smoke d~ tpe to r activation, considera t ion shou ld he t, lven to activatin~ the matem by two or m o r e smoke detec tors or on-alarm-verlf ication.

Automat ic de tec t ion devices s h o u l d n o t he conner t , -d directly to the smoke mana~ ,ement swtom wi thout f i i r ther conce rn for tht; inte~rltv of t he de tec t ion system. Intem4t¢ o f t he det,-cti0n system is

• 4LY..~ Snot. tvne smoke detectors can be mu~! on or n e a r low ceilinwa o f l a ~ e - v b l u m e snares, n rov lded tha t t he d,.t,.ctora are a c c e ~ b l e for-servicln~ a ~ ! nos i t loned ~ on cons idera t ion o f the effects o f stratifies/ion a n d air cur renm ~nu,¢l ~ n~ttnrai mad mtghaaL~.fam~

2~;.2.4 Projected beam4vne smoke detec tors can be u sed on or nea r h ~ h ce i l i n~ o f lar~g-volume sna re s and po*i t lnned to m'olect the beam horlzontal lv o r i n o the r accentable orientatiom~ SUmltlcat ion a n d natura l o r mechan ica l air cur ren ts can n eem l t~ t~ the use o f addl t ional ~roiected b e a m s a t in t e r im levels o f the l a r ~ - vo lume snace where c, eillnc, heiMats would cont r ibute to t h e delay in initiating_ smoig¢ manacement~

2~.2 .5 Automat ic sur inkler water flow shou ld a l ~ be used to activate the smoke m a n a g e m e n t ~ t ~ ' n . It is i m n o r m n t tha t tho sur inkler s~tmla he zoned with the smoke de tec t ion s~at*pm in t h e i~trge-volume snace so tha t the correc t smoke m ~ e m e n ¢ re snouse is effected. T h e hei~dht o f t he large-volume *paCe a n d the-

621

N F P A 92B - - MAY 2 0 0 0 R O P

location of snrinklers shou ld be analyzed in order to es t imate snr inkler activation resnonse t ime. Snrinkler activation t ime o n be too slow to effectively initiate smoke m a n a g e m e n t where snrinlders are located several stories above the floor of the snace. Tl~e eaua t ions of Chante r ~ shou ld be used to analyze eaqh case. Snrinlder water flow shou ld never theless be one of the smoke m a n a g e m e n t svstem initiating means , even if only as a backup system. Snrinkler activation can nrovide an effective t~rimarv initiation m e a n s where sorinklers are located on lower ceilings.

2-5.3 Manual Activation. A m e a n s of manua l ly start in~ and s t unn ing the smoke m a n a g e m e n t system shou ld be located so as to be -accessible ~o the fire d e n m x m e n t .

2-6 Smoke Manastemeut System Reliability.

2-6.1 Fault Analysis. Every smoke m a n a g e m e n t system shou ld I~¢ subjected to a fault analysis so as to determine: the irqpact Of a failure, i m n r o o e r oneraf ion, or partial onera t ion of e~ch major svstem c o m n o n e n t on i n t ended svstem onerat ion. Of parti~;ul~r concern are those systems tha t are in t ended t9 maintg in a nr¢~sure or flow balance between adiacent soaces to control the m o v e m e n t of smoke. Should it be f ound tha t i_he faulty ooera t ion of a c o m n o n e n t will cause reversal o f the s m o k e flow or lowering o f the smol~e interface laver to dange rous levels, the degree to which its onera t ion can be r educed an~ the prgbabili~y of such occurr~oce sfiould be de te rmined .

2-6.2 ll.eliabiUty. Reliability of the smoke m a n a g e m e n g sy s t e~ d e n e n d s on the soecific reliabilitv of the individual comnonen t s . funct ional d e p e n d e n c e o f the c o m n o n e n t s on one a n o ~ e r , and degree of redundancy . Reliability of the individual c o m n o n e n t s (i,¢,, hardware, softwi~re, a n d interfaqes with 9 ther sv~ems~ involves both their ne r fo rmance du r ing normal onera t ing condi t ions as affected bv env i ronmenta l factors, over the-life of the system, and their ability to withstand the stresses e n d u r e d durin~r a

the system is onerat ional and will reliably ne r fo rm when needed . Means shou ld be nrovided for oer formin~-ner iodic tests of the smok~ m a n a g e m e n t system in order to verify the systerq pefformance~ Systems shou ld be des igned to pe rmi t t;¢sting wi thout any snecial euu iDment o ther t h a n what is provided with the

Because access for ne r fo rmance verification m e a s u r e m e n t s is oft~la difficult, it is glesiral~l¢ that. where possible. i n s t rumen ta t ion be comnlete lv built-in or partially built-in a n d nartiallv nrovided as portable moni tors .

Chapte r 3 Calculation Procedures

3-1 Introduction.

3.1.1 Design Approaches . Three d i f ferent smoke m a n a g e m e n t system des ign approaches are descr ibed as follows:

(a) Scale mode l i ng us ing a r educed scale physical mode l following establ ished scaling laws. Small-scale tests are conduc t ed to de te rmine the r equ i r emen t s and capabilities of the mode led smoke m a n a g e m e n t system.

(b) Algebraic, closed-form equat ions derived primari ly f rom the correlat ion o f large- a n d small-scale exper imenta l results.

(c) C o m p a r t m e n t fire mode l s us ing both theory and empirically derived values to est imate condi t ions in a space.

"Each approach has values a n d limitations. None is totally satisfactory. While the results ob ta ined f rom the different approaches shou ld normal ly be similar, they are n o t usually identical. The state of the art involved, while advanced, is empirically based, a n d a final t heo ryp rovab le in f u n d a m e n t a l physics has no t yet been developed. The core of each o f the calculat ion m e t h o d s is based on the e n t r a i n m e n t of air (or o ther s u r r o u n d i n g gases) into the rising fire-driven p lume. A variation of approximate ly 20 pe rcen t in e n t r a i n m e n t occurs between the empirically derived e n t r a i n m e n t equat ions c o m m o n l y used, such as those indicated in this chapter , or in zone-type c o m p a r t m e n t fire models . Users m igh t wish to add an appropr ia te surety factor to exhaus t capacities to accoun t for this uncertainty. A brief discussion o f the values of the several approaches follows.

3-1.1.1 Scale Modeling. Scale mode l i ng is especially desirable where the space be ing evaluated has project ions or o ther unusua l a r r a n g e m e n t s tha t p reven t a free-rising p lume. In a scale model , the mode l is normal ly propor t iona l in all d imens io n s to the actual building. The size of the fire a n d the in terpre ta t ion of the results are, however, governed by the scaling laws, as given in 3-1.2. A l though sound , the app roach is expensive, t ime-consuming , and valid only within the range o f tests conducted . Because this approach is usually reserved for complex s t ructures , it is impor tan t tha t the test series cover all o f the potential variations in factors such as posit ion an d size of fire, location a n d capacity of exhaus t and intake f l o w s , . . , . ~ o m in internal t empe ra tu r e (stratification or f loor-cei l in .~.~. .~pi~ture gradients) , a n d o the r variables. It is likely tha t ~ : : . w i l l n o t be appralsable us ing scale models .

3-1.1.2 ~ . r a i c " ' ~ i o n s . Algebraic equat ions , as conta ined in this ~ ' ~ d e afS~_o._. ~ m e a n s o f calculat ing individual

that ~ 'b l lec t i . . .1 .~- -~e used to establish the design r, of a smo l~"h system.

g f a c t ~ re~rem~ '~a t s m n a g e m e n t T h e equat ions t ~ . . ~ e d . ~ cons idered to be the mos t accurate, s ,mple, a l g e " ~ x p l ~ s s m n s available for the p roposed purposes . In ge~ner~..~.~y are l imited to cases involving fires tha t b u rn at a c o n s t a n t ° ~ . . . ~ h e a t release ("steady fires" as descr ibed in 3-2.2) or ~ . ~ s tha t ir i : ~ a s e in rate o f hea t release as a func t ion of the square

. ~ . ~ . ~ ' s t e a d y fires" as descr ibed in 3-2.3). T h e equat ions are ~ 7 $ ' ~ $ i ' p r i a t e for o ther fire condi t ions or for a condi t ion that ~:nitial~'grows as a func t ion of t ime but after r each ing a m a x i m u m , bu rns at a steady state. In mos t cases, jud ic ious use o f the

~ at ions can reasonably overcome this l imitation. Each of the at ions has been derived f rom exper imenta l data. In some

cases, there is only l imited test da ta a n d / o r the da ta has been collected within a l imi ted set o f fire sizes, space d imens ions , or points of m e a s u r e m e n t . Where possible, c o m m e n t s are inc luded on the range of da ta used in deriving the equat ions presented. It is impor tan t to consider these limits.

Caut ion shou ld be exercised in us ing the equat ions to solve the variables o ther than the ones p resen ted to the left of the equal sign, unless it is clear how sensitive the result is to m ino r changes in any of the variables involved. Where these restrictions p resen t a l imit that obstructs the users ' needs , cons idera t ion shou ld be given to combin ing the use of equations with ei ther scale or c o m p a r t m e n t fire models. Users of the equat ions shou ld apprecia te the sensitivity of changes in the variables be ing solved for.

3-1.1.3" C o m p a r t m e n t Fire Models . Compu te r capabilities sufficient to execute some of the family of c o m p a r t m e n t fire models are widely available. All c o m p a r t m e n t fire models solve the conservat ion equa t ions for dist inct reg ions (control volumes) . C o m p a r t m e n t fire mode l s can be general ly classed as zone models or f i e ld - (computa t iona l f luid dynamics-)- (CFD) models .

3.1.1.$.1 Zone Models . Zone models are the s impler models and can usually, be run on personal computers . Zone mode l s divide the space into two zones, an uppe r zone tha t conta ins the smoke and ho t [~ases p roduced by the fire a n d a lower zone, which is the source o t e n t r a i n m e n t air. The sizes of the two zones vary dur ing the course of a fire, d e p e n d i n g on the rate of flow f rom the lower to the uppe r zone, the rate o f exhaus t o f the uppe r zone, an d the t empera tu re o f the smoke a n d gases in the uppe r zone. Became of the small n u m b e r o f zones, zone mode l s use eng inee r ing equat ions for hea t a n d mass t raasfer to evaluate the transfer of mass an d energy f rom the lower to the uppe r zone, the hea t an d mass losses f rom the uppe r zone, and o ther features. Generally, the equat ions assume that condi t ions are un i fo rm in each respective zone.

In zone models , the source of the flow into the u p p e r zone is the fire p lume. All zone mode l s have a p lume equat ion. A few models

622

N F P A 92B - - MAY 2 0 0 0 R O P

allow the user to select 3Jnong several p lume equat ions. Most cu r r en t zone mode l s are based on an axisymmetr ic p lume.

Because p resen t zone mode l s a ssume tha t the re is no pre-existing t empera tu re variation in v_he space, they canno t direcdy handle stratification. Zone mode l s also a s sume tha t the ceiling smoke layer forms ins tandy and evenly f rom wall to wall. This fails to accoun t for the initial lateral flow of smoke across the ceiling. The resul t ing error can be signif icant in spaces having large ceiling areas.

Zone models can, however, calculate many impor tan t factors in the course of events (e .g , smoke level, t empera ture , composi t ion, a n d rate of descent) f rom any fire tha t the user can describe. Most zone mode l s will calculate the ex ten t of hea t loss to the space boundar ies . Several models will calculate the impact of vents or

mechanica l exhaust , a n d some will predict the response of heat- or smoke-ac tua ted ,detection systems.

3-1.1.3.2 ~ CFD Models . CFD ~ - - ~ models , also referred to as . . . . . . . . . : ^ ~ 1 ~..:.~ ~ . . . . . :~" (C.r~D~) field models , usually require large-capacity c o m p u t e r workstations or ma i n f r ame compute r s and advanced expertise to opera te and interpret . CFD ~ ! ~ models , however, can potentially overcome the l imitat ions of zone mode l s and c o m p l e m e n t or supp lan t scale models .

As with zone models , CFD_f ie~ mode l s solve the fundamen ta l conservat ion equations. In CFD field models , t he space is divided into m a n y cells (or zone.,;) :rod use the conservat ion equat ions to solve the m o v e m e n t of heat and mass between the zones. Because of the massive n u m b e r of zones, CFD field~ mode l s avoid the more general ized eng inee r ing equat ions used in zone models. T h r o u g h the use of small cells, C.CFD field models can examine the s i tuat ion in m u c h greater detail aztd accoun t for the impact or i rregular shapes and unusua l air m o v e m e n t s tha t c anno t be addressed by ei ther zone mode l s or algebraic equat ions. T h e level of r e f inemen t exceeds that which can usually be observed or derived f rom scale models .

3-1.2 Scale Models .

3-1.2.1" In this guide the emphas i s o f scaling activities is placed on mode l ing ho t gas m o v e m e n t t h rough bui ld ing conf igurat ions due to fire. Combus t ion and f lame radiat ion p h e n o m e n a are ig~:.~g.:...c:d. Fire growth is no t mode led . A fire needs to be s p e c f f i e ~ of a steady or rime-varying hea t release r a t e . . ~ . , : ~ : ~ * ~ . . . . :.."~i!!~::..:x.

3-1.2.2 Based on the relat ionships in Table 3-1.2.2, a sc'~.'{~,~.fl/i~lel'~.~ can be developed. T h e mode l shou ld be m a d e l ~ L e n o " ~ i ~ achieve t u rbu len t flow of the full-scale system~..,:~mm~X.:,c.x re la t ing full-scale condi t ions (F) to those i . ~ . ~ : c a l ~ l pr~s=:':~(m)~O ~" presen ted in Table 3-1.2.2, condi t ions exist.

Table 3-1.2.2

Geometr ic posi t ion T e m p e r a t u r e Pressure difference

Velocity Total hea t release rate Convective hea t release rate Volumetr ic exhaus t rate The rma l propert ies of enc losure

x , = x F t ~,,/~'1 ":~i!i~"" T= = T F ..:i?.Y:" Ap= = App ( IJlp) v, = ~ ~ ( t.lt~) '/~

= ~ (t,,Ig,~?/~ Q, ., ~..~- = Q~, O~3- ( t . l t , ) ~/~ ~:. ~ = Q,~ ~ # ( t . , I t , W ~ (kp~)~,., = ( kpO ~,j, (t.llF) °'~

where: c = specific hea t of enclosure materials (wall, ceiling) k = the rmal conductivity of enclosure materials (wall, ceiling) l = length Ap = pressure difference Q = hea t release rate t = t ime T = t empera tu re (ambien t and smoke) v = velocity V = volumetric exhaus t rate x = posit ion

~_ = density c = convective F = full-scale m = small-scale mode l

w = wall

3-1.3 The r ema inde r of this chapter presents the algebraic equat ion-based calculat ion p rocedures for the various design parameters , as re fer red to in the previous sections. The calculation p rocedures r ep resen t an accepted set of algebraic equat ions an d related in format ion available for this edi t ion of the guide.

3-1.4 Es tabl i shment o f Two-Layer Envi ronment . A delay in activating exhaus t fans can allow smoke to descend below the des ign he igh t of the smoke interface. Initial smoke accumula t ion at low levels can also be aggravated by initial vertical t empera tu re stratifications tha t delay t ranspor t o f smoke to the uppe r reaches of the large volume s p a c e . However, with the exhaus t and air makeup systems activated, a clear lower layer can be expected to develop in a g r e e m e n t with the des ign assumptions .

3-1.5 SI Units. SI forms of the equat ions conta ined in this chapter are p resen ted in Append ix D.

3-2 Design Fire.

3-2.1" All of the des ign calculations p resen ted in this guide are d e p e n d e n t on the hea t release rate f rom the fire. Thus , as a first step, the des ign fire size needs to be identified. T h e design fire size is d e t e r m i n e d based on an eng inee r ing analysis of the characteristics of the fuel a n d / o r effects i nduced by a fire. In addit ion, fires can he cons idered as steady or unsteady.

3-2.2 Steady F~.:~ ~ e a d y fire is de f ined as a fire with a cons tant . . : . ~ . .

hea t release g~,-X{bs such, the fire is expected to grow qmckly to some l i m i t , : : i ~ . ~ u ~ x t e n s i o n is restricted ei ther due to fire control ~ e s ( ~ a l or automat ic) or a suff icient separa t ion d i s t a n : ¢ ~ ~ ~ . ~ b u s t i b l e s be ing p r e s e n t

* l~:~ect o f S p r i ~ e r s on Fire Size. Unless there is reason ~ " : ~ c t : . - ~ : f i r e will con t inue to spread after sprinkler activation, t h e | ~ ~ . ~ f S~rinklers on the des ign fire size can be accoun ted for by ass d ' ~ g tha t the fire stops growing when sprinklers are actuated. ~ ' ~ e r words, the design fire is the es t imated fire size at ~:.¢.. momeri~g.-~tgf sprinkler actuation. It is a s sumed tha t the fire ' : ~ . ¢ . S : . _ 4 6 bu rn at this size until t he involved fuel is consumed , ~ ! ~ ' ; ~ 1 1 e r effect of the sprinkler spray on the b u r n i n g process.

..... " However, if tests for ~e prevail ing ceiling he igh t show tha t fire in the combust ible Laterial will be quickly suppressed with the instal led sprinkler

protect ion, combus t ion can be a s s u m e d essentially to cease when the sprinklers operate.

3-2.2.2 Separat ion Distance. The des ign fire shou ld be d e t e r m i n e d by considef in~ the type o f fuel, fuel spacing, an d configurat ion. The selection of the des ign fire shou ld start with a de te rmina t ion of the base fuel package, tha t is, the m a x i m u m probable size fuel package that is likely to be involved in fire. The design fire shou ld be increased if o ther combust ibles are within the separa t ion distance, R, indica ted in Figure 3-2.2.2(a) and de t e rmined f rom Equat ion (1). Note tha t if the base fuel package is no t circular, an equivalent radius needs to be calculated by equa t ing the floor area covered by the fuel package with tha t sub t ended by a circle of the equivalent radius. The ent ire floor area covered or inc luded between commodi t ies shou ld be cons idered in the calculations, for example , if the fuel package consists of the furn i tu re i tems il lustrated in Figure 3-2.2.2(b), the area of the fuel package includes tha t covered by the furn i ture as well as the area between the furn i tu re items.

R = [ Q / ( 1 2 ~ t q " ) ] ' /2 (1)

where: R = separa t ion dis tance f rom target to center of fuel package (ft) Q= hea t release rate f rom fire (Btu/sec)

q" = inc ident radiant hea t flux requi red for nonpi lo ted ignit ion (Btu/fff . see)

623

N F P A 92B - - MAY 2000 R O P

Hemisphere

E l e m e n t o r i e n t e d

F i g u r e 3 - 2 . 2 . 2 ( a ) S e p a r a t i o n d i s t a n c e , R .

\ F u e l

< ' - - i t e m s / :; . . . . . . . . . [

: + . . . . . . . . . . . . . I - . ~ + - ^1 . . . . . + ^ C^ . + I ~ ^ . 3 ^ . : ~ C . . ^ I . . . . . . . 1.-..~

~ l , : V ~ : ~ n + :+ On ~*11 / ¢ * + + + + ra m+~ ~ A ,i.+ t}+,,/c+~-++~ c . . . . . . . . *~1+

nee l l nnne ; - - ^ I l l ^ - - A . - - . : J ^ - - + ; - - I . . . . . . . . . . . + ; . , ^ I . . [ C 1

3-2.5 Min imum Desi~,n F i r e S i z e C a u t i o n . D e s ! ~ z ~ e ~ c r ~ m . . . . . . . . + . . . . I . . . . . . + : ~ . . I g . . . . . . . . . . + . ~ - - + + k ~ + C . . . . . . . k . . ^ + : k l . . . . ,=ll

t.+ . . . . . . . . . . . : - + " " ~ ^ ' ~ in t~c s~:cc, tkcrc~ 7 Emi~ng ~ c mtc +f .hc : t r c l c = e . . . . . . . . . . . . . . . n-.'g~ . . . . . . - - - - " ^

~ . . . : ~ . + : _ ~ e ~ _ ^,, f .a.ay m--n' .a!n du r ing "d~c lifc of " ~ c . . . . . . . ~, . . . . . f i m c s o , da}~ of the ;reck, or z e n a n a af Lhc }'ca.-.

3-2,5,1 The selected desi_~a fire size shou ld represen t a credible worst case scenario . Desigmers a n d analysts are s t rongly caut ioned ~gainst selecting modes t fire sizes based solely on the type or l imited a m o u n t of combust ib les tha t are present or expec ted .

3-2.5.2 In low ceiling spaces (ceiling h e i g h t less than 25 ft) where sprinklers are provided, the design fire consists ei ther of a steady design fire or a fire tha t grows to some steady th reshold size. for example , due to opera t ion of an au tomat ic suppress ion system.

3-2.5.3 In h igh ceiling spaces (ceiling he igh t at least 25 ft) where

3 -3 and Sprinkler A c t u a t i o n .

Figure 3-2.2.2(b) F u e l i tems.

3-2.2.3 Design Fire Size. Specification of a f ixed desigt~:*.:~ :+":+;~+++ "e s i~.'..~ applicable to all s i tuat ions is no t realistic. T he type ~ nou . ~ o f fuel shou ld be cons idered when de t e rmin ing the c l e s ' l ~ " ;!:" +¢ s ~ i Further , a s t andard size des ign fire canno t be r e c o m m e n ~ ~.~. ue t'~ the lack of available da ta in North Amer ica to i ~ . . q tha +~+++-+. des ign fire is only exceeded in a l imited pro. :~+."P~h+6"fi~ .+ ~i ~.+-+,.+++ inc luding e i ther atria or covered malls. .:..:~ . . . . . +;-'~% "T.-m"

++.:++-.'+'.:~ :'~,.. ~ '~ i • 3-2.3 Uns teady Fires. An uns teady fire ++ ~ . . a t varies:~ ith respect to t ime. A t -squared profile is of ten ~ . e d for.$ 6steady fires. Then , the hea t release rate at any t ime is ~ ' ~ b~..-'$+Pquation (2):

Q= lO00 (t/ t~) ~ (2)

where: Q = hea t release rate f rom fire (Btu /sec) t = t ime after effective ignition (see) tg = .growth t ime (see) g = u m e interval f rom the t ime of effective ignition until the fire

exceeds 1000 Btu/sec .

See Append ix C for fu r ther informat ion on t-squared profile fires.

Due to the dynamics of secondary ignitions, a t -squared profile can be used for eng inee r ing purposes until large areas become involved. Thus , a t -squared profile is reasonable up until the fire growth is+ l imited e i ther by fire control activities or a sufficient separa t ion dis tance to ne ighbor ing combust ibles to prevent fu r ther ignition. After this dine, it is a s sumed tha t the fire does no t increase in size.

3-2.4 Data Sources for Hea t Release Rate.

3-2.4.1 Recendy, a l imited a m o u n t of hea t release rate da ta for some fuel commodi t ies have been repor ted [2,3]. (See Appendix B.) However, fu rn i tu re cons t ruc t ion details and materials are known to substantially inf luence the peak hea t release rate, such tha t hea t release rate da ta are no t available for all furni ture i tems nor for "generic" furn i tu re items.

can be est imated f rom the t empera tu re rise g e n e l ~ e ~ by t~¢ fire at those locations. Tbe t empera tu re rise d e p e n d s on the vertical dis tance above the base of the fire a n d the radius f r o m the fire center l ine axis. NFPA 72. National Fire Alarm Code. provides a p rocedure for de t e rmin ing hea t de tec tor spac ing (for he igh ts less than $0 ft) based on the size and growth rate of the fire to be detected, various ceiling heights , and a m b i e n t tempera tures . The under ly ing theories , assumpt ions , l imitations, and known a n d potential sources of errors for es t imat ing the response t ime of smoke and hea t detectors are identif ied and discussed in ISchifiliti & Pucci l . An e n ~ n e e r i n g analysis is n e e d e d for ceiling heights greater than ~0 ft.

f r+m c+..'~..=+n Pads +~mt ::'+'-!J c~u:c dctcc+-+n ~)" a rca=+nm~! 7 +~-~.- . ~°°-+:+;'- . . . . ,+ . . . . . . . : . . . . . . . : . . . . ,., o n o ~ ~ . . . . . . : . . . . h, -+-,ln°+'~.

+ P " . . . . . . . . . . . . . ~ ' I : " . . . . . . . . . . . . l ~ + " x + l " t " . . . . . . . . . . . . I

~ o q * f ' ^ : t : - - P.r . . . . . . .z c ' : _ ^ . a T . . . . . . . . . . r~._^.~__.+__.v.. ~3Jith no

?Jlo;'An~ e^- +&crm~2 ling and can~'-cd':c N c a = l a . ~ , +r.. . . . . . . . . . . • . . . . . L " . . . . . . . . . . . . IF" . . . . . . .

. ~ + ; . . . . . . . . . . + . + ~ + , ~ A + + / D T I \

• h ~ ~ + + . . . . . . + . . . . . . A . . . . . . I ^ . : + . r o l A . . . . . . . . + ~ + : . . . . . I . . . . C

+ k . . . . A . . - + : . . . . . . . + + . - P . . . . . . . . . ~ * ~ I . ~ - + ~ I . I + - - : ^ 1 o . . . . . . . ~ . . . . . . " I " . . . . . . . . . . . . . . . . . . . . . . . . . L " + ' ` ' u ~ ' + " + = . + . ~4 . / ~ /+o~ 1 / c j 11 ~ - l - { - ~ . ] _ . p_~ . c_ - l - ~ - J_

624

N F P A 9 2 B - - M A Y 2 0 0 0 R O P

a t . . . . . . . . . . . . . . . . . . . . . . L ' I Y . . . . . . . . . . O ~

~ + + . . . . . . . . . . . + . . . . . + + I . . . . . . . . " . . . . + . . . . . . . . . . = . . . . . . . . . ~ " I ~ ^ - - ~ + : . . . . . . , . : ' ~ 1 ^ ~ 1 : T ^ k l + Q m O O 1 k + . ^ , - I . . . . . : ~ 1 . 1 - - - -

. . (<2

t ~ j . . . . . . . . . . I : " + • + . . . . . - - - . . . . . . . . . . . . . . . . t h . . . . . . . . . e g k ~ + . . . . . ~ * I - . + ~ . + . . . . . . . . . . . . A +I++. . . . . ~ I . I = . + . . . . . * . . . . . + : ~ - -

~* ~ r r . . . . . . . . . . . 7 . . . . , = ~ , . . . . . . . . . . . " ~ " . . . . . . . . . • = - • o ~ 0

. . . . . . . , . . . . . . . . . . . . t " . . . . . . . C . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

+ . , , = ~ . , + + ~ . + . . o ~ + . , . . + . . . . . + + , . . . . . . + ~ ~ + . , + , ~ + . + + + + + x , , ~ + + + , ~

+J + ~1 ¢1 o I T - - . + ^ ~ J . . ~ ' . ' - - . T I . . . . . . . . . : - - ~ + . . . . + . . . . . + . . . . . +

m t c

; ~ £ ^ + " + : ~ 1 " ~ ' ^ " : + ' 1 ^ " 3 ; _ ~P+kl ^ ~ 9 C) O ¢1+ I - . ~ . ^ + . 3 . . . . . ~ I , I ^ -

".LL:2::y:LL^L:?~:':.:+7..::.3 : : : " ": : - : T Z : . . : : L +LU::." %~ . . . . . .

r ~ T ~ . . . . . t +'~ . . . . . . . . . . . . . . - I : _ + i ~ ^ + ~ , . , . . . . u . + t " ^ d ] f f c : . e ~ c e

t o 1 13, . . . . I . . . . . . . : ~ . , I : ~ . . + ^ + . I : + I ~ + ~ l , . l i ~ I : - - . : ~ + ^ ~ ^ I . + ' . . ~ . ^ _

b e u = c d . The z . = c s c ! a t e d t i m e f o r a c t u = f i c = , .~ c . z~ b e e = f i m a t e ~ ~}" ,,.:--~+,..+~ v ~ . ~ . . . . . . . . . . . . . : ^ ~ (~.) , " ' " '+ ' . ++'~.~ t e m p e ~ t u r e r~zc being t .hc a e t e . . - - . . ' n c d

a r n ~ ' c n t t e m p e : ' a t u r e .

o ~ + T ' I + . ^ + . . . . . ~ . . . . + 7 + 1 - . . . . . I . . . . . A ~ + I + . ^ ~ ^ : 1 : . . . . . I ~ ^

L:;+_:::~ ~:':~:':::C~: ~+'L:::::5 .r:'~+~ 7==a-;';'~,~', +-+++ _. thc =mc!.c !a.)'zr dec= na +. a a - a t i f / p r c m = = u r c ; 7 ( ;oo Sc=~=r+ # 4).

+t ,+ , ~ .__*._: m ~ _ ~? . . . . .-1: . . . . . . . : ' : - - e , + ' ^ : - ~ ' " - - + : - " ' ~ " +...1":.=.:] . . . . . . . : ~ + ^ 1 . . ~ ICI + I ~ ^ ~ . . . . . . . . . . - + , = . - ^C . 1 , . . . . . I . . . . . : + L " t ! ~ . C + t " . t + f . . . . . . . . . . . ! + " + ' . . . . . . . . V . . . . . . . . . . . . . . . . . . . . . . . . . . . . I n

d a t a + F c r X £ !"+0:

Delete existing equation ( 3 )

v = + 0 _ . . i , ] . . . . ~ / ~ ~

=d

tend: t e e : ' e r ~ S m a t e + ~ e t e m F c r a t ' a r e r+~e a t = d e - - n e e d *,5mez.

. . . . . . + . . . . . . . . . . . . . +:, . . . . . . . . . . . . . . . . . . . 2-~+_+^ . . . . . . . . . . . . .

• k . . . . . . + : ~ - - f l ira, I ' P ~ . + l f ' l t ~ t C'JI~'$ <:/1 I . . . k . l ' l ~ + ~ ~ + , . I ~ . - - • A I U Q p * . . . + : ^ .

A.Ig + 1 £~ ~ A / 7 ~ X : + A : . - - + . . . . . . : J - - - - E , . . h e + K - - . + . . . . . . + . 1 . + . + k - - ~ + = = , + ~ = ~ , - + 1 . u . . . . . . . . . . . . . . . . . l " • ' ~ . . . . . . . . . ~ . . . . . . . . . . . .

IY . . . . . . . . . 1 . . . . " 1 . . . . . . t + ~ 1 + + J . . . . . . . . . . . . . . . . . . . . . . . ~ . . . . . I : - - , , : A ~ - - + , . + 1 . . . . + - - : + k + I . . . . . . t - - : + : ~ l l . , k . . . . , ~ _ J l ~ . . . . . . . . . . . . . + - - , d l • " ~ + ' ~ r~ . ~`+. , " . . . . . . . ~, . . . . . . . . . . . . . . . ; . . . . . . . . . . . r . . . . . . t +" . . . . . .

. . . . . . . . . . . : . . . . c . . . ~ + k : . . . . : - - - - : ~ " * '+'^ c c . n ~ ! d e r c d "st,~ty -

9t 0 I:[ 1 T - - ^ + ~ . r i . . I P t . - - . ~ " ^ + . + . . . . . . . A P + - ^ . ' . ^ ^ ~ " . . . . . : ^ ~ 1 0 ~ . 1 + k ~

. . . , . ~ - . , = + o . . . . o . v = . . . . . . . . . " ~ " . . . . . . . . . . . . . . . . b J . . . . . . . . . . . . . . . . . .

e ; + S m a t c d ~ a f ' d = c t l a ~ c f "Z inc ~ e d ~ n +&care+-c~M g e n c . - a l i z a + . 2 c : : c f ~ h e l i m i t e d a m a u n t a f e x p c r l m e n + . a t d a t a :

T ' + + ~ : . . . . . + : ~ + t . . . . . , : + k ~ . . + + ~ + : . . . . . . . *h . . . . T ~ , . + ~ . + + + ~ I + + : I t + +

~ + + : ~ o + ~ + I . . . . . . . , . , + ~ . ~ I A ~ + ~ . . , , : + + I . . + I : - - : + ^ A + ^ : l : ~ + F11+'11 . . . h + . ~ + ~ ) . . . . . . . . . . . . 1 " . . . . . . . . . . . ~ . . . . . . . . . . . . . . . . . . . . . . = ~ t = + J , . . . . . , +

. . . . . . . . . ~ . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . c h a n g e 7-ar t. . . . . v . . . . . : ^ _ t a x "z mcz: :ccurztc "~ ? , /H2 £ 7.! , t £ A = ° O : : c = : d t ~ e : ' e z t = ! z f i c n r o t e d ~ . e : = a t e r = e c c d 1 . 0 a i r ch-~--':gc ~ c r h a u r .

3-4_* Stratification o f Smoke•

</ A I ~P I . . . . . . . . . . . - I . . . . . . . . . ~C . - - ~ , . ^ : ~ +L .~ ~ l . . . . : . . . I . . . . . . 1 ^ ~ ÷

+ I ~ - . - 1 1 : . . . . A e l ^ ^ _ , . . . . I. ^C+k . . . . . . . . . r 1 1 1 ^ - - •

. . . . . . . . . . . . ;•~+~'- . . . . . . . . . . . . g . . . . . . . . . . . ,, . . . . . -7~+ ' . . . . • . . . . , ~ t " . . . . . . y c ~ r l 7 a f t e r :g==lt=Cn, d e p e = d : z g c : ~ c : c = = ; ' c c t : ; ' c ~ e a t : c l c z ~ c r z t c e+-=d xt-.c m-==~ 'c= : t e m p c ~ v - r c ~'aN=+.2c= "= ~ e c p c = s p = c e . +'12".'= W,~,=i:'~,,~,7;+:,, -:3 ̂ ~+.~"~ ~ e ~ e " : ' : ^'a f rC~ . . +~'^ ~ : ^ ~ - C T ! ~ . . . . . . ,. ^c } f l ^ ~ ^ T J

_ _ _ ; • " P c . + - + . r , l , . . . . . . . . v . . . . . e , . . . . . . . . . . . . . .

Delete existing equation (5)

. ram_=. . . . . . . . . . . . . . . . ~, . . . . . . . . . . . . . . . . . . . . . . . . . . , '+f=~ ,, .~, . . . . . . . . + 1 . . . . . . , . : ~ ~ + P + 1 . . - I - . ^ ^ + ~ I . . . . + ~ I U + . . / . ^ . ~

A ' - I t ' / . a . _ ~ + ^ +C . I - ~ - _ . ~ ^C ~ I ~ . : _ - - + + . . . . . . . . . . . . : ÷ 1 + . . . . . . ~ + ~

I-.+1+1-,+ I o ] ~ l C + X

. . . . . . . . . . . ? " ' + L " . . . . . . . . . . . . . . . . . + . . . . . . . . )

Delete existing equation (7)

...t.:~+. ~ _ , . . . . . r . . . . . i . , - ~ ~ _ _ . . . r^. n .-:" . . . . rs-v Z:Z"..:.Y27+:.2: ~ L 7 2 ~ : ~ - - - - ; b L S " ~ . . . . . . . . . . . ~ . . . . . . + * . . . . .

Delete existing equation (8)

$-4.1 General. The uotential for swatification relates to the difference in temperature between the smoke and sprrounding aiF at any elevation [111. The m a x i m u m he ight to whk;h p lume fluid (smoke~ wiU rise. esneciallv early after imaitlon, deuends o n the convective heat release rate and t h e ambient temperature v'triatlen in the open space.

Of oarticular interest are those situations where the temperatur c of the air in the uDDer nort ion o f the large open snace is ~reater than at lower level's'before the fire. This-can occur as a re-suit of p, solar load where the ceil ing contains glazing materials. Comoutational_ methods ar , i+vallable to assess the potential for intermediate stratification.

One case of interest is denicted in Figure 3-4. In this case. the temoerature of the ambient air is relalively constan t UP to a he ight above which there is a laver o f warm air at un i form temperature. This situation can occur if the upper port ioq of a mall. a lr ium, o r other large space is unoccutf ied so that the air in that port ion is left u n co n d i t i o n ed .

3-4.2 Sten Function Temoerature . Where the interior air has a discrete temperature c h a ~ e at s o m e elevation above f loor level, then the ootential for stratification can be as+essed by applying th e o l u m e centerl ine temnerature correlation, Where the o l u m e

625

N F P A 92B ~ MAY 2 0 0 0 R O P

q e n t e r l i n e t e m p e r a t u r e is e o u a l to t h e a m b i e n t t e m p e r a t u r e , t h e n l u m e is n o l o n g e r b u o y a n t , loses its ab i l i ty t o r i s e , - a n d s t r a t i f i e s a t t h a t hei t th t .

3 - 4 . 3 I m n a c t o f S t r a t i f i c a t i o n o f S m o k e o n S m o k e M a n a g e m e n t S y s t e m I ) e s k r n . O n c e a s m o k e e v a c u a t i o n svs t em h a s s t a r t e d in a n a t r i u m 91" 9 t ~ e r l a r g e s n a c e , t h e s t r a t i f i c a t i o n c o n d i t i o n will b e e l i m i n a t e d by t h e r e m o v a l o f t h e h o t laver . T h e p r o b l e m f a c i n ~ t h e d e s i g n e r is h o w to e n s u r e t h a t t h e n r e s e n c e o f s m o k e is n r o m p t l v d e t f e t e d t h r o u g h all o o t e n t i a l p r e - f i r e t e m n e r a t u r e prof i les . U n d e r s g m e c o n d i t i o n s , suc[a as n i g h t s a n d c o l d days . it i s~o robab le t h a t a ~ t r~ t i f ica t lon c o n d i t i o n will n o t b e p r e s e n t a n d a n y s m o k e n l u m e will p r o m n d v r ise to t h e r o o f o r ce i l i n~ o f t h e v o l u m e , in w h i c h case d e t e c t i o n a t o r n e a r t h e t o p o f t h e v o l u m e w o u l d b e r e s p e n s i v e . In o t h e r cases , s u c h as h o t s u m m e r days o r days wi th a h i g h ~olar l o a d . t h e n l u m e m a y n o t r e a c h t h e t o p o f t h e v o l u m e a n d t h e s m o k e c a n s o r e a d a t a level l o w e r t h a n i n t e n d e d , in w h i c h case d e t e c t i o n n e a r t h e tot) o f t h e v o l u m e w o u l d n o t r e s o o n d a n d t h e s r a o k ¢ m a n a g e m e n t svs tem w o u l d n o t b e s t a r t e d . Tl~ere is n o s u r e way o f id¢Bt~fying w h a t c o n d i t i o n will b e p r e s e n t a t t h e s t a r t o f a f i re . T h e f o l l o w i n ~ a r e two d e t e c t i o n s c h e m e s t h a t c a n o r o v i d e f o r p r o m p t d e t e c t i o n r e ~ a r d l e s s o f t h e c o n d i t i o n p r e s e n t a t t he t i m e of f i re i n i t i a t i on .

Q = h e a t r e l e a s e r a t e o f f i r e ( B t u / s e c ) Q, = c o n v e c t i v e p o r t i o n o f h e a t r e l e a s e r a t e ( B t u / s e c )

Q , = J" (1 - ~ ) Q d t f o r s t e a d y fires: Q n = (1 - ~ t ) Q t (Btu) f o r t ~ f i re

Q n = ( 1 - g , ) ( / , t~ /S (B tu )

Q,= po c t T o A ( H - z ) (B tu ) t = t i m e f r o m i g n i t i o n (see) Tx= a b s o l u t e a m b i e n t t e m n e r a t u r e (R)

A T = t e m p e r a t u r e r ise in s m o k e l aye r (°F) V= v o l u m e t r i c v e n t i n g r a t e ( f t S / m i n ) Yi= m a s s f r a c t i o n o f spec i e s i (Ib s p e c i e s i/lb o f s m o k e ) z = h e i g h t f r o m t o p o f fue l to s m o k e l a y e r i n t e r f a c e (ft)

0 [ = t ~ f i r e g r o w t h c o e f f i c i e n t ( B t u / s e c s)

p o = d e n s i t y o f a m b i e n t a i r ( l b / f t s)

~ 0 ~ = c o m b u s t i o n e f f i c i ency f a c t o r (-), m a x i m u m va lue o f 1 [21]

~ = to ta l h e a t loss f a c t o r f r o m s m o k e l a y e r to a t r i u m b o u n d a r i e s ,

m a x i m u m v a l u e o f 1, m a x i m u m t e m p e r a t u r e r ise will o c c u r i f ~ = 0

(a~ BeamDetection of the Smoke La~er at Various Levels. T h e ~ m . ~ i . . . . . : _ . ~ ^ _ ~ ' f - ~ _ , t. . . . . . . . . . : . . . . . , .^ c . . . . L. . . . . , .^ p u r p o s e o f th i s a p p r o a c h is to uuici~lv d e t e c t t h e d e v e l o p m e n t o f a , . . . . . : . . . . . . . ~ ' ~ ' ~ . . . . :^ o~: . . . . . I- . . . . I. . . . . . . . . . . . . . . . . : ^~

- - a ¢ ~ ^ a : ~ : ~ _ ^ ~c .^_ : . . :_: . :~, . . . . . . : ^_ . , . ^ . ^ : . . . . . , .^ s m o k e l aver a t w h a t e v e r t e m p e r a t u r e c o n d i t i o n exists. O n e o r a n ~ t o r a ~ ~ : . , , . . . . . . . . . . . . . . . v . . . . . . . , . . . . . . . . . . . . . . .

fill: . . . . . . .;::':*.^ .~..;~"-':,~.;$~...t.:~L. , t . . . . . b^ ] . . . . . : - ,^. .¢^~^ I . . . . I . . . . .~ :~ m o r e b e a m d e t e c t o r s a r e l o c a t e d a t t h e r o o f level. A d d i t i o n a l . . . . ~, ~x.-,x.~.~'~ . . . . r~-:~ . . . . . . . . . . . . . . . . . . . ~ . . . . . . . . . . . . . . . . . . . . . . . . d e t e c t o r s a r e l o c a t e d a t o t h e r levels l o w e r in t h e v o l u m e T h e e x a c t . . . . . ~ - - t ~ : , ~ . . . . . . . . ~:~.-..~ . . . . . . . . . . . . . . . . . . . . . . . . . :~- .

. . . . . . . -:: .. . ~:,:<~.,

p 0 s m o m n g o f t h e b e a m s ts a f u n c t m n o f t h e snec t f i c d e m g n b u t _ , ~ : : ' . . . : . : . . . . . ~v" . . o - • • e - - . . . . ; . . . . . . . - . . . . - ~ k r m t l ~ . ~ o t r~rs t l n ~ l c a t m n o t a m o g e : ~ 7 c . . = t o t . a c e at A n y S~QUI(1 i n c l u o e p e a m s a t t h e o o t t o m o t a n l a e n n n e o -/~.':2&. "~:.'::::::, "

~ ] ~ 1 ~ : - ' . . . . ' : : :~ : : : :~ u n c o n d i t i o n e d s p a c e s a n d a t o r n e a r t h e d e s i g n s m o k e level wi th ~-.-~:: ' "-'::"

it) t e r m e d ] a t e severa l b e a m p o s i t i o n s a t o t h e r levels. .-. $-6.1 " ~ . . . a l . ~ . , .The p o s i t i o n o f t h e f i r s t i n d i c a t i o n o f s m o k e ":,.'.:'~?:~..~.:... i n t e f f -aee -~ i~ i~ : " t ime c a n b e d e t e r m i n e d f r o m t h e r e l a t i o n s in 3-6.2

(b) //corn Detection of the Smoke Plume. T h e p u r p o s e o f th i s " ~ . : . . S . e c t l o ~ i i ' ~ 7 . T h e r e l a t i o n s a d d r e s s t h e f o l l o w i n g t h r e e a p n r o a c h is to d e t e c t t h e r i s i n g ~ l u m e r a t h e r t h a n t h e s m o k e laver . ~:.:"~. " ~ i : . " F o r th i s a p p r o a c h , a n a r r a n g e m e n t o f b e a m s c lose e n o u g h to e a c h :'% ..::#' .................... . . . . . i~ ~ . ' : ' ~ o t h e r to e n s u r e i n t e r s e c t i o n o f t h e n l u m e a r e i n s t a l l ed at~'~': . . : '!~,: "~ilIi!:" (1) gNo s m o k e e x h a u s t is o p e r a t i n g (see 3-6.2)

.(~) T h e m a s s r a t e o f s m o k e e x h a u s t e q u a l s t h e m a s s r a t e o f b e l o w t h e lowes t e x p e c t e d s t r a t i f i c a t i o n level. T h e s p a C ~ b e i ~ ! ":-:.~, . . . . . . . . - . . . . . . . . . . ~ .,:z::::, - ....-¢ . : ~ o k e s u p p l i e d f r o m t h e p l u m e to t h e s m o k e l a y e r (see 3-7 I) oasecl o n m e WlCim oI m e p e a m a t t i le l eas t e l eva r lo r r : i a t~ /e a D ~ E ~ , ~?....!::. , ~ , ~ _ , , , . , . , . , ~

• :"" ":::.':~:::, Z-.':-"'-":':::::i::%::,. ":ik ¢: L31 l n e m a s s r a t e o i s m o k e e x n a u s t is less t i t a n m e r a t e o r s m o k e o f f i r e p o t e n t t a l - " ~ i ~ ":"::ii~ii~ i: s u p p l i e d f r o m t h e p l u m e to t h e s m o k e l a y e r (see 3-7.2)

, • u . . . . .~ . . . . r , ^ ~ : : . ":::::.'::", 3 5 S m o k e L a v e r P r o n e r t ~ e s • • . x " . - " . . . . . . . . . ~ : . . ~ ' " ' ~ ~ - : ; ~ ' $-6.2 ] J ~ h . [ ] g e s i t o - ~ o f F i r s t I n d i c a t i o n o f S m o k e : ~ ) ' c : . - - - tc . . : .cc y . . . . . . . ~ . . . . . . . 7 ~ ' : . . . . . 7 . . . . -~. ~ .'7.-.'~'" ~. . . . . . ~ '~ ; .~ . . : . ' . - : : : ' w~th N o S m o k e E x h a u s t O p e r a t m g .

(e .g . , C O , H C ! , H C N ) " : : :..~.c!:e ]:)'er::':" E " ~ : o n s to ~ l a t e t h e s m o k e l a y e r d e p t h , a v e r a g e t e m p e r a t u r e r ise, ~ . . a l d e n ~ , a n d spec i e s c o n c e n t r a t i o n s d u r i n g t h e s m o k e f i l l i n g ~ . . ¢ . , a..~."" t h e quas i - s t e a d y v e n t e d s t a g e a r e p r o v i d e d in T a b l e 3-5. T h ~ i ~ h a t i o n s a p p l y f o r f i res wi th c o n s t a n t h e a t r e l e a s e r a t e s a n d ~ u a r e d f i res . T h e s e e q u a t i o n s c a n a l so b e u s e d t o c a l c u l a t e t h e i ! o n d i t i o n s w i t h i n t h e s m o k e l a y e r o n c e t h e v e n t e d c o n d i t i o n s exist .

3-6.2.1 S t e a d y F i res . F o r s t e a d y f i res , t h e h e i g h t o f t h e in i t ia l i n d i c a t i o n s o f s m o k e a b o v e t h e f i r e s u r f a c e , z, c a n h e e s t i m a t e d f o r a n y t ime , t, f r o m E q u a t i o n ( 0 2 3 , w h e r e c a l c u l a t i o n s y i e l d i n g z/H > 1.0 m e a n t h a t t h e s m o k e l a y e r h a s n o t ye t b e g u n to d e s c e n d .

z/H = 0 .07 - 0 .28 In [ ( t(~/S/l-r/3)/(A/It 2) ] ( 0 2 3

T a b l e $ - 5 E q u a t i o n s f o r C a l c u l a t i n g P r o p e r t i e s o f S m o k e L a y e r

e - - ^ ~ - r e . n : _ _ S ~ c U n v e n t e d F i r e s

P a r a m e t e r s S t e a d y F i r e s T - s q u a r e d F i r e s V e n t e d Fi re s

A T . T ~ [ e x p ( Q . / Q o ) ] - 11 _T~[exp ( Q o / Q ) I - l i [00(1 .Z , ) Q ~ / ( p o c p F )

D (D,Qt)/[ZoAHe4(H.z) ] (D=a t~)/ [$c, DH, A(H-z)] (60D,,Q)/(Z~AH, V )

E ~Qt ) / [ OoXoAH, A(H-z)] ~oe d)/13Oox.AH, A(H-z)] ( 6 0 f Q ) / ( p o Z ~ A H y )

w h e r e : A = h o r i z o n t a l c ros s - sec t iona l a r e a o f s p a c e ( f t ~) c, = soec i f i c h e a t o f a m b i e n t a i r ( B t u / l b * ° F ) /5 = L - ~ l o g ( l J / ) , o p t i c a l d e n s i t y

2 D~ DV/m/, = m a s s op t i c a l d e n s i t y (f t / I b ) m e a s u r e d in a tes t s t r e a m c o n t a i n i n g al l t h e s m o k e f r o m a m a t e r i a l t e s t s a m p l e

3 V = v o l u m e t r i c f l ow r a t e ( f t / s e c ) L f = y ie ld f a c t o r o f spec i e s i r ib spec i e s i/lb fue l ) H = ce i l i ng h e i g h t (ft) AH~ = h e a t o f c o m p l e t e c o m b u s t i o n ( B t u / l b )

w h e r e : z = h e i g h t o f t h e f i r s t i n d i c a t i o n o f s m o k e a b o v e t h e f i re s u r f a c e

(ft) H = c e i l i n g h e i g h t a b o v e t h e f i re s u r f a c e (ft) t = t i m e (see) Q = h e a t r e l e a s e r a t e f r o m s t e a d y f i re ( B t u / s e c ) In = n a t u r a l l o g A = c ros s - sec t iona l a r e a o f t h e s p a c e b e i n g f i l l ed wi th s m o k e (f t 2)

6 2 6

N F P A 9 2 B - - M A Y 2 0 0 0 R O P

Equadon (0_33 is based on exper imenta l da ta f rom investigations us ing un i fo rm cross-sectional areas with respect to he igh t with A / H z ratios in the range f rom 0.9 to 14 a n d for values of z/H>_ 0.2 [7, 10, 12, 13, 14]. This equa t ion is for the worst case condit ion, a fire away f rom any walls. T h e equat ion provides a conservative est imate of hazard because z relates to the he igh t where there is a first indicat ion of smoke, ra ther than the smoke layer interface posi t ion.

3-6.2.2* Uns teady Fires. "I~e descen t of the he igh t of the initial indications of smoke can also be es t imated for certain types o f uns teady fires, for example, t -squared fires. From basic theory and l imited exper imenta l evidence, the he igh t of the initial indicat ions of the smoke above the fire surface, z, can be est imated for a given t ime accord ing to the following relation, where calculations yielding z /H > 1.0 m e a n tha t the smoke layer has no t yet begun to descend .

z /H = 0.23 [ t/( tg ~/s Pi 4/~ (A/H2) s/s) ]-i.4s 0"0_43

where: z = he igh t of the first indicat ion of smoke above the fire surface

(ft) H = ceiling he igh t above the fire surface (ft) t = t ime (sec)

where: m = total fuel mass c o n s u m e d (Ib) At = dura t ion of fire (sec) H, = hea t o f combus t ion of fuel (Btu / lb) tg = growth t ime (sec)

3-6.2.4* Varying Cross-Sectional Geometries and Comp lex Geometries. Equations (07) and (~08.) are based on exper imen t s conduc ted in un i fo rm cross-sectional areas. In practice, it is recognized tha t spaces being evaluated will n o t always exhibit a s imple un i fo rm geometry. T h e descen t o f the first indicat ion o f ~ smoke layer- in varying cross sections or complex geomet r ic spaces can be affected by condi t ions such as s loped ceilings, variations in cross-sectional areas of the space, a n d project ions into the r ising p lume. Where such irregularit ies occur, o the r m e t h o d s of analysis shou ld be considered. O the r m e t h o d s of analysis, which vary in their complexity bu t may be useful in deal ing with complex an d n o n u n i f o r m geometr ies , are as follows:

(a) Scale models (see 3-1.1 and 3-1.2). (b) CFD -~-¢¢4d mode l s (see 3-1.1). (c) Zone mode l adaptat ion. A zone mode l (see 3-1.1.3,1)

predicated on smoke filling a un i fo rm cross-sectional geomet ry is modif ied to recognize the chang ing cross-sectional areas o f a space (see 3-1.1). The e n t r a i n m e n t source can be modi f ied to accoun t

tg= growth t ime (sec) for expec ted increases or decreases in en t r a inmen t due to geome'tr ic c o n s i ~ s , such as projections.

Equat ion 0-04) is based on exper imenta l da ta f rom investigations (d) Sens i t i~ t . 'na lys '~" An i r regular space is evaluated us ing with A/H2 ratios in the range from l.O to 23 and for vaiues of z/ H>_ Equat ions ( ~ f f ~ , ( i O _ 8 . ) at and between the limits of a 0.2 [10]. Equat ion (J, O4) is based on un i fo rm cross-sectional areas m a x i m u m ~ g h t ' ~ : m i n i m u m he igh t identifiable f rom the with respect to height . This equat ion is for the worst case g e o m e ~ i ~ : ~ . e s p a i ~ ' ~ n g . e q u i v a l e n t he igh t or volume condit ion, a fire away f rom any walls. T he equa t ion also provides a c o n s ~ ' ~ ' $ ' ~ Z " : ~ ' - " conservative est imate of hazard because z relates to the height .-.-::':'.-::';" :.:, ~'i'i~:" where there is a first indicat ion of smoke, ra ther than the smoke ...~.:~.~ P ~ o n of Smt~'e Layer Interface with Smoke Exhaust layer interface position. "()p~:':::::iii '~:.

"'::'~:::~::.:..

3-6.2.3 Mass Consumpt ion . T h e equat ions presen ted in 3-6.2.1 ~. 3-7 6 , g . ~ . ~ t e o f Smoke Exhaus t Equals Mass Rate o f Smoke ~nd e3~te2r2i a r e t i u ? ! ~ l o i ? : s ~ a e l U c ~ v f i ~ g e , t h t e h P t S l t ~ l ° ~ ~ Y " ~ ; . ~ : . u . . ~ l l i e d ~ r the smoke exhaus.ts.ystem h .as.operated for a

s , , • z v :~: . : ~ e n t .1~,¢1oa ot ume , an e q u m n n u m p o s m o n ot m e smoke re.c[uired to sustain the steady hea t release rate over the t ime per iod "?"ii!-, ~ a c e will be achieved if the mass rate of smoke exhaus t is of interest can be de te r ra lned as follows ~ ~ u a l e o s : ........ !~g::..'-'~ou l ~ " t h e mass rat f moke suDplied by the p l u m e to the base

.,:.i~#~-~ "?-{$i or* t hd ' smoke layer. Once achieved, this posit ion shou ld be m= QAt /H~ :~..:.y, (~lg.~ ~,'..-,W~tintained as long as the mass rates r ema in equal. See Section 3- ..$.:::::::. .:.#. . ,::;

.~.:-:.:.:-:-:-:.. :--:.. .... ~.,. i~.~ for the mass rate of smoke su hed to the base of the smoke . :i::"".'.'-:i:i..'-~--., .:.::~:~)~'.':-.~. '~.~ . P P . where • , • "!.'.'~.~;::" " ' : ~ layer for different p lume conf igurauons . m = total fuel mass consumect (lb) "~i:~--:,$:.,, "-':'-'.".-':'.-'.".~ Q = hea t release rate (Btu /sec) ..*:-'?.:.'~ii~i~:--"~:::.-.., ":~i-:.-"~ . . . . . . 3-7 6~ .2 Mass Rate o f Smoke Exhaust Not Equal to Mass Rate of At = dura t ion of fire (sec) .~':~" " ":-"~::-. ":~:."-!:."-~::.:-'.:'¢" - • • ..:, .~.:.:.:.:.:.:.. -.~x.:.w Smoke Suoolied. With a vreater rate of mass sunt~lv t h a n exhaust , H, = hea t of combus t ion of fuel ( B t u / l ~ ":;~i?:. .--:f;': an equi l ibr ium posit ion o~ the smoke layer interface will n o t be

¢~'~"::~;~':.~, . ':~.ii~":. achieved. The smoke layer interface can be expected to descend, vor a t-s unfed fire, the total mass consum"" ~.ver tlae t'ti~e • - • ,q " . ~ : . . , ~.~ t hough at a slower rate than ff no exhaus t was provided (see 3-6.2), eriod ot interest can be deterrmnect as- " " ' ~ ~ ~ - - P • %!:...'..~..-., :.# Table g4Gg-,$.~:.7~ include~ in format ion on the smoke layer

. . . . ~ . . . . . ~, ":":'~i.:'."-'~'i'": .... , . . . . posi t ion as a func t ion of t ime for axisymmetr ic p lumes o f steady m = ~ a r / t tx,~g ) ~i~i;" K~-~-q~ fires given the inequali ty of the mass rates. For o ther p l u m e

4:: configurat ions, a c o m p u t e r analysis is required.

Table 3-7.2 Increase in Time for Smoke Layer Interface to Reach (Axisymmetrlc Plumes and Steady Fires)

t / t o

m / m r = 0.25 0.35 0.5 0.7

Selected Position

0.85 0.93

z / H

0.2 1.12 1.19 1.3 1.55 1.89 2.49

0.3 1.14 1.21 1.35 1.63 2.05 2.78

0.4 1.16 1.24 1.4 1.72 2.24 3.15

0.5 1.17 1.28 1.45 1.84 2.48 3.57

0.6 1.20 1.32 1.52 2.00 2.78 4.11

0.7 1.23 1.36 1.61 2.20 3.17 4.98

0.8 1.26 1.41 1.71 2.46 3.71 6.25

627

N F P A 92B - - MAY 2 0 0 0 R O P

where: z = design height of smoke layer interface above fire source H = ceiling height above fire source t = time for smoke layer interface to descend to z t o = value of t in absence of smoke exhaust [see Equation (9)] m = mass flow rate of smoke exhaust (minus any mass flow rate

into smoke layer f rom sources o ther than the plume) rn, = value of m required to maintain smoke layer interface

indefinitely at z [see Equation (14)]

$ . ~ t Rate of Smoke Mass Production. The height of the smoke layer interface can be mainta ined at a constant level by exhausting the same mass flow rate f rom the layer as is supplied by the plume. The rate of mass supplied by the p lume will depend on the configuration of the smoke plume. Three smoke plume configurations are addressed in this guide. The exhaust fan inlets should be sized and distributed in the space to be exhausted to minimize the l ikelihood of air beneath the smoke layer from being drawn through the layer, sometimes referred to as phtggi-ng plugholin~. To accomplish this, the velocity of the exhaust inlet should no t exceed a value to cause fresh air to be drawn into the smoke layer.

3 ~ g . l Axisymmetric Plumes• An axisymmetric plume (see Figure -g-g--/- ~.-.~-d) is expected for a fire originating on the atrium floor, removed from any walls. In this case, air is ent ra ined f rom all sides

The p lume mass flow rate, m, above the limiting elevation is predic ted from:

m= 0.022 QcllSzJlS + 0.0042 Q, (z > zt) ( t -4~ where:

m = mass flow rate in p lume at height z ( lb/sec) z = height above the fuel (ft)

The plume mass flow rate below the flame tip is predicted from:

m=0.0208 Q?l~z (z<_ zt) (t-59.)

3-~-.1.3 The rate of mass supplied by the plume to the smoke layer is obtained f rom Equation (-t69.) for clear heights less than the flame height [see Equation (-l-g3] and otherwise f rom Equation (t-48). The clear height is selected as the design height o f the smoke layer interface above the fire source.

8-~g.1.4 It should be no ted that Equations (-1-4.8.) and (t-g~) do no t explicitly address the types of materials involved in the fire, o ther than through the rate of heat release. This is due to the mass rate of air ent ra ined being much greater than the mass rate of combustion products genera ted and due to the amount of air entrained only being a function of the strength, that is, rate of heat release, of the fire...:..

and along the entire height of the p lume until the plume becomes submerged in the smoke layer. - . . . . . . . . v . . . . . . . . . . . . v . . . . . . ~, . . . . . . . . . . v . . . . . . . . .

. . . . . .... ~'¢~2 ~•.L~wE~^-. : . . ^c . _ ^ , . ^ .

D el ~ x ! ~ s h~ g e q u a t i o ~ l ~ ' ~

\ ) I ~ '~ ' . .~ l~,g . l .5 6 ~ can be located near the edge or a corner of the H ~.~'~" .~ \ / I Z. '~:~: : ~ . . . C . ~ In this case, en t ra inment might no t be f rom all sides ":'~x " ::;::::~¥'k • • [ I ,~ ,~i~i.fi '~,~me, resulting m a lesser smoke productaon rate than

I I ~ - . . ~ : " ~ W h ~ ' n t r a i n m e n t can occur f rom all sides. Thus, conservative I I .::#'-*:" ' ~."::~, ~ d e s i g n calculations should be conduc ted assuming that

] I ~ ~i !'" ~ . ~ a i n m e n t occurs f rom all sides.

;t.2 B a l c o n y Spill

X~.\\'x~3 * ~ - ~ . ~,~}~.~..~: 3-~fl-.2•1" A balcony spill p lume is one that flows under and around .~, ~ " ~ : ~ a balcony before rising, giving the impression of spilling f rom the ~.~ .~ .~"

~ ? ~ ~ .~ balcony, f rom an inverted perspective (see Figure 3--78.2,1). Figure 3-8 1 Ax~/mmeei ' i e • ~ ' i ~ ' ~ e . ~ Scenarios with balcony spill plumes involve smoke rising above a

• "~*Zh. ~ . fire, reachimz a ceilinlz, balcony, or o ther significant horizontal 3-~ g 11 The mass rate of smoke p roducuon es.L.~nated • . . . . :non c .a~.~. ~ , projection, ~ e n traveilng horizontally toward the edge of the based on. the rate of entraaned mr, smce the mass l ~ f "balcony." Characteristics of the result ing balcony spill p lume combusuon products genera ted f rom. the fire is g e ~ l y much less d e p e n d on characteristics o f the fire, width of the spill plume, and man m e rate ot mr en t ramea m me pmme. he ight of the ceiling above the fire. In addit ion, the path of

3-~g.1.2" Several en t ra inment relations for axisymmetric fire plumes have been proposed. Those r e c o m m e n d e d here in were those first derived in conjunct ion with the 1982 edition of NFPA 204M, Guide for Smoke and Heat Venting. These relations were later slightly improved by the incorporat ion of a virtual origin and also compared against o ther en t ra inment relations [2,15].

The following en t ra inment relations are essentially those Sented in NFPA 204~, Guide for Smoke and Heat Venting [2]. cts of virtual origin are ignored since they would generally be

small in the p resen t application and thus far can only be adequately predicted for pool fires. The definit ion of a l imiting elevation, cor responding approximately to the luminous flame height, is given as:

~, = o . s s s 0~ 2/~ ( ~ )

where: z, = limiting elevation (ft) Q~ = convective port ion of heat release rate (Btu/sec)

horizontal travel f rom the plume centerl ine to the balcony edge is significant.

For situations involving a fire in a communicat ing space immediately adjacent to the atrium, air en t r a inmen t into balcony spill plumes can be calculated from Equation (t=7~10):

m = 0.12 ( QH a ) l/s (g./~ + 0.25/-/) (t-g.L0.)

where: m = mass flow rate in p lume ( lb/sec) 1~¢= heat release rate of the fire (Btu/sec)

= width of the p lume as it spills unde r the balcony (ft) Z b = height above the balcony fit) H ffi height of balcony above fuel (ft)

Equation (t-gig.) is based on Law's interpretat ion [16] of smaib scale exper iments by Morgan and Marshall [17]. Equation (t-gLQ) should be regarded as an approximat ion to a complicated problem.

628

l N F P A 92B - - M A Y 2 0 0 0 R O P

~ "~"~ ~u':'~4<:~

Section

~.-~!-"!~i ~i!iii!~:~i~i~.~:~i~i~! ~!~!'-'."-!~! : iiiiiiii~i~iii:-'!liiii::~:~.~i:-:.~i~:~k%i ~.--..-';.:~.;i ?

~ ~ w _ _ __~ o,,r'

+ + Front view with draft curtains

Delete existing Equation (11)

where: W = the width of the plume w = the width of the opening from the area of origin b = the distance from the opening to the balcony edge

3-~ .3 Window Plumes.

3-~g.3.1 Plumes issuing from wall openings, such as doors and windows, into a large-volume, open space are referred to as window plumes (see Figure 3-8.3.1) . After room flashover, the total heat release rate can be expected to be governed by the airflow rate through the wall opening from the open space, i.e., the fire is "ventilation controlled." The heat release rate can be related to the characteristics of the ventilation opening. Based on experimental data for wood and polyurethane, the average heat release rate is given as [20,21]:

Q= 61.2 A m H~ 1/~ ( t -812)

where: Q = heat release rate (Btu/sec) A~ = area of ventilation opening (fC) H~ = height of ventilation opemng (ft)

.....~:-."-,h.. This assumes th#~".',~.eat release is limited corn partment~.~.fue! ~enecation excess fuel ~ c" tsi outside ~ ~ t. The methods only va~..'~..~S~omp~i ents.having

by the air supply to the is limited by the air supl~ly, and

de the compartment using air entrained ~'r in this section are also

a single ventilation opening.

Zw

Front view without draft curtains

F'~ure ~.,~_._._._._._._~1 Balcony spill plume.

3-~.~.2.2 When z~ is approximately 13 times the width, the balcony spill plume is expected to have the same production rate as an axisymmetric plume. Consequently, for zb>13W, the smoke production rate from a balcony spill plume should be estimated using Equation (t-48).

3-8_g.2.3 The width of the plume, W, can be estimated by considering the presence of any physical barriers protruding below the balcony to restrict horizontal smoke migration under the balcony. In the absence of any barriers, visual observations of the width of the balcony spill plume at the balcony edge were made in a set of small-scale experiments by Morgan and Marshall [17] and analyzed by Law [16]. In these experiments, the fire was in a communicating space, immediately adjacent to the atrium. An equivalent width can be defined by equating the entrainment from an unconfined balcony spill plume to that from a confined balcony spill plume. The equivalent width is evaluated using the following expression:

Side Front

Figure 3-8.3.1 Window plume.

3-~ ~t.3.2 The air entrained into the window plume can be determined by analogy with the axisymmetric plume. This is accomplished by determining the entrainment rate at the tip of the flames issuing from the window and determining the height in an axisymmetric plume that would yield the same amount of entrainment. As a result of this analogy, a correction factor addressing the difference between the actual flame height and the equivalent axisymmetric plume height can be applied to the axisymmetric plume equation according to the following relation:

a = 2.40 a ~/5 n=l / s _ 2.1 H~ (~-o~_1.~

where: Q = heat release rate (Btu/sec) A m = area of ventilation opening (ft ~) H~ = height of ventilation opening (ft)

Then, the mass entrainment for window plumes is given as:

m = 0.022 Q)/s (z~ + a) s/3 + 0.0042 Q~ (~014)

where: z~ = height above the top of the window.

629

N F P A 9 2 B - - M A Y 2 0 0 0 R O P

Subs t i t u t i ng fo r 0,~ f r o m E q u a t i o n (-1-g_J2,),

m = 0.077 ( A H~ 1/~) ~/s (z , + a) s/s + 0.18 A~ H~ 1/~ (-24-15)

T h e v i r tua l sou rce h e i g h t is d e t e r m i n e d as t he h e i g h t o f a f i re sou rce in t he o p e n t h a t gives t he s a m e e n t r a i n m e n t s as the w i n d o w p l u m e a t t he w i n d o w p l u m e f l a m e t ip. F u r t h e r e n t r a i n m e n t above t he f l a m e t ip is a s s u m e d to be the s a m e as for a f i re in t he open . W h i l e th i s d e v e l o p m e n t is a r e a s o n a b l y f o r m u l a t e d m o d e l fo r w i n d o w p l u m e e n t r a i n m e n t , t h e r e a re no d a t a ava i l ab le to va l ida t e its use. As such, t h e a c c u r a c y of t he m o d e l is u n k n o w n .

3-8.4 P l u m e Wid th . l = ~ ' - c = c c c,f P!'--'mc Cc.---'~-.:t v - t ~ W~.~2~. As a p l u m e rises, i t e n t r a i n s a i r a n d ~ t s~ widens . -m.^ ~t . . . . . :_t . ,

ovcrc21 F l u m e 2";mn:ctcr ca-: bc c:~dmatc~ : : r, ~ . t . ~ j .

D e l e t e e x i s t i n g E q u a t i o n (22)

- - t ~ . . . . . . . . . . . . . . k . . . . . . . . . . . . . . . . . . v . . . . . . . , ' x - - , ,

"_ ,U;L~.7~Z, " - ' ' r . . . . . . . , - ,

c ^ - n : . . . . . . . . A k . , + I - . _ _ _ + _ _ : . . . . . ^ z . _ _ ^ t ^ : . ^ I ^ - - . i - . ^ ^ _ + : _ ^

le:g'& c.f :~c F!umc. TSu~, g_G_enerally the total plume diameter can be e s t i m a t e d as:

a= Z ~ • (~ l~}

i n t e r e s t w h e n a t r i a a r e t e s t e d bv rea l f i res as d i s cus sed la ter . The ggn t e r l i ne t e m n e r a t u r e can be a n o r o x i m a t e d f r o m

l- - I I /3

Z~p=Za+9.11 L g f p~a J - ~

(18)

w h e r e :

x).a.= a b s o l u t e c e n t e r l i n e p l u m e t e m p e r a t u r e a t e l eva t ion z (°R.

Tz= a b s o l u t e a m b i e n t t e m o e r a t u r e (°R. °K)

p~= dens i ty of a m b i e n t a i r ( I b / f t s. k g / m s)

g = a c c e l e r a t i o n of gravi tv (~2.2 f t / s ~. 9.8 m / s ~) z = h e i g h t above toD of fue l (ft. m) C.= soec i f ic h e a t oi= a i r (0.241 B t u / I b * ° F . 1.005 k l / k g * ° C I

~-~* N u m b e r o f E x h a u s t In le t s . W h e n t he s m o k e laver d e o t h b e l o w a n e x h a u s t i n l e t is re la t ive ly shal low, a h i g h e~thaust r a t e can [ ~ I ~9 e n t r a i n m e n t of c o l d a i r f r o m the c lea r laver . T h i s p h e n o m e n o n is c a l l ed o l u g h o l i n g . T h e n u m b e r of e x h a u s t in le ts n e e d s to be choser~:!~o {he m a x i m u m f low ra t e s fo r e x h a u s t w i t h o u t o l u g h o l i n g a re r ~ d e d . Acco rd ing ly . m o r e t h a n o n e e x h a u s t i n l e t m a v l a e ~ e d . T~e m a x i m u m m a ~ f low ra te . w h i c h can be e f f i c i endv ~ ' a ~ ' ~ s i n g a s ing le e x h a u s t in le t , is raven as [CIBSE

.,,.~.~.-.:.:-::-.::.~. ~ ,::.:::::~::~ .~.:~"

~ _ ~ . 3 5 4 f l d ~ T, _ T O ~1,~ . . . ~ . . . @ : . ~ , ~::;~::. T, T,

:. "%~i::i~:, ~ l u m e d i a m e t e r f i t) '~":: '~ whe re : %i~j~.-.:..-.'.-.@" . . . , ~ ":-':*-':~"

Z = b e m h t f i t) "~i...'-..'::.'.. . . . . ..::.:::" / q = d i a m e t e r c o n s t a n t "~ , ~ ' ~ x i m u m mass r a t e of e x h a u s t w i t h o u t p l u g h o l i n g f i b / s )

. . . . . . . . . . . . . . . ~:~'.~;.~ ..,.x..>~:~:#.:~T_~~= ~ g s o l u t e t e m p e r a t u r e o f t he s m o k e laver (°R~ i n e c u a m e t e r c o n s t a n t can r a n g e t r o m u .za to o.a. i t IS ..4:~:-$~.~::~ ~::~ ~" " - " aml~ ien t t e m p e r a " ~ " • ~ . ~ " " ~ i ~ "?::. d~ = a n s o m t e t u r e I-K/ r e c o m m e n d e d t h a t va lues o f K , be c h o s e n so t h a t t he r~lt in~'~: . , ' : '-.'-'.x .~ . . . . . . . - . . . . . . . . .

• - 6:al.':. ~ ~ " '~:~.".~.'.~= G e D m O I s m o k e l a v e r b e l O W e x h a u s t i n l e t l i t } ca l cu l a t i ons a r e conse rvauve : ~ . .,~.~,-:o,~ :~x :,'-~-~" -

*'+" " : ~ ' ~ ~ . ~ x ~ ~.">" ~ (Beta) = e x h a u s t l o c a t i o n f ac to r ( d i m e n s i o n l e s s ) • . "::!i~.~*:*" "~.-'..<~

= 0.5 resul t s In a conservatave e s t i m a t e of p l u m e c o n t ~ t h .... -....,~::.~%,..:.. "::%':.~::~,~

walls . ..:..;.-':-*'"":i::!-!!.~i.::$'~. " ~ Based o n l i m i t e d i n f o r m a t i o n , s u g g e s t e d va lues of B a re 2.0 for a

:.:.: . . . . . . :q~-j~ii~-:~ "%~" .... c e i l i ng e x h a u s t i n l e t n e a r a wall. 2.0 fo r a wall e x h a u s t i n l e t n e a r Kg = 0.25 resu l t s in conserva t ive e s t i m a t ~ i ~ e n c o n s i ~ g I ~ ' n t he ce i l ing , a n d 2.8 for a c e i l i n g e x h a u s t i n l e t fa r f r o m a n v walls. I t

d e t e c t i o n of t he s m o k e p l u m e . ""~" "'~*~..,. "~ii - is s u g g e s t e d t h a t d /D be g r e a t e r t h a n 2. w h e r e D is t he d i a m e t e r o f "~'~':'~i~i?.--'.-. f f t he in le t . For r e c t a n g u l a ~ e x h a u s t in le ts , use D = 2ab/(a + bL w h e r e

3-8.5 P l u m e T e m n e r a t u r e . ' . ~ . : . y a a n d b a re t h e leng{h a n d wid th o f t h e inlet .

3-8,5.1 Average T e m o e m t u r e . Based o n t he f i rs t l a ~ : o f T h e m a x i m u m v o l u m e t r i c f low ra t e w h i c h can be e x t r a c t e d t ~ e r m o d y n a m i c s , t h e a v e r a g e t e m o e r a t u r e of t he o l u m e is t h r o u g h a n e x h a u s t inlet• is g iven as:

where:

~ =average p l u m e t e m p e r a t u r e a t e l eva t i on z (deg-F) - - amb ien t t e m p e r a t u r e (deg-F)

_~ =speci f ic h e a t of o l u m e gases (0 .24 B t u / I b deg-F) m = mass f lo r ra te of t he n l u m e ( I b / s e c )

T h e mass f low ra te of t he o l u m e can be c a l c u l a t e d f r o m e o u a t i o n (8) o r (9 ) . E o u a t i o n (8) was d e v e l o n e d for s t r o n g l y buovata t o lumes , a n d for smal l t e m o e r a t u r e d i f f e r e n c e s b e t w e e n t he o l u m e a n d a m b i e n t • e r ro r s d u e to low b u o y a n c y c o u l d be s ign i f ican t . Th i s t o n i c n e e d s f u r t h e r study, a n d . in t he a b s e n c e of~bet ter da ta . i t iS r e c o m m e n d e d t h a t t he o l u m e e o u a t i o n s n o t be u s e d w h e n th i s t e m o e r a t u r e d i f f e r e n c e is smal l (< ~t°F).

3-8.5.2 C e n t a ' l i n e T e m o e r a t u r e . T h e t e m p e r a t u r e f r o m e o u a t i o n (17) is a mass f low average , b u t t h e t e m o e r a t u r e var ies over the Dlume cross-sect ion. T h e p l u m e t e m p e r a t u r e is tzTeatest a t t he c e n t e r l i n e o f t he n l u m e , a n d t he c e n t e r l i n e t e m p e r a t u r e is o f

Vma~ = 0 .537 fldS/2 4To(T, - T o ) {20)

w h e F e "

. V ~ = m a x i m u m v o l u m e t r i c f low r a t e a t T ( ~ a / m i n )

W h e n t he e x h a u s t a t an i n l e t is n e a r th i s m a x i m u m f low ra te . a d e o u a t e s e o a r a t i o n b e t w e e n e x h a u s t in le t s n e e d s to be m a i n t a i n e d to m i n i m i z e i n t e r a c t i o n b e t w e e n t h e f lows n e a r t he inlets . O n e c r i t e r i o n for t he s e n a r a t i o n b e t w e e n in le t s is t h a t i t be a t leas t the d i s t a n c e f r o m a s ing le i n l e t t h a t w o u l d r e s u l t in a rb i t r a r i ly smal l Velocity b a s e d on s ink flow. U s i n g 40 f o m as t he a rb i t r a ry velocitv. t he m i n i m u m s e o a r a t i o n d i s t a n c e for in le t s l o c a t e d in a wall n e a r t he ce i l i ng (o r i n t h e c e i l i n g n e a r t h e wal l ) is

Smi ~ = 0 . 0 2 3 f l - ~ - ~ ~l~

63O

N F P A 9 2 B ~ M A Y 2 0 0 0 R O P

where: ~ai= = min imum edge-to-edge senaration betweeu inlets (ft) V = volumetric flow rate (~S/min) ~'= exhaust location factor (dimensionless)

It is suggested that the min imum senaration between inlets 9f enuation (21) be adhered to whenever VJ V ~ half.

3-10" Volumetric Flow Rate. For practical reasons, exnressine the stIIoke product ion rate in terms of-a volumetric rate (~S/min~ might be preferred over ~, E0ass rate. This nreference can be ~;commp~lated by dividing the mass flow rate by the density of smoke..

V= 60 m / Q (22)

where:

~_ = density of smoke (Ib/'ft s)

The volumetric flow rate ~¢termined using the above eauat ion is at the smoke laver temnerature . For a smoke managemen t system 0,~;~imled to onerate unde r euuil ibrium conditions (see ]-7.1 }. the smoke exhaust system should be designed to nrovide sufficient volumetric exhaust canacitv at the tempera ture of the smoke laver.

3-110 Maximum Air Supply Velocity. The supply velocity of the makeup air at the per imeter o f the large, open space m u s t - ~ be limited to sufficiently low values so as no t to deflect the fire p lume significantly, which would increase the air en t ra inment rate, or disturb the smoke interface. A maximum makeup supply

(1 m/see ) is velocity of about 200 f t / rnin . .re.commended, based on flame deflection data [22]. " " "

• • o 0 •

may not be detr imental .

, .~ . 3-12 0 Opposed Airflow Requirements . ~.:~ "~i~ ~"

3-120.1 To prevent smoke originating in a cummin sp~:..:~{:..[~.,. commumcat v ~ c e :~ f rom propagat ing into the large space, the

must be exhausted at a sldiicient rate to cause ..~.~..~,%rag~ velocity in the opening f rom the large space ~ ~ . l ~->. ,. limit. The limiting average velocity, v, can . .~ c a l c u i ~ : ~ ~!~':'"~':'!~

~.:#~ .~ #~ .-~:, -% ~ (~_~) v = 38 [gH ( T f - To) / ( Tf + 460) ] '/= i$.-':i$:~

where: :~'~ .....

v = air velocity ( f t /min) #: = acceleration of gravity ($2.2 f t / sec ~ ) = he ight of the opening (ft)

~jo = tempera ture of heated smoke (°F) = tempera ture of ambient air (°F)

For example, with H = 10 ft, T~ = 165°F (considered realistic for sprinklered spaces) and T o = 70°F, the limiting velocity becomes 270 fpm. For the same conditions with Tf = 1640°F (considered realistic for unspr inklered spaces), the limiting velocity becomes 594 f t /min .

3-120.2 To prevent smoke originating in the large-volume space f rom propagat ing into the communicat ing space, air must he supplied f rom the communicat ing space at a sufficient rate to cause the average air velocity in the opening to the large space to exceed a lower limit [i.e. the limiting average velocity (v,) in Equation (g~24)]. Two cases can be differentiated. In one case, the open ing to the communicat ing space is located below the position of the smoke layer interface and the communicat ing space is exposed to smoke from a plume located near the per imeter of the open space, in which case the limiting average velocity, v,, can be estimated from:

v, JoCpm) --= 17 [ Q,/z] ~/s (gg24)

where: v,= limiting average velocity (fpm) Q = heat release rate of the fire (Btu/sec) z = distance above the base of the fire to the bot tom of the

open ing fit). (See Figure 3 4 0 ~ 3-12.2.)

v, should not exceed 200 f t /min . This equation should no t be used when z < 10 ft. In the other case, the opening to the communicat ing space is located above the position of the smoke layer interface, in which case Equation (24) is used to calculate the limiting average velocity (setting v = v,), where Tf - T o is the value of A T f rom Table 3-5 a n d T f = A T + T o.

L a r g e - v o l u m e space.

Communicating space

Measurement of distance above base of fire to bottom of opening.

Chapter 4 Equipment and Controls

~ ' . 1 The dynamics, buoyancy, plume, and stratification of the ~otent ia l fire, together with the width and height of the large- volume space must all be cons idered when selecting the smoke management system. Generally, the HVAC systems designed for these spaces do no t have the capacity for use as a smoke management system, nor are the supply and exhaust air grilles located for their p roper use in such a system. In most cases, therefore, a dedicated smoke managemen t system should be considered•

4-1.2 Some existing large-volume spaces that have glass walls or skylights have been repor ted to exper ience temperatures up to 200°17 (93°C) because of solar loads. Any building materials located in such areas need to be capable of operating in this heated environmenu

4-2 Exhaust Fans. Exhaust fans should be selected to operate at the design conditions of the smoke and fire. While dilution with " ambient air might significantly cool down the fire temperature, there can be instances where the direct effects of the fire will he on the equipment .

4-3 Makeup Air System. The simplest me thod of introducing makeup air into the space is th rough direct openings to the outside such as doors and louvers, which can be opened upon system activation. Such openings can be coordinated with the architectural design and be located as required below the design smoke layer. For locations where such openings are impractical, a l ~ = d - m e c h a n i c a l supply system can be considered. This could possibly be an adaptat ion of the building's HVAC system if capacities, outlet grille locations, and velocities are suitable. For such systems, means should be provided to prevent supply systems from operating until exhaust flow has been established to avoid pressurization of the fire area. For those locations where climates are such that the damage to the space or contents could be extensive during testing or f requent inadvertent operat ion of the system, consideration should be given to heating the makeup air.

4-4 Control Systems.

631

NFPA 92B - - MAY 2000 ROP

4-4.1 Simplicity. Simplicity should be the goal of each smoke management control system. Complex systems should be avoided. Such systems tend to confuse, might not be installed correc.tl~, might not be properly tested, might have a low level of reliabdity, and might never be maintained.

4-4.2 Coordination. The control system should fully coordinate the smoke management system interlocks and interface with the fire protection signaling system, sprinkler system, I-IVAC system, and any other related systems.

4-4.$ HVAC System Controls. Operating controls for the HVAC system should accommodate the smoke management mode, which must have the highest priority over all other control modes.

4.4.4 Response Time. :]'he smoke management system activation should be initiated immediately after receipt of an appropriate activation command. The smoke management system should activate individual components such as dampers and fans in sequence as necessary to avoid physical damage to the equipment. Careful consideration should also be given to the stopping of operating equipment in proper sequence as some fans take a long time to wind down, and the closing of dampers against airflow can cause serious damage. The total response ume, including that necessary for detection, shutdown of operating equipment, and smoke management system start-up, should allow for full operational mode to be achieved before the conditions in the space exceed the design smoke conditions.

4-4.5_* Control System S~Fc-:-'=: Verification and Instrumentation. Every system e~ed~ should have means of ensuring it will operate if ~ d e d a c f i ~ t e d . The means and uf£.f..q.~l~ will vary according to the complexity and importance of the system. S-:Fc:-A:L~n ~z'-cc: -̂ .--~ "~.cl'.:~c "~c fc!!:;'.~=g:

^ x I ~ . 4 ~ ^ - - - - A ; . . . . . . . ' ^ : ^ - - ^ C . k - - . . . ' - - : . . . . . . : . . . . . ~ . 4 A ^ . . ' . ^ . ( - ~ . . . . . . . . . . . . r . . . . . . . . . . . . . . . . . . . . . r~, - ~ - ' r . . . . . . " , . . . . . . . . . .

: - - . I . . A ^ * I ~ . . . . . . . . . . e ~ . . ~ : . . . . . . . . . A . . . . . * . . . . . ~ - I I . : . . . . : +

+5"- , ~ , . . . . . dcw.ce= ~r me~q~ == zp~r~r:a te .

4-4.6 Manual Control. Manual control of all systems should be provided at a centralized location. Such controls should be able to override any interlocking features built into the automatically operated system. See NFPA 92A, Recommended Practice for Smolm Control Systems, for devices that should not be overridden.

4-5 Electrical Services.

4-5.1 Electrical installations should meet the requirements of NFPA 70, National Electrical Coder.

4-5.2 Normal electrical Dower serving air conditioning systems will generally have sufficient reliability for nondedicated zoned smoke- control svstems.

. . . . . . . . . . . . . . . . . . . . . . . . t . . . . . . . . . . . . pc . . . . . . . . . ~re~.er C . . . . . : - - . . . . . ! ~ ' = . . ^ ~ I * . ^ - - c . . . . . : - - . . . . . . ~ ^ - - ^ 2 - - - : . . . . . ' - - - -

u~l:.v/!!no. T ~ c oT=tc.":'~ : ~ c : : : ~ b c I ~ c : t c ~ ~ : =.-ca= ~ = t ;"~u!?q. n e t be .~m.7.gccl f: . . . . . . . . . . . . . . . . . ~ . . . . . m. :Fzcc.

4-5.$ Whether Or not standby power is needed should be considered for smoke-control systems and their control svstems.

4-6 Materials.

4-6.1 Materials used for systems nrovidin~ smoke control should g9nform to NFPA 90A. Standard for the InstaUation of Air- Conditionin~ and Ventilatin~ S~stems. and other aut)licable NFPA documents.

4-6.2 Duct materials should be selected and ducts desi~,ned to convey smol~e, withstand additional nressure (both Dosftive and negative~ bv the suonlv and exhaust fans when ooeratin~ in a smoke-control moc[e_ and maintain their structural integrity durin~ the neriod for which the system should onerate.

~i ,$ Euuinment including, but not limited to. fans. ducts, and balance damners should be suitable for their intended use and the nrobable temneratures to which they may be exposed.

5L

4-70th, building serving t

C Systems. When other systems in the )art of the smoke management system area, refer to NFPA 92A, Recommended ..S3s~ns, for guidance.

Testing

5-1.1 mr provides recommendations for the testing of lent systems. Each system should be tested against a criteria. The test procedures described herein the following three categories:

:Component system testing Acceptance testing Periodic testing and maintenance

5-1.2 It is recommended that the building owner, designer, and authority having jurisdiction meet durin~ the planning stage of the project and share their thoughts and objectives concerning the smoke management system contemplated and agree on the design criteria and the pass/fail performance tests for the systems. Such an agreement will help overcome the numerous problems that occur during final acceptance testing and facilitate obtaining the certificate of occupancy.

5-1.:$ Contract documents should include all acceptance testing procedures so that all parties have a clear understanding of the system objectives, testing procedures, and pass/fail criteria.

5-2 Component System Testing.

5-2.1 General. The intent of component system testing is to establish that the final installation complies with the specified design, is functioning properly, and is ready for acceptance testing. Responsibility for testing should be def ined clearly prior to component system testing.

5-2.2 Prior to testing, the party responsible for this testing should verify completeness of building construction, including the following architectural features:

(1) Integrity of any partition, floor, or other member intended to resist smoke passage

(2) Firestopping (3) Doors and closers related to smoke control (4) Glazing that encloses a large-volume space

5-2.3 The operational testing of each individual system component should be performed as it is completed during construction. These operational tests will normally be performed by various trades before interconnection is made to integrate the overall smoke management system. It should be documented in writing that each individual system component 's installation is complete

632

N F P A 9 2 B ~ M A Y 2 0 0 0 R O P

and the component is fimctional. Each component test should be individually documented, inducting such items as speed, vo lume, sensitivity calibration, voltage, and amperage.

5-2.4 Testing should include the following subsystems to the extent that they affect or are affected by the operation of the smoke management system:'

(1) Fire protective signaling system (see NFPA 72, NationalFire Alarm Code)

(2) Energy management system (3) Building management system (4) HVAC equipment (5) Electrical equipment (6) Temperature control system (7) Power sources (8) Standby power (9) Automatic suppression systems (10) Automatic operating doors and closures (11) Other smoke-control systems (12) Emergency elevator operation

5-3 Acceptance Testing.

5-$.1 The intent of acceptance testing is to demonstrate that the final integrated system installation complies with the specific design and is functioning properly. Representatives of one or more of the following should be present to grant acceptance:

(1) Authority having jurisdiction (2) Owner (3) Designer

All documentation from component system testing should be available for inspection.

5-5.2 Test Parameters. The following parameters need to be measured during accepumce testing:

(1) Total volumetric flow rate (2) Airflow velocities (3) Airflow direction ~ , , (4) Door-opening forces ~ , (5) Pressure differentials (6) Ambient temperature

5-3.$ Test Equipment. The following equipme to perform acceptance testing-

h t t

electronic manometer ( instrument ,.5 Pa) and 0-0.50 in. w.g. ( 0-125 Pa) ibing

(2) Scale suitable for measuring door--open (3) Anemometer, induding traversing equipme (4) Ammeter (5) Door wedges (6) Tissue paper roll or other convenient device for indicating

direction of airflow (7) Signs indicating that a test of the smoke management system

is in progress and that doors should not be opened (8) Several walkie-talkie radios have been found to be useful to

help coordinate equipment operation and data recording

5-3.4 Testing Procedures. The acceptance testing should consider inclusion of the procedures described in 5-3.4.1 through 5-3.4.6.

5-3.4.1 Prior to beginning acceptance testing, all building equipment should be placed in the normal operating mode, including equipment that is not used to implement smoke management, such as toilet exhaust, elevator shaft vents, elevator machine room fans, and similar systems.

5-$.4.2 Wind speed, direction, and outside temperature should be recorded for each test day. If conditions change greatly during the testing, new conditions should be recorded.

5-3.4.3 ff standby power has been provided for the operation of the smoke management system, the acceptance testing should be conducted while on both normal and standby power. Disconnect the normal building power at the main service disconnect to simulate true operating conditions in this mode.

55.4.4 The acceptance testing should include demonstrating that the correct outputs are produced for a given input for each control sequence specified. Consideration should be gaven to the following control sequences so that the complete smoke management sequence is demonstrated:

(1) Normal mode (2) Automatic smoke management mode for first alarm (3) Manual override of normal and automatic smoke

management modes (4) Return to normal

5-3.4.5 It is acceptable to perform acceptance tests for the fire protective signaling system in conjunction with the smoke management system. One or more device circuits on the fire protective signaling system can initiate a single input signal to the smoke management system. Therefore, consideration should be given to establishing the appropriate number of initiating devices and initiating device circmts to be operated to demonstrate the smoke management system operation.

5-3.4.6 Much can be accomplished to demonstrate smoke management system operation without resorting to demonstrations that use smoke or products that simulate smoke.

5-3.5 Large-Volume.Space Smoke Management Systems.

5-3.5.1 The I ~ e space can come in many configurations, each of wh ic [~ -~ . t s own peculiarities. They can be tall and thin; short and ~ " e; " ~ b a l c o n i e s and interconnectin~ floors; be open or c l o s e ~ ~ l ' ~ . ~ o r s ; have corridors and stmrs for use in evacu: ~ ' ~ a v e o ~ ' ~ ' ~ o s e d walls and windows (sterile tube); and be ~'portion o t ~ o t e l , hospital, shopping center, or az .~S~. . . .g if ic smoke gt~ criteria must be developed for nagement

5-$.5.2%~K~ffy the exact location of the perimeter of each large- volume s ] ~ J ~ . ~ o k e management system, identify arty door

" ~ t h a t space, and identify all adjacent areas that are m ~.~en]ngs ir i and that are to be protected by airflow alone. For ings, the velocity must be measured by making

~ippr0#tiate traverses of the opening.

.~. '5.3 With the HVAC systems in their normal mode, measure ~ffressure differences across all door barriers and airflow velocities at interfaces with open areas. Using the scale, measure the force necessary to open each door.

5-8.5.4 Acdvate the smoke management system. Verify and record the operation of all fans, dampers, doors, and related equipment. Measure fan exhaust capacities, air velocities through inlet doors and grilles, or at supply grilles if there is a mechanical makeup air system. Measure the force to open exit doors.

5-$.5.5 Measure and record the pressure difference across all doors that separate the smoke management system area from adjacent spaces and the velocities at interfaces with open areas.

5-3.6 Other Test Methods.

5-3.6.1 General. The test methods previously described should provide an adequate means to evaluate the smoke management system's performance. Other test methods have been used historically in instances where the authority having jurisdiction requires additional testing. These test methods have limited value in evaluating certain system performance, and their validity as a method of testing a smoke management system is questionable.

, 5-3.6.2 As covered in the preceding chapters, the dynamics of the fire plume, buoyancy forces, and stratification are all major critical elements in the design of the smoke management system. Therefore, to test the system properly, a real fire condition would be the most appropriate and meaningful test. But there are many valid reasons why such a fire is usually not [~ractical in a completed building. Open flame/actual fire testing might be dangerous and should not normally be attempted. Any other test is a compromise. If a test of the smoke management system for building acceptance is mandated by the authority having jurisdiction, such a test condition would become the basis of design and might not in any way simulate any real fire condition. More importantly, it could be a deception and provide a false sense of security that the smoke management system would perform adequately in a real fire emergency.

633

N F P A 92B - - MAY 2 0 0 0 R O P

Smoke bomb tests do NOT provide the heat, buoyancy, and en t ra inment of a real fire and are NOT useful to evaluate the real per formance of the system. A system des igned in accordance with this doc um e n t and capable of providing the in tended smoke managemen t might no t pass smoke bomb tests. Conversely, it is possible for a system that is incapable of providing the in tended smoke managemen t to pass smoke bomb tests. Because of the impracticality of conduct ing real fire tests, the acceptance tests described in this documen t are directed to those aspects of smoke managemen t systems that can be verified.

5-3.7 Testing Documentat ion. Upon complet ion of acceptance testing, a copy of all operational testing documenta t ion should be provided to the owner. This documenta t ion should be available for reference for periodic testing and maintenance.

5-3.8 Owner ' s Manuals and Instruction. Informat ion should be provided to {he owner that defines the operat ion and maintenance of the system. Basic instruction on the operat ion of the system should be provided to the owner 's representatives. Since the owner might assume beneficial use of the smoke managemen t system wherever there are complet ion of acceptance testing, this basic instruction should be completed prior to acceptance testing.

5-3.9 Partial Occupancy. Acceptance testing should be pe r fo rmed as a single step when obtaining a certificate of occupancy. However, if the building is to be completed or occupied in stages, acceptance tests of the entire system should be conducted in order to obtain temporary certificates o f occupancy.

5-3.10 Modifications. All operat ion and acceptance tests should be per formed on the applicable part of the system wherever there are system changes a n d modifications. Documenta t ion should be updated to reflect these changes or modifications.

5-4 Periodic Tesdng.

5-4.1 During the life of the building, maintenance is essential to ensure that the smoke managemen t system will per form its in tended funct ion under fire conditions. Proper main tenance of the system should, as a minimum, include the periodic testing of all equ ipment such as initiating devices, fans, dampers , c o . ~ . ~ , doors, and windows. The equ ipment should be m a i n ~ i : , . accordance with the manufacturer ' s recommendationm~:See N ~ A 90A, Standard for the Installation of Air-Conditioning ~tila~'.. ~ ~_~:~.,,~,, Systems, for suggested main tenance practices. "~#ji~.~:,,.-.~.":~- " ~ "%-..'.:'$~::.

5-4.2 The periodic tests should de te rmine t h ~ . C ; ~ l e u ~ continue to operate in accordance with t h e . ~ ' p r o v e a ~ preferable to include in the tests both t h ~ . ' ~ a s u r e m e n fa l l , quantities and the pressure " " s-.~ ~':~::':'::'::- t~.~: differentaal .., ..:~...-~:...-.~

(1) Across smoke barrier openings .#:: (2) At the air makeup supplies (3) At smoke exhaust equ ipmen t j~":

All data points should coincide with the acceptance test location to facilitate comparison measurements .

5-4.3 The system should be tested at l e~ t semiannually by persons who are thoroughly knowledgeable in the operation, testing, and main tenance of the systems. The results of the tests should be d o c u m e n t e d in the operations and main tenance log and made available for inspection. The smoke managemen t system should be operated for each sequence in the current design criteria. The operat ion of the correct outputs for each given input should be observed. Tests should also be conduc ted under standby power, if applicable.

5-4.4 Special arrangements might have to be made for the introduct ion of large quantities o f outside air into occupied areas or computer centers when outside tempera ture and humidity condit ions are extreme, and such uncond i t ioned air might damage contents. Since smoke management systems can override limit controls such as freezestats, tests should be conduc ted when outside air condit ions will no t cause damage to equ ipmen t and systems.

Chapter 6 Referenced Publications

6-1 The following documents or portions the reof are referenced within this guide and should be considered as part of its recommendat ions . The edit ion indicated for each re ferenced documen t is the current edit ion as of the date of the NFPA issuance of this guide Some of these documents might also be referenced in this guide for specific informational purposes and, therefore, are also listed in Appendix F.

6-1.1 NFPA Publications. Nadonal Fire Protection Assodation, 1 Batterymarch Park, P.O. Box 9101, Quincy, MA 02269-9101.

NFPA 70, National Electrical Code °, 4.0061999 edition. NFPA 72, National Fire Alarm Code*, 400~1996 edition. NFPA 90A, Standard for the Installation of Air-conditioning and

Ventilating Systems, 4-00~1999 edition. NFPA 92A, Recommended Practice for Smoke-Control Systems, 4002~

2000 edition. NFPA 101 ®, Life Safer 3 Code ®, ~ 0 4 2000 edidon. NFPA 204M-, Guide for Smoke and Heat Venting, -1-0Ot- 1998 edition.

6-1.2 Other Publications.

6-1.2.1 UL Publications. Pfingsten Road, ~ b r o

UL 555, S t a ~ J o r Saf,

Control S ~ s ~ 19~..":~.'...

Underwriters Laboratories Inc., 333 ~k, IL 60062.

Fire Dampers, 1999. y Leakage Rated Dampers for Use in Smoke

Explanatory Material

anoroach [CooDer et al. 1982 and Peacock an d Babrauka, (1991)] is to use linear ]nteroolat ion of the point measurements . U~in~ temperature data. the interfaces are at the heights at which the temperatur~ i* as follows=

z -~__c~- r~ + r~.

where."

~mL = the temnerature in the smoke laver = the tempera ture in the cold lower laver

_C = an interoolat ion constant with values of 0.1--0.2 for the first incfication of smoke and 0.8-0.9 for the smoke laver interface, resnectivelv.

A-l-4 Transition Zone. (See aho A-3-8.1.2 for further details. I

A-1-5.4.1 The per formance obiective of automatic snrinklers installed in accordance with I~FPA 13. Standard fur ¢h¢ Installation of St~rinkler S~stems. is to provide fire control whicl~ is def ined as follows: L~mitin~ the size of a fire by distribution of water so as to decrease the heat release rate and ore-wet adjacent combustibles. while controll ing ceilin~ ~as temoeratures to avoid structuvo,] damage. A limited number of investi~-ations have been undertaken in which full-scale fire tests were conc/ucted in which the ~pHnkler system was chal lenged but nrovided the expected level o f oerformance. These investigations indicate that. for a fire control situation, the heat release rate is limited but smoke can continue to be oroduced. However. the tempera ture of the smoke is reduced.

Full-scale sprinklered fire tests were conducted for open-plan office scenarios [Madrzvkowski and Vettori 1992. Lou~heed-199-7]. These tests indicate that there is an exnonential decav]n the heat release

634

N F P A 92B - - MAY 2 0 0 0 R O P

rate for the snrinklered fires after the sDrinklers are activated and achieve control. The results of these tes~ ~lso in~iicate that; design fire with a steady-state heat release rate of 500 kW provides a ¢0nserv~0ve estimate fol" i~ spdnklgred open-plan 9ffice.

There is lilnited full-scale test data available for use in determining design fire size for other sorinklered occuDancies. Hansell and Morgan I~RE 2581 provide conservative estimates for the ¢onvectivg heat release rate based on OK fire statistics: 1 MW for a s_.12dnkere d office, 0.5-1.0 MW for a sDrinklered hotel bedroom and 5 MWfo[ a sprink]ered retail occuoancv. These steady-state desires fires assume the area is fitted with standard resnonse sndnklers.

Full-scale fire tests for retail occupancies were conducted in AOstralia lBenn etts et al.l. These tests indicated that for some commofi I'etail O~30ets ¢qlQthing and book stores) the fire is controlled and eventually extinguished with a single sDrinkler. These te~t# also indicated that the sprinklers mav have difficulty ~uppres~iog a fire in a shop such as a toy store with a high, fuel load.

temperature= and -:~nd :'¢!vc:'-e: =-e n~t g!:'zn. T~..c:: dam :'.¢¢d :v bc ~:cd :;-=.h c~'a~c.n._

A-1-6.$ One source of data is the ASHRAE Handbook of ~'undamentals. Chanter 26. "Climatic Design Information." It is su~e~%~d that Lhe 99.6% headng dry bull~ (DB/ temuerature and the 0.4% cooling DB temnerature be used as the winter and ~umrqe I" design ~ondition~ resnectiveiv. It is also suggested that the ~1% extreme winql velocitv be used as the desires con-cl-ition. Where ~vailabl¢. more ~ite-sDetfic data should be consulted.

#-2-4.1.3 A sDreadsheet model or other time step model can be ¢gBstructed using the algebraic ~quations contained in Section $ in order to calculate the nosition of a smoke laver interface over time. both with and without-smoke exhaust in operation. This annroach involves the calculation of the mass of smoke entering the smoke layer, the temDerature of the smoke entering the laver, and volume of smoke removed from the laver bv the mechanical exhaust. The ~teps usgd to determine the Dosition of the smoke laver interface are as follows:

Full-scale fire tests were conducted for a variety of occunancies (retail. cellular offices and libraies) in the United Kingdom [ Ghosh 1997].

(a) Select the time step for the calculation. A.£ (b) Determine the design fire (e.~.. steady-state, growing fire.

F~ll-scaie fire tests were conducted for comDact mobile storage svstems used for document storage. Information on tests conducted in 1979 on behalf of the Library of Congress is orovided ill Aooendix G of NFPA 910 Protection of Libraries and Li-brarv Collections. Subseouent full-scale fire tests were conducted for the Librarv of Congress-Archives II and the National Library of Canada and showed that fires in compact mobile are difficult to extln~uish [Lou~heed et al 1994].

&-1-5.4.2 During the initja~ active phase of the fire with the sprinklers ooeratlng, the smoke laver remains stratified under the

the moke 1 will I ra id d s toroughqut the volume a~ buoyancy decays,

application density of 0.1 g p m / f t ~ and <500 kW at an appligation densltv of 0.2 ~nm/ f t ~. For higher heat release rates, the smoke temperature will be above ambient and will be buoyant as it leaves the sDrinklered area.

For low heat release rate sorinkered fires, the smoke is mixed over tOe height of the comnartrnent. The smoke flow through large oDenings into an atrium will have a constant temnerature with

With higher heat release rates, a ho t UDDer laver will be formed. The temnerature of the UDDer laver will be between the ambient temperature and the oDeratina temnerature of the sorinkler. If the smoke is hotter than the sprinkler 9perating temp¢l'ature, f~rth~r sDrinlders will be activated and the smoke will be cooled. For design DUrDOSeS. a smoke temDerature eGuivalent tO the operating temperature of the sDrinklers can be assumed.

A 1 $.~ Wcatkcr !:-£~rmafic.n f~r ma=:y N~r '~ ?~mcrica=: and :~mc . . . . . . . . . . . . . . . . . . ~ . . . . . . . . . . . . . . . . . . . . . ~ . . . . . . . . . . . . ~ j

F:~d~.m~r~:a::. ,~?.v.:t ;':cz:kcr data "; co!looted at municipal airpcTt~

~veraging this value with the incremental mass from the previo¢~ tUar_m~

(e) Calculate the temperature of the smoke entering the under laver. The eouations for calculating temnerature can be founcl in

v

Tabte ~5~ (f) Add the mass of smoke enterin~ the UDDer laver to the total

mass of smoke to obtain the new total mass of smoke in the UDDer

ml* mi_e_m ~

where."

m z = to~al mass of smok¢ of smgke at eqd of previous time steD

m 2 = new total smoke mass in upper laver (kg)

(g~ D~;termine the new temperature of the upper laver via conservation of ener~,v. The higher the laver temperature, the greater its volume angel lower the smoke laver interface. Therefore, it is conservative to assume no heat losses f rom the UDDer laver to the compar tment boundaries over time.

T~ --__LmlZ1 + miT1)Z_~°C

(h) Determine the densitv of the new uDDer laver:

p ~ ~ + 273) / ( T: + 273) ( k g / m 3)

(i) Determine the volume of the new UDDer laver:

635

N F P A 9 2 B - - M A Y 2 0 0 0 R O P

~__~ze_tmLt

(i) Subtract the volume of the smoke removed, if anv. via mechanical ventin~ over the riven time sten to determine the final uDDer laver volume

(1~) Determine the new smoke laver interface oosition as a function of the final under laver volume, and the tteometrv of the sln?ke reserv?ir. For re¢tangular geometries, t h e s m o k e laver position is calculated as follows:

Layer interface (m) = ceilin~ height - - (final upper laver volume LE~.)/area of reservoir)

(B Return to step (d) and use the newlv calculated laver interface to calculate the mass of smoke enterimt the smoke laver in the subseouent time steP.

Existing A-3-1.1.3 remains unchanged.

Existing A-3-1.2.1 remains unchanged.

Existing A-3-1.2.2 remains unchanged.

The following examples are included to provide insight into the w~y that the Froude mo~leling scaling relations are used. -

F, xample 1, What scale model should be used for # mall where the

~,xampl¢ 5, In a ?ne tenth scale model, the following clear heights were observed: 9.5 m at 96 seconds. 1.5 m at 85 seconds and 1.0 m at | 52 seconds. What are the corresvondimt clear heights for the full scale facility?

F?r the first clear height and time hair of z a ffi 2.5 m at t z ffi 26 s:

and

2.5(10/1) = 2 5 m

t F = t . [ ~ - [ = 26(1011) i n = 8 2 s tl )

The other clear hei~rht and time hairs are claculated in the same lll~nner. ~nd thev are all listed belo~.

Scale Model Observation Full Scale Facilltv Prediction

Qlear Height (m~ ~ Clear Height (m~

~ . ~ . ~ ' ~ . 15~ smallest area of interest is the floor to ceilin~ height on the 10 480 balconies which i s 3 m ? - n ~ : ~ % : r ~ "

Existi t unchanged.

~ote tha~ il~ is essential that the flow in model is fvlly develope0 l-sca r onen-nlan offices [Madrzvkowski l~urbulent flow. and to achieve this it i~ suggested that areas of a l t~e t to ' i~1992 and L ~ l t h e e d 19971 "have shown that. once the interest in the scale model be at least 0.3 m. The corresoondin~ ~- : snri~l~ri-~'tx~/~ control of t-he fire bu t are no t immediately able to floor to ceilin~ height of the model should be at least 0.~ m. Set I e ' x t i n ~ ' i t due to the fuel configuration, the heat release rate will ffi 0.3 m an0 /1: = 3 m. then /~/-/1: ffi 0.1. Therefore. the model can b ~ d e cr eas e"~. "~'~'~aat~e nti ailv follows~ 9he tenth ~cale. ~ ~ - . . . . ~ . ~

a s

Exara~le 2. The desi£m fire for a snecific facility is a constant fire " ~ ~ ' ~ ' O a c t e - k t - - 9f 5000 kW, What size fire will be needed for a one tenth s c a d ~

1~/-~ =0.1 . ~ . _ . . . - 4 ~'" ~( t i : the heat release rate at time t after smSnkler activation ( B t u / s e c ~

(1 .~5/2 f f ~ > ~ . , ~ : . . .(~ffi the heat release rate at sprinkler a~ztivation~ (Btu/s~ " < ~ ¢ = time after sDdnkler activation (sec) Q~ = Qe/TM / = 5000(0.1)'/ .~+~+15.8~".~ :#~:<" t l F ) ~ . ~ " ~ 'Sg" k = decay c o n s e n t (sec'll

~ ' ~ . :~ Estimates for the decay constant for office occunancies orotected Examole 3, The smoke exhaust rate for a full fft~/iff~, facik'~ is 250 with a discharge density of 0.1 t rom/~ ~ are 0.0025 for situations with

mS/sec, what is the smoke exhaust rate for a one ~ : a l e li~tht fuel Ioacls in shielded a r e ~ [Madrzvkowski and Vettori 19921 model? "7

," \ 5 / 2

Vfan, m "- V f £ ) = 2 5 0 ( 0 . 1 ) 5 n = 0 7 . 9 m 3 / s e c ) Examble 4. The walls of the full scale facility are made of

concrete. What is the impact of constructin~ the walls of a one tenth scale model of ~-¢os'um board?

of brick is 1.7 k W i m "4 K ~ s.

The ideal thermal orooerties of the model can be calculated as

(kpc).,,. = (kpc)~, r = (1.7)(0.1) o.9 = 0.21kW2m-4K-2s

The value for ~ w s u m board is 0.18 kW ~ In "4 K ~ s which is close to _ . _

the ideal value above, so that the ~t~rnsum board is ~ood match. It should be noted that usin~ class windows for video and PhotograPhs would be more imnortant than scalin~ of thermal properties.

and 0.00155 see'* for situations with h e a w loads rLou~heed 19971.

Delete existing A-3-B.2.2.

A-$-4 Another case for which a solution has been deveioned is deDicted in Figure A-$-4. In this case. the ambient interior air wi(hin the large snace has a constant temnerature madien t 0 ;¢mperaturechange per uni t height~ f rom floor level to the ~Hing. This case is less likely than temneratures that anuroximate._

~¢n function. For the linear temperature profile, the maximum h~i~l~t that smoke will rise can be cierived from the vioneerin~ work of Morton. Tavtor. and Turner f l l l .

Figure A-3-4 (Figure was unavailable for the ROP)

. I ~ (DT/ d,zY~/s

where'. a~ = maximum height of smoke rise above fire surface (ft~ .Q. = convective vort ion of the heat release rate tBtu/sec~

AT/d~ffi rate of ¢liange of alnbient temnerature with resnect to heicht ( °F/ft]

636

N F P A 9 2 B ~ M A Y 2 0 0 0 R O P

T h e convective Dortion of the hea t release rate. O. can be est imated as 70 percen t of the total hea t release r~t¢,

The m i n i m u m O_ reou i red to overcome the ambien t t e m p , ra ture difference and d~ive tlae smoke to the ceiling (z = /i3 follows readily f rom Eoua t ion f5/.

= m i n i m u m convective hea t release rate to overcome stratification (Btu / sec )

/-/= ceiling he igh t above fire surface fit)

_AT o = difference betweeq ambien t t emnera tu re at the ceiling and ~mbient t empera tu re at the level of the fire surface

Alternativelv. an express ion is provided in t e rms of the amb ien t t ¢ m p e r ~ u r e incr¢~lse frgl B floor to ceiling, which is_lust sufficient tO prevent a p lume of hea t release. O~. f rom reach ing a ceiling of

Finally. ~ a third plternative, the m a x i m u m ceiling clearance to

wlaich a n l u m e of stren~_h. O~. can rise for a # v e n A.T,. 2 follows f rom rewriting the above eoua t ion :

Exist ing A-3-5 remains u n c h a n g e d .

Existing A-3-6 remains u n c h a n g e d .

Existing #.-3-6.2.2 remains u n c h a n g e d .

Existing A-3-6.2.4 remains u n c h a n g e d . ~-~"-q-~-~-- , ~ : ":*:;:~

Delete exist ing A-3-7.1.5. "A{~,_ ~,,:.,_

At equil ibrium, the he igh t z in Eouat ion (8/ is the location of the smoke laver interface above the fuel level (see Ficure 1-41. The t ransi t ion zone is located below this level. For an efficient smoke m a n a g e m e n t system, the den t h of the t ransi t ion zone is annroximate lv 10% of the a t r ium height . In the transi t ion zone. the tglnpera~re and 9 ther smoke paramete rs decrease linearly with he igh t between the smoke laver interface he igh t a n d the lower edge of the transi t ion zone.

Existing A-3-7.2.1 r ema ins u n c h a n g e d . R e n u m b e r as A-3-8.2.1.

A-3-9 T h e eouat ions for Dlugholing were originally developed for I~i~tural vents lThomas . P.H-. Hinc-klev P .L. . -Theohald . C~R. and Sims. D.L., Investigations into the flow of ho t gases in roof venting. Fire Research Technical Paner No. 7. HMSO. London . 1963.1 It- has also been aDnlied to mechanica l smoke exhaus t system by Hincklev (62). - 'i 'he numer ica l factors inc luded in Eouat ions (19

assume the exhaus t inlets ;~re located nea r a wall. Larger factors can be used if the inlets are located near the center of the smoke reservoir [Sce &ct i~s 2.8 and 1.5 for Eauations (19 a~d 20). r esbectivel~ 1.

Al though the equat ions were develoned for natural venting. nhvsical and numer ica l mode l i ng studies conduc t ed io lndv bv )~SHRAE and N]RC [Lougheed and Hadi i sonhoc leous 1997. Loughced . Hadi isonhocleous . McCar tuev a n d Taber 1999 an d HadiisQphoclegtjs. L o u g h e e d and Can 1999] indicate they are also aoolicable to mechan ica l exhaus t systems. These s tudies used oh-vsical models, which were 5.5 m and 19.9 m in he igh t with volumetr ic flow rates o f u o to 25 m S / s e c for a single exhaus t inlet (average exhaus t inlet velocities o f u n to $0 m / s e c L Th e nhvsical mode l results indicated tha t the smoke d e n t h could be reduced to anoroximate lv 10% of the clear he igh t bv us ing mul t ip le exhaus t inlets to min imize the mass /vo lume t r i c flow rate at each exhaus t inlet. The numer ica l mode l studies indica ted tha t the results could be scaled to h ighe r atria.

Bv increas ing the n u m b e r of exhaus t inlets, the velocity at each exha~s~ inlet ~;gvld be reduced. T h e h ighes t efficiency fqr the ohvsical m o d f l exhaus t System was obta ined if the inlet velocity w ~ i imited go 10 m / s e c or less. It is also r e c o m m e n d e d tha t t h e ratio of the smoke laver dep th to the d iamete r of the exhaus t inlet (d/D) be greater t han 2 (for-rectan~alar exhaus t inlets, use D = 2ab/(a+b). where a a n d h are the l eng th -and width o f the exhaus t onen ing .L

system. The considera t ions ou t l ined in this sect ion are impor tan t when dea l ing with system in which the desima reou i rement - fo r the clear he igh t is lust below the exhaus t inlet lleight[

25

20

15

10

5

0 ,

.o.9

_= o >

E E

T = 5 o c / . . . . . . . . . . T = I0°C

T = 2 5 ° C - - , , / / ( T=50°C i : / / _ _ T = 75 °C / / / / T = IO0 °C / / / . - - T= 125:0 ; ~ ) / / / T= 150 C / / 7 7 . / "

• ' / . . ' / / / ) 2 / / " f ' f J , f * S

" ~ ' ° / / " o" ~" . . n / ~ / . ' ~ / ' , " ...

.- ...... -

- , . - : - " , . . . . 1 2

Smoke depth (m)

Figure A-3-9.

A-3-10 Densi t 7 of smoke is approximate ly equal to the density of air. The denst ty of air at 68°F at sea level isO.075 l b / f t s. Th e densi ty of air at a n o t h e r t empera tu re can be calculated from:

637

N F P A 92B - - MAY 2 0 0 0 R O P

P/P0 = 5 2 8 / ( 4 6 0 + T)

w h e r e :

P0 = 0 .075 ( l b / f t a)

p = d e n s i t y o f s m o k e a t t e m p e r a t u r e ( l b / f t s)

T = t e m p e r a t u r e o f s m o k e (°F)

A - 4 4 . 5 V e r i f i c a t i o n dev ices c a n i n c l u d e t h ¢ fo l l owing :

(1) E n d - t o - e n d v e r i f i c a t i o n o f t h e w i r i ng , e o u i D m e n t , a n d dev ices i n a m a n n e r t h a t i n c l u d e s n r o v i s i o n f o r nos i t ive c o n f i r m a t i o n o f ~qt iva t ion . p e r i o d i c t e , t ing . a n d m a n u a l o v e r r i d e o p e r a t i o n

.{2) T h e o r e s e n c e o f o n e r a t i n g Dower d o w n s t r e a m o f all c i r c u i t d i s c o n n e c ~ - -

{ 3) Pos i t ive c o n f i r m a t i o n o f f a n a c t i v a t i o n bv m e a n s o f dug~ o r e s s u r e , a i r f low, o r e o u i v a l e n t s e n s o r s t h a t r e s n o n d to loss 9 f o p e r a t i n g Dower . n r o b i e m s in t h e D o w e r o r c o n t r o l c i r c u i t w i r i n g . ai-rflow r ~ t r i c t i o n s , a n d f a i l u r e o f t h e bel t . s h a f t c o u D l i n g , o r t n o t o r

(4) Posi t ive c o n f i r m a t i o n o f d a m n e r o p e r a t i o n bv c o n t a c t , o r o x i m i t v , o r e o u i v a l e n t s e n s o r s t h a t r espQI ld to loss o f o p e r a , p c

L-" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Z " . . . . . . . . . . . . . . . . . . . . . . . .

• ~=cr r r - g ~ t ~ a ; ' c t c c c n v c ~ "n fc r : ' n . a ' 2c : f r c m t.hc~c tab!c= i n t c "~hc fc:~, , r e q u i r z d ~)" c q u = f i o n : F r c z c n t c d c ! : c w ~ c r z in ~ i ~ d o c u m e n t .

T h e ~ ' .=rn!ng r a t e e f .'x..ateri~= c~ ' : b e r e l a t e d t c t.he h e a t r e ! t r e e ~ ' ^ o f ma=c r i a ! : ~) . . . . t . : ~ . . . _ _ ,t. . . . . . ~ . . . . . . . t,.. ,~.^ ^~w^_,:..^

g ! ; ' en a~ a n e ~ e r g 7 r c ! e ~ e r a t e t- . . . . . . . . . . . . . . . . . . . . . . . . . . . . o _ 1 . , . ~. . . . . . t . . . . d = L a r ~ _ r..^~, t . . . ~ : _ ~ in L . ~ z r~A1

A n n e n d i x B P r e d i c t i n g t h e R a t e o f H e a t R e l e a s e o f F i r e s

B- I I n t r o d u c t i o n . T h e f o l l o w i n g p r e s e n t s t e c h n i o u e s f o r e s t i m a t i n g t h e h e a t r e l e a s e r a t e o f v a r i o u s fue l a r r a y s l ike ly to b e p r e s e n t in b u i l d i n g s w h e r e ~ m o k e v e n t i n g is a p o t e n t i a l f i r e safety. p r o v i s i o n . It p r i m a r i l y a~ldresses t h e e s t ima ! iQn o f fue l c o n c e n t r a t i o n s f o u n d i n re ta i l , s t ad ia , o f f ice itIlO s i m i l a r l oca t io I l s t h a t m a v involve l a r g e a r e a s a d d r e s s e d b v th i s g u i d e . Conve r se ly , N F P A 204. Guide for Smoke and Heat Venting. a d d r e s s e s t h e t w e s o f fue l a r r a y s m o r e c o m m o n to s t o r a g e a n d m a n u f a c t u r i n g l o c a t i o n a n d o t h e r w o e s o f b u i l d i n g s i t u a t i o n s c o v e r e d bv t h a t s t a n d a r d . N F P A 9 2 B is a n o l i c a b l e to s i t u a t i o n s w h e r e t h e h o t l ave r d o e s ug~

Dower o r c o m o r e s s e d a i r . n r o b l e m s in t h e Dower . c o n t r o l c i r cu i t , e - - " " ~. . " " "s " o r p n e u m a t i c l ines , a n d f a i l u r e o f t h e d a m n e r a c t u a t o r , l i n k a g e , OF • • .~ : :# .~

d a m o e r i t se l f " " " w e e " " e (5)- P e r i o d i c a c c e n t a n c e t e s t i n g in a c c o r d a n c e wi th C h a p t e r ~ " v v ':': " "s ( 6 ) O t h e r dev i ce s o r m e a n s as a p p r o D r i a t e "v -- "

I t e m s ( l ) t h r o u g h (5) d e s c r i b e m u l t i o l e m e t h o d s t h a t m a y b e u s e d . e i t h e r s i ng ly o r in c o m b i n a t i o n , to ver i fy t h a t al l por f ior l$ o f ~,.~x~ ~ _ _ . _ _ - . . _-..^_ __.._, _ . _ ~ : . . _c a_._ ~ . " " - - o " _ _ ~ . e . " t h e c o n t r o l s a n d e q u i p m e n t a r e o o e r a t i o n a l . F o r e x a m n l e . c o n v e n t i o n a l ( e l ec t r i ca l ) s u p e r v i s i o n m a y b e u s e d to ver i fy t h e i n t e g r i t y o f t h e c o n d u c t o r s f r o m a f i r e a l a r m sys tem c o n t r o l u n i t to

- " w ' " " • ¢ ! : . ( 1 ) Lhe r e l ay c o n t a c t ~thm ~ f e e t o f t h e c o n t r o l sys tem i n o u t (see 2VFPA ~4:';:~.~ . . . . . ~,;~.:.-:-::~- . . . . 72. National Fire Alarm Code °. Section 3-91 ~t ld e n d - t o - e n d ~ . ~ . A c t u ~ : ~ e s t s o t S.lrq[l~ r a r r a y s . . . . . . . . . v "~ " , - . ~ ; ' ~ , . :$k "':','T:.~?._,X~algommms Der ived / r o m tes ts o I a r r a y s n a w n g s l m n a r tUgl~ e r m c a t l o n m a v De u s e d to venv¢ o o e r a t l o n t r o m t n e c o n t r o l sy s t em .:s~ ,-:.-:-~:.:.--..:.:.:-:.x.:.- . . . - . . . . " . . . . . . ~ - , . ~ - - "~-S: .~,'ld ~ i ! ~ n s l o n a l c h a r a c t e r i s t i c s InPUt to m e cieslre(1 e n a resu l t . I t o a t l e r e n t sys t ems a r e u s e d . m . ":!~:.~::.:: ..... . . .

(4) r ,~a l cu lauons b a s e d o n t e s t e d DroDertl¢~ a n d m a t e r i a l s a n d verifv d i f f e r e n t n o r t i o n s o f t h e c o n t r o l c i r c u i t a n d / o r c o ~ i i : : % :: - - • - • ..:.'.: • ~--:-:'~ 5gkex~eg~¢d f l a m e f lux e q u i p m e n t , t h e n e a c h sys t em w o u l d b e r e s p o n s i b l e for~z'fildicatlt~ ::.:':i:-~i:~, . . . . . . . . . . . . . .

~- " - - . • " • • - .:;:-***:.. :-~.--. :::~ ~.-'s:::-iOl M a m e m a t l c a l m o o e t s o i n r e s o r e a a a n d a e v e l o n m e n t o H - n o r m a l c o n c u t l o n s o n its r e s p e c t i v e s e g m e n t . ~a'.-:.-.'~:.-:i~.¢:.~:: ~ : . ..:::., ::::.. - - . . . . ""-'~.".-'."t-~.::. # ::~ ":~:~a~.::-

-:~t-:':: ~ : : B-$ A c t u a l T e s t s o f t h e A r r a y I n v o l v e d W h e r e a n a c t u a l ca lo r i f i c E n d - t o - e n d ve r i f i c a t i on , as d e s c r i b e d in s e c t i o n A , * a , ~ . m o n f f ~ : , b o t h "'" . • the e l e c t r i c a l a n d m e c h a n i c a l c o m p o n e n t s o ~ - ~ o n t r ~ l ~ i i ! J . ! k . : : . - ' : . ' ~ t e s t o f t h e SDecific a r r a y u n d e r c o n s i d e r a t i o n h a s b e e n c o n d u c t e d sys tem. E n d - t o - e n d v e r i f i c a t i o n n r o v i d e s o n ' r i v e c o n ~ t i o n ~ t a n d t h e d a t a is in a f o r m t h a t c a n b e e x p r e s s e d as r a t e o f h e a t t h e d e s i r e d r e s u l t h a s b e e n a c h i e v e d d u ~ e t i m e t h a ~ "*: r e l ease , t h e d a t a c a n t h e n b e u s e d as i n p u t f o r t h e m e t h o d s in th i s c o n t r o l l e d dev i ce is ac t i va t ed . T h e i n t e n t t ) ~ : ~ t o - e n d v ~ i f i c a t i o n g u i d e . S i n c e a c t u a l t es t d a t a s e l d o m p r o d u c e s t h e s t e a d y s t a t e g o e s b e v o n d d e t e r m i n i n g w h e t h e r a c i r c u i t fa t l~z- i~sts . ~ i n s t e a d a s s u m e d f o r a l i m i t e d - ~ r o w t h f i re o r t h e s o u a r e o f t i m e ~ r o w t h ~ c e r t a i n s w h e t h e r t h e d e s l r e d e n d r e s u l t (i.e. a i ~ d a m n e r a s s u m e d f o r a c o n t i n u o u s - g r o w t h ( t - s q u a r e d ) f i re . e n ~ n e e r i n ~ a s c e r t a i n s w h e t h e r t h e d e s i r e d e n d r e s u l t fi .e. a i ~ . : ~ d a m D e r p o s i t i o n ) is a c h i e v e d . T r u e e n d - t o - e n d verif icat ion~' .-"~erefore~ r e a u l r e s a c o m n a r i s o n o f t h e d e s i r e d o n e r a t i o n to ~ e a c t u a l e n d resu l t .

A n " o p e n " in a c o n t r o l wire. f a i l u r e o f a f a n bel t . d i s c o n n e c t i g n 0 f a s h a f t CouDling. b l o c k a ~ e o f a n a i r t i l te r , f a i l u r e o f a m o t o r , o r o t h e r a b n o r m a l c o n d i t i o n w h i c h c o u l d D r e v e n t D r o n e r o p e r a d ~ , is n o t e x p e c t e d to r e s u l t in a n o f f - n o r n ~ l i n d i c a t i o n w h e n t h e c o n t r o l l e d dev ice is n o t a c t i va t ed , s ince t h e m e a s u r e d r e s u l t a t t h a t t i m e m a t c h e s t h e e x n e c t e d resul t . If a c 0 p d i t i o n t h a t p r e v e q t s D r o n e r o D e r a t i o n Derslsts d u r i n g t h e n e x t a t t e m p t e d a c t i v a t i o n o f t h e - d e v i c e , a n o f f - n o r m a l i n d i c a t i o n s h o u l d b e p r o v i d e d .

E x i s t i n g A-5-3.6 .2 r e m a i n s u n c h a n g e d .

AFFcaa-: .x ~-. H e a t R : ! : . ~ c R a t e ~ a ' ~

, v ~ : . ~ . . . . a : _ : . . . . . . . . . . , , r . . . . . . . . . ~ , : . . . . t , L : . - . - - ' ~ r w ' ?

. . I . . . . . , t I . . , ~ 2~ ~ - - . I ~ . . I ^4 ; ^ . ; - - 1 " ^ - - -~2^ - - - I ~ . , ~z . . . . . . . . I . ,

. . . . "~^':-~o . . . . . . . c._^ T ~ c f c ! ! c ; ; ' i n g *.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

J u d g e m e n t is u s u a l l y n e e d e d to d e r i v e t h e a c t u a l i n o u t n e c e s s a r v i f e i t h e r o f t h e s e a n n r o a c h e s a r e u s e d . (See A o o e n d i x C f o r fur t Jaer de t a i l s r e l e v a n t to t - s a u a r e d f i res . ) I f a c o m p u t e r m o d e l t h a t is a b l e to r e s p o n d to a r a t e o f h e a t r e l e a se ver~ns t i m e c u r v e is u sed . t h e d a t a c a n b e u s e d d i r e c d v . C u r r e n t l y t h e r e is n o es tab l i shec l c a t a lo~ o f tes ts o f snec i f i c a r r ays . S o m e t e s t d a t a c a n b e f o u n d in t e c h n i c a l r e n o r t s . A l t e rna t ive lv . i n d i v i d u a l tes ts c a n b e c o n d u c t e d ,

M a n v f i re tests d o n o t i n c l u d e a d i r e c t m e a s u r e m e n t o f r a t e 9 f h e a t r e l ease . In s o m e cases , i t c a n b e d e r i v e d b a s e d o n m e a s u r e m e n t o f m a s s loss r a t e u s i n g t h e f o l l o w i n a e a u a t i o n :

Q = mhc (~-1)

( O = r a t e o f h e a t r e l e a se in kW. 3= d e n s i t y in k g / s . h c = h e a t o f c o m b u s t i o n in k l / k g )

In o t h e r cases i t c a n b e d e r i v e d b a s e d o n m e a s u r e m e n t o f f l a m e h e i g h t as fol lows:

Q = 3 7 ( / , + 1.02 D) 5/~ (B-~)

( O = r a t e o f h e a t r e l e a s e in kW. L = f l a m e h e i g h t in m. D = f i re d i a m e t e r in m )

638

N F P A 9 2 B ~ M A Y 2 0 0 0 R O P

B-4 Actual Tests of Arrays Similar to that Involvggt, Where an actual calorific test of the soecific arrav under consideration cannot be found, it may be possible to find data on one or more tests tha~ are similar to the fuel of concern in imnortant matt¢l~ such as ~/pe of fuel. arrangement , or i~nition scenario. The morg the act~lai tests are similar to the fuei of concern, the hi~her the confidence that can be olaced in the derived rate of heat-release. The additiq[i of engineeri-n¢ iudmnent, however, may be needed to adjust the test data t-o that at~t)roximating the fuel of concern. If rate of heat release has not been direcdv measured, it can be estimated using the method described for estimating burnin~ rate from flame height in Section B-3.

11-5 Ahtorithms Derived from Tests of Arrays Havin~ Similar Fuels and Dimensional Cb~acteristics.

B-5.1 Pool Fires. In manv cases, the rate of heat release of a tested array has been divided by a common dimension, such as occuoied floor area. to derive a normalized rate of heat releas¢ per unit area, The rate of heat release of oool fires is the best documented and accented algorithm in thi~ class.

An eouation for the mass release rate from a nool fire is as follows- [Balprauskas 199~];

m " = ,, ~ -kBDx m o t t - e ) (B-3)

The variables 5 for eouation B-3 are as shown in Table B-5.1 lBabrauskas 19951.

Delete existing Table B-1.

The mass rates derived from eouation B-3 are converted ~9 rates of heat release nsin~ eouation B21. and the h o t of combu, t jon from the Table B-5.1. ~'he rate of heat relea#¢ per uni t area times the area of the oool fields heat release data for the a n t d n a t e d fire.

!]-5.2 Other Normalized Data. Other data based on burning rate per unit area in test~ have been ~eveloped. "gables B-5.2{~) ~md B-5.2(b) list the most available of these dam.

Mateml

_C~'og~ics*

LPG (mostly C:_H~_ Alcohols

metl-anol (CH.OH~ __maao l_~ /a~ la3_ Simnle ortranic fuels A

4.4 29_Q~ 1.9

H

~ x n z e n ~ f f ~ . ".'~:. '~."-:~ ~-~ 17.00o 0.017 ~ - - -,.:~..:~ :...'~. _ _ _ Z L C ~ - - - - f f ~ " ~!i~.::.. :"~~ 41 19.000 0.015

_____~ xa n~ J_C_~lt~)_ 65 11.000 0.00S~* diethvl e ther ( C:_Hx0_O_)_ 45 14.500 0.017 0.21

Petroleum products b e n z i n e 46 19.000 O.0O98 1.1 _ _ _ . g a ~ l i a e 46 19.000 0.011 0.64 k e r o s i n e 51 18.500 0.008 1.1 _ _ _ j 2 = ~ 47 18.500 0.01 1.1 ~ _ d £ - 5 51 18.500 0.011 0.49

warts former oil. 47 20.000 0.008** h~.rocarbon f u e l oil. h c_~.w~_ 59B 6_~ 17.000 0.0072

crude oi!_ ~ l ~ 18.000 0.O045 B0.0092

So ~IL4L ___p_Q lymet hyimet bacrvtat e 74 10.000 0.0041

_ _ _ ~ t ~ r o ~ u ~ _ _ L c ~ 56 18.500 0.00~7 _____~lystyrene ( C. I:~.~ 66 17.000 0.007

1.0

*Fog_gools on cky land. n g t over water. **E~fi mate unc er lain sin ce o n lx l~d ;~ .~4ao in~ a~U able. HValue i n ~ e n d e n t of diam eter in turlml ent regime.

639

N F P A 92B - - MAY 2 0 0 0 R O P

Table B-5.2(a) Unit Heat Release Rate for Commodities Heat release rate per unit floor area of fully involved combustibles, based on

negligible radiative feedback from the surroundings and 100 percent combustion efficiency.

Btu/sec°fC of Floor Area

Commodity Wood pallets, stacked 11/2 ft high (6-12% moisture) 125 Wood pallets, stacked 5 ft high (6-12% moisture) 350 Wood pallets, stacked 10 ft high (6-12% moisture) 600 Wood pallets, stacked 16 ft high (6-12% moisture) 900 Mail bags, filled, stored 5 ft high 35 Cartons, compartmented, stacked 15 ft high 150 PE letter trays, filled, stacked 5 ft high on cart 750 PE trash barrels in cartons, stacked 15 ft high 175 PE fiberglass shower stalls in cartons, stacked 125 15 ft high PE bottles packed in Item 6 550 PE bottles in cartons, stacked 15 ft high 175 PU insulation board, rigid foam, stacked 15 ft high 170 PS jars packed in Item 6 1,250 PS tubs nested in cartons, stacked 14 ft high 475 PS toy parts in cartons, stacked 15 ft high 180 PS insulation board, rigid foam, stacked 14 ft high ~ PVC bottles packed in Item 6 PP tubs packed in Item 6 .4# ~ 5 - - PP & PE film in rolls, stacked 14 ft high ~'~g 550 Methyl alcohol ... ~t,.~, " % Gasoline ~ 6 ~g ' " Kerosene Diesel oil P ~ a " ~ e 1~ Note: PE ffi Polyethylene q~

PS = Polystyrene PU ~'~,~ PV = Polyvinyl chloride

Table

~-~ -'::i..':~-.~?~.. where: ~::.-.'-x::: ".:.:.::-:.::Y:~'~!~::'!:: .,.:-::- maximum heat r e l e ~ r a t e (F ~ . g.:,:.Y:" Qm

q = heat release d e n s i t ~ e c / : ) ":.~..:-'~. A = floor area (ft 2) '%~!!~.,..

-~-.-.::.) ~..-".~ The following beat release rates per ~ . . ~ o r area are for fully involved combustibles, assuming 100

percent efficiency. The growth times she ffi are those required to exceed 1000 Btu/sec heat release rate for developing fires assuming 100 percent mbustion efficiency.

(PE = polyethylene; PS = polystyrene; PVC = polyvinyl chloride; PP = polypropylene; PU ffi polyurethane; FRP = fiberglass-reinforced polyester.)

Warehouse Materials

Wood pallets, stacked 1 1/2 ft high (6-12% moisture) Wood pallets, stacked 5 ft high (6-12% moisture) Wood pallets, stacked 10 ft high (6-12% moisture) Wood pallets, stacked 16 ft high (6-12% moisture) Mail bags, filled, stored 5 ft high Cartons, compartmented, stacked 15 ft high Paper, vertical rolls, stacked 20 ft high Cotton (also PE, PE/Cot, Acrylic/Nylon/PE),

garments in 12-ft high rack Cartons on pallets, rack storage, 15-30 ft high Paper products, densely packed in cartons, rack

storage, 20 ft high PE letter trays, filled, stacked 5 ft high on cart PE trash barrels in cartons stacked 15 ft high FRP shower stalls in cartons, stacked 15 ft high PE bottles packed in Item 6 PE bottles in cartons, stacked 15 ft high

Growth Heat Time Release (sec) Density (q)

150-310 110 90-190 330 80-110 600 75-105 900

190 35 6O 2OO

15-28 2O-42

40-280 470

190 750 55 250 85 110 85 550 75 170

Classification (s-slow)

(m-medium) (f-fast)

m-f f f f f

m-f m.-~

f

w

640

N F P A 92]8 - - MAY 2 0 0 0 R O P

I Wa.ho. . leon, ued?

Growth Time

PE pallets, stacked 3 ft high PE pallets, stacked 6-8 ft high PU mattress, single, horizontal PF insulation, board, rigid foam, stacked 15 ft high PSjars packed in Item 6 PS tubs nested in cartons, stacked 14 ft high PS toy parts in cartons, stacked 15 ft high PS insulation board, rigid, stacked 14 ft high PVC bottles packed in I tem 6 PP robs packed in Item 6 PP and PE film in rolls, stacked 14 ft high Distilled spirits in barrels, s tacked20 ft high Methyl alcohol Gasoline Kerosene Diesel Oil ÷

* Hre growth rate exceeds classification criteria. For SI Units: 1 ft = 0A~05 m.

B-5.S Other Useful Data. There are other d a ~ that are no t normalized that mkrht be useful in develooing the rate of h~ut release curve. Examnles are-included in the Tables B-5.3(a) thromrh B-5.$Ch),

Table B-5.S(a) Maximum Heat Release Retea from

(Stu/sec~ Medium wastebasket with milk cartons 100 Large barrel with milk canons 140 Upholstered chair with" polyurethane foam 550 Latex foam mattress (heat at room door) 1200

1~o 30.55 110

8 55 105 110

7 9

10 40

2.g.40 °

n ~ t

.

lio

~ 0

r;

Classification ( x i o w )

(m-medlum) ff-f.a)

f

f s

f f

Table i

Cigarette 1.1 g (not v " q ~ W puffed, laid on solid surface), bone dry, conditioned to 50% R,H. ,~ Methenamine pill, 0.15 g Match, wooden (laid on solid surface) Wood cribs, BS 5852

Par t2 No. 4 crib, 8.5 g No. 5 crib, 17 g No. 6 crib, 60 g No. 7 crib, 126g

Crumpled brown lunch bag, 6 g Crumpled wax paper, 4.5 g (tight) Crumpled wax paper, 4.5 g (loose) Folded double-sheet newspaper, 2"2 g (bottom ignition) Crumpled double-sheet newspaper, 22 g (top ignition) Crumpleddouble-sheet newspaper, 22 g (bottom ignition) Polyethylene waste-basket, 285 g, filled with 12 milk cartons Plastic trash bags, filled with cellulosic

0.2-14 k0e Time duration of significant flaming

b Total b u m time in excess of 1800 sec ¢ As measured on simulation burner d Measured fi'om 25 mm away e ResUlts vary greatly with packing density 1 in, = 25.4 mm 1 Btu/sec = 1.055 W 1 oz = 0.02835 kg = 28.$5 g

2 2 I Bm/f t ~sec = 11.$5 kW/m

Ilura Flame m ~

5

5 45 80

1 ~ 0

1200 gO

20-S0

.

14

(~90 g)

IOO0 1900 20OO 0400 1200 1800 550O 40O0 74O0

17,000 . 50,000 120,000

to ~50,000

190 200 190 ~ 0 8O 25 20

100 4O 20

200b

200b

550 200

Heat Flux

42

~5 4

18-20

15d 17d 20d

35c

641

N F P A 92B - - MAY 2 0 0 0 R O P

Existing Table B-6 remains unchanged. Renumbered as Table B-5-$(c).

Existing Table B-7 remains unchanged. Renumbered as Table B-5-B(d).

Table B-5-3(e) Effect of Fabric Type on Heat Release Rate in Table B-5-3(a) (within each i~roup all other construction features were kept constant) [3]

Full-Scale Peak q

Specimen (kW) Paddinlg Fabric Group 1

F24 700 cotton (750 g /m ~ FR PU foam F21 1970 polyolefin (560 g/m ~) FR PU foam

Group 2 F22 $70 cotton (750 g /m ~) cotton batting F23 700 polyolefin (560 g /m ~) cotton batting

G r ~ p 3 28 none FR PU foam 17 530 cotton (650 g /m ~) FR PU foam 21 900 cotton (110 g /m ~) FR PU foam 14 1020 polyolefin (650 g /m ~) FR PU foam 7, 19 1340 pol)'olefin (360 I~/m~/ FR PU foam 1 Ib/ft ~ = 48.83 g /m z 1 oz/fF = 305 g /m ~ ~ !~ 1 Btu/sec = 1.055 kW .¢~ '\'~

Table B-5.3(f) Effect of Padding Type on Maxim (within each trrouo all other construction

Rate i~"Table B-5.3(d) ~t constant) [3]

Peak q '¢~ ,x Specimen (kW~ ":~:",~%%~.~ a d d in g L , = : : . " : ~ ' . Fabric

Group 1 ~ . 4~.-"' F21 1970 F R ' ~ J ~ , ~ ~ $ ~ " polyolefin (560 g /m ~) F23 1990 N ~ ' ~ ~ ' + polyolefin (560 g /m ~)

Group [ ~".'::. F21 197~)ff ~=~'~:~i~i FR P~i~'o~i o..~n " polyolefin (560 g /m 2) F25 ~.~,~, ~" cotton' :~l ing polyolefin (560 g /m 2)

F24 70( ~ " 1 ~ ' foam cotton (750 g/m ~ ) F22 ~ - . ~ ~ . . . . c o t t o n batting cotton (750 g/m ~)

12, 27 .-::'!:: 1" ~, ~ ' . l~r~ PU foam polyolefin (360 g/m ~) 7, 19 ~ . . . _ 1540~ :U FR PU foam polyolefin ($60 g /m ~) 15 ': "::!%.-:-:-.:ii~.-.. 120 :~ i' neoprene foam polyolefin (360 g /m ~)

%'%6rou~1 %.:.~:...:.x. .,~ 20 % ~ . . f f NFR PU foam cotton (650 g/m ~) FR PU foam 17 "~..~:g~ cotton (650 g/m~!

22 .~-i~ ~: 3 neoprene foam cotton (650 ~[/m 1 lb/fi ~ = 48.83 g/m ~ 1 oz/fF = 305 g /m ~ 1 Btu/sec = 1.055 kW

Table B-5.B(g) Effect of Frame Material for Specimens with NFR PU Padding and Pol),olefin Fabrics [3]

Mass Peak q Specimen (k~) (kW) Frame F25 27.8 1990 wood FS0 25.2 1 0 6 0 polyurethane F29 14.0 1950 pol):prop]:lene 1 lb = 0.4556 kg 1 Btu/sec = 1.055 kW

642

N F P A 9 2 B ~ M A Y 2 0 0 0 R O P

Table B-5.3(h) Considera t ions for Selecting Heat Release Rates for Design

Cons tan t Heat Release Rate Fires T h e o b a l d

(industrial) 260 k W / m ~ Law

(offices) 290 k W / m ~ Hansell & Morgan

(hotel rooms) 249 k W / m ~ Variable Hea t Release Rate Fires NBSIR 88-3695 Fire Growth

Fuel Conf igura t ion Rate Compu te r Work Statiot~

free burn slow-fast c o m p a r t m e n t very slow

Shelf Storage free burn m e d i u m up to 200 sec,

fast after 200 sec Office Module very s low-medium

NISTIR 483 Peak Heat Release Fuel Commodity Rate (kW)

Compu te r Work Station 1000-1300 NRS Monograph 173

Fuel Commod i ty Peak Heat Release (kW) Chairs 80-2480 (<10, metal f rame) Loveseats 940-2890 (370, metal f rame) Sofa 3120

il-6 Calculated Fire Descriotion Based on Tested Propertie#,

B-6.1 Background. It is nossible to make general est imates of the rate of heat-release o f bu rn in~ materials based on the fire propert ies of tha t material . T he fire nrooer t ies involved can l~er d e t e r m i n e d bv small-scale tests. T he m o s t imoor tan t of these tests are calor imeter tests involving bo th oxv~en deple t ion calorinaetry and the applicat ion of extern-al hea t flUX to the sample while de t e rmin in~ t ime to ignit ion, rate of mass release, and rate of hea t

(approx. 26 Btu/sec-f t 2 ) (approx. 29 Btu/sec-f t 2 ) (approx. 25 Btu/sec-f t 2)

Products Usin

~-6.2 Discussion o f Measu red Prooer t ies . Table B-6.2 lists th¢ type o f fire nrooer t ies obtainab]e f rom the cone or F~gtory Mutual -ca lor imeters and similar ins t ruments .

In Table B-6.2, the rate of hea t release (RHR). mass loss. and t ime to imaition are func t ions of the externally aDolied iBcident vadi~mt hea t f lux imnosed on the tested samt~le. The purpose of the externally appl ied f lux is to s imulate the fire env i ronmen t s u r r o u n d i n g a b u r n i n g item. In general , it can be est imated tha t a f ree -burn ing fuel package (i.e.. one that bu rns in the ooen and is no t a~ected-bv enerav feedback f rom a ho t gas laver of a hea t source o ther than its own flame) is imnac ted by a f lux in the range of 95 k W/n l ~ 1;o 50 k W / m ~. If the fire is in a space and conditioq~ are aooroach ing flashover, this can increase to the range of 50 k~0)/m ~ to 75 k W / m ~. In fully developed. Dost-flashover fires. a range o f 75 k W / m 2 to over 100 kW/ 'm ~ can be exoected. The (ollowing is a discussion of the individual orooert ies measu red or derived and the usual fo rm used to repor t the

ta) Rate of Heat Release. Rate of hea t release is de t e rmined

L J 1 1 :..-. ~ . ~ 0 kW/m a . . . . 2S kW/m 2 ........ 50 ,#! ,

6~ 1200 /,,, !_. "~:" 1000 l ~ " ~

8oo ;

6oo ] | 400 ! ~ ~

- .

oi/J , . . 200 400 600 800 1000 1200 1400 1600

Time (sac)

Figure B-6-2(a) Typical ~raphic ou tou t o f cone calor imeter test, nroDertv test da ta as the basis of an analvtlcal evaluation of the rate of hea t release involved in the use o f a tested material. The approach out l ined in this sect ion is based on tha t pl 'esented bY Nelson and Forssell [19941.

Table B-6.2 Relation o f Calor lmeter-Measured Proner t ies to Fire Analwi~

Rate of hea t release H Mass loss H Time to ignit ion H Effective thermal oroDerties* Heat of combust ion*- Heat o f gasification* Critical imaition flux* Ignit ion temp.*

~ a m e lgatfiaa

xxx

xxx xxx xxx xxx

xxx

xxx xxx xxx xxx

H-Pronertv is a func t ion of the externally anol ied inc ident flux, * Derived Drooerties f rom calor imeter measu remen t s .

xxx xxx

xxx xxx

643

N F P A 92B ~ MAY 2 0 0 0 R O P

Often only the neak rate of heat release at a snecific flux is renorted. Table B-6.2(a) is an examole.

Material

PMMA

Pine

Sample A

Sample B

Sample C

Sample D

] 'able B-6.2¢a) Average Maximum Heat Release P-~tes tkW/m ~)

4.4

Orientation

Horizontal Vertical Horizontal Vertical Horizontal Vertical Horizontal Vertical Horizontal Vertical Horizontal Vertical

2.2 B t u / s / f t ~ Exposinl{ Flux

57 49 12 11 11 8 12 5.~

6.2

B t u / s / f t ~ Exposing Flux

79 63 21 15 18 11 15 18 19 15

11

6.6 B t u / s / f t 2 Exposing Flux

114 114 23 56 22 19 21 29 22 15

11

(b) Mass Loss Rate (m). Mass loss rate is determined by a load ~;~lJ: The methgd of reoortin~ is identical to that for rate 'of heat release. In the tvoical situation whene the material has a consistent h~gt of combustion, the curves for mass loss rate and rate of heat release are similar in shane.

(c) Time to Ignition (aA. Time to i maition is renorted for each individual test and am~l]'ed flux level conducted.

(d) Effective Thermal Inertia (kDc). Effective thermal inertia is a measurement of the heat rise resnonse of the tested material to the heat flux imDosed on the sample. It is derived at the time of ignition and is based on the ratio of the actual incident flux to the (ritical ignition flux and the time to ignition. A series of tests at different levels of applied flux is necessary, to derive the effective thermal inertia. Effective thermal inerda derived in this manner

(a) Thermally Thin Materials. Relative to i~nition f rom a constant incident hea t flux. hi. at the exnosed surface and with relatively

(B-5)

(g) Critical Ignition Flux (a ). Critical ignition flux is the rnh3imum level-of incident flux on the samnle needed to ignite the samole gfiven an unlimited time of anolicatJon. At jnciden-t flug leveis less than the critical ignition flux. ignition does not take paa~

(h) Ignition Temberature (T.I. Ignition temperature is the surface temperature of a satanic at which flame occurs. This is a samole material value that is indeDendent of the incident flux. It i~ clcrivable from the calorimeter tests, the LIFT aonaratus test. arid other tests. It is derived from the time to imaite in a given test. the aoolied flux in that test. and the effective thermal inertia of the samole. It is renorted at a single temoerature. If the te~, jIl¢lqclCS a nilot flame or snark, the reported tempera~ture is for niloted imaition: if there is no nilot nresent, the temoerature is-for autoignition. Most available data is for piloted ignition.

B-6,3 Ignition. Eouations for time tO ignition, t.~ are L, iven for v - . v . . ~ - v .

both thermally thin and thermally thick materials, as defined m B- 6.3(a) and (b). For materials of intermediate deoth, estimates for tig necessitate considerations beyond the scqpe of this presetliat, ion [Drvsdale 1985. Carslaw and Iaeger 19591.

The time to i~nition of a thgrmallv thin material subiected to i I lddent flux alaqve a critical incident flux is;

tig = p c l (T,,-To)

i I t

(b) Thennall~ Thick Materials. Relative to the taroe of iaxaition test described in B~6.3(a). a satanic of a inaterial of a-thickn¢$s, I, iS considered to be thermally tfaick if the increase in temperature of the unexnosed surface is relatively small comnared to that of the exnosed surface at t = tie. For examole, at t = ti~.

~ B T~ < 0.1(T.:atmi.~.. .~= 0.1¢T,;B Ta) (B-7)

Eouation B-7 can be used to show that a material is thermallv thick I'Carslaw and lae~er 19591 if:

1 > 2 ( t 10) ]/2 (B-8)

For example, according to eauat ion B-8. in the case of ~m ignidon test on a slaeet of m a o l e o r oal~ wood. if fig = 35 s is measurecl in a niloted itmition tesL then. if the samnle thickness is trreater than anoroximatelv 0.0042 m. the unexoosed surface of tl~e sample can be exoected to be relatively close to T_ at t = t, and the satanic i~ considered to be thermally thick.

Time to ignition of a thermally thick material subjected to incident flux above a critical incident flux is:

. . , , ] 2 IB-9)

644

N F P A 92B ~ MAY 2000 R O P

It should be no ted tha t a Darticular mater ia l is no t intyinsicaily thermal ly th in or thick (i.e.~ the character is t ic of be ing thermal ly thin or thick is no t a mater ia l characteris t ic or propertv) but also d e p e n ~ on the thickness of the par t icular sample (i,¢,, a par t icular mater ia l can be i m p l e m e n t e d in e i ther a thermal lv thick or thermai lv th in configurat iol l ) ,

(c) Pro~a~'ation Between Sebarate Fuel Packages. Where the concern is (or nronatrat ion I~etween individua] seoarated fuel nackages, incid~en(fiux can be calculated using t radi t ional radia t ion heat transfer orocedures [Tien et al ] 995L

The rate of radia t ion heat transfer f rom a f laming fuel nacka~e of total energy release rate. O. to a facing surface e l emen t of an exposed fuel vacka~re can be esfimatecl from:

qm~"= X~Q/4 Br 2 ¢B-lo~

T! qme = I n o d e n t flux on exposed fuel. Xr ffi Radiant fract ion of

exoosing fire. O ffi rate of hea t release of exoosing fire. and r = radial distance from center of exDosing fire to ex~nosed fuel.

B-6.4 Estimatin~ Rate of Heat Release. As discussed in B-6.2. tests have demons t r a t ed tha t the c n e r w feedback from a burn in~ fuel package ranges f rom anoroxima(e lv 25 k W / m ~ to 50 k W / m ~. For a reasonable conservative-analysis, it is r e c o m m e n d e d tha t test da ta develoned with an inc iden t flux of 50 k W / m ~ be used. For a first order anorox imat ion , it shou ld be assumed tha t all of the surfaces tha t can be s imul taneously involved in bu rn ing arc re leasing energy at a rate eoual to tha t d e t e r m i n e d by test ing the mater ia l in a fire oroner t ies c a lo r ime t e r with an inc iden t flux of ~iO k W / m ~ for a free- bu r n ing mater ia l and 75 k W / m 2 to 100 k W / m ~ for nost-flashover condit ions.

In making this estimate, i t is necessary to assume tha t all surfaces

Flame soread is the movemen t of the f lame f ron t across the surface of a mater ia l that is b u r n i n g (or exoosed to an igni t ion flame) where the exnosed surface is no t vet fully involvec1. Phvsicailv. f lame soread can be t rea ted as a succession of igni t ions resul t ing from the hea t ener~v o roduced by the bu rn ing nor t ion of a mater(al, its f lame, and anvot-her inc iden t hea t e n e r ~ imnosed unon the u n b u r n e d surface. Othglr sources of inc iden t enerav include ano t he r b u r n i n g object, h igh t emvera tu re trases t h a t c a n accumula te in the u o o e r oor t ion of an enclosed soace, and the radiant hea t sources used in a test aoDaratus such as the cone ca lor imeter or the Lib-q" mechanism. For analysis ournoses, f lame st)read can be divided into two categories, tha t which moves in the same d i rec t ion as the f lame ( concu r r en t or wind-aided flame snread) and that which moves in any o ther d i rec t ion (lateral or onnosed flan1~ soreadL ConcurreBt f lame spread is assisted bv the inc ident hea~ flux from the f lame to un ign i t ed oor t ions of the bu rn ing material . Lateral f lame soread i s no t so assisted and tends to be much slower in nrotrression unless an external source of hea t flux is present . Concur ren t flam¢ snread can be exoressed as

(~i tt L v =

laains unchanged .

~a ins unchanged .

Append ix E.

he (B-11~

The resul t ing mass loss rate is then mul t in l ied bv the derived effective heat o f combust ion and the bu rn ing a rea exoosed to the inc iden t flux to n roduce the es t imated rate of hea t release as

Q/' = th"hc A. f~-tz)

B-6.5 Flame Svread. If it is des i red to predic t the ~rowth of fire ~s v

it propagates over combust ib le surfaces, it is necessary to est imate flame soread. The comouta t ion of f lame snread rates is an emerg ing technology still in an embryonic stage. Predict ions should b-e cons idered as order of magn i tude estimates.

~.y Appendix E Example Problems Illustrating the Use of the Equations in NFPA 92B

This appendix is not apart of the recommendations of this NFPA document but is included for informational purposes only.

Given: Atr ium with un i fo rm rec tangula r cross sect ional area-

Height 120 ft Area 20,000 sq tt A / H A 2 1.4 Design Fire (steady state) 5000 B tu / s ec Highes t Walkin~ Surface 94 ft

1. De te rmine the t ime when the first indica t ion of smoke is 6 feet above the highest walking surfacer,

a. Use Eaua t ion 9

z/H=O.67-O.281n~Q'/3/H4/3)/(4/H 2 )]

Z 100ft H 120ft Q 5000 B tu / sec

~ (1 /3) 17.1 (4/3) 591.9

A / H ^ 2 1.4

0.83 = 0.67 - 0 .281n[(17.1t /591.9) / (1 .4) l

0.16 = -0.281a[0.03t/1.391

0.10 = -0.281n [0.02tl

645

N F P A 9 2 B ~ M A Y 2 0 0 0 R O P

-0.57 = In[0.02tl

0,56 = 0.02t

t = 28 seconds

b. Use the mass flow me thod , based on Eouat ion 14.

Two calculat ion m e t h o d s will be used. T he first calculation will a s sume a smoke densi ty of 0.075 Ib/ft^3. This is eouivalent to smoke at a t empera tu re of 70 17. T h e second calculation assumes the l~yer t empera tu re is equal to the average p l u m e tep~perature at the he igh t of the smoke laver interface. In both cases, no hea t loss f rom the smoke laver to the a t r ium boundar ies is assumed. A t ime interval of 1 second is chosen for each case.

iA Calfulaf ion ] - No smoke d¢osity corr¢ction

Steo 1 Calculate mass flow (Ib/sec~ at z = H usin~ Fxauation 14.

Step 2 Convert mass flow to vo l ume flow {ft^3/sec] us ing Eauat ion 16. a s sumi ne smoke t emoera tu re is 70 F.

Sten 3 Assume the smoke volume p roduced in the selected t ime ipterval i8 instantly and un i formly distr ibuted over the a t r ium area. De te rmine the deo th of the smoke laver, clz frO. deposi ted dv r iog l;he selected t ime oeriod.

Steo 4 Calculate the new smoke laver interface he igh t (fO

Repea t steps until the smoke laver interface reaches the des ign

2. De te rmine the volumetr ic exhaus t rate requ i red to keep smoke 5 ft above the h ighes t walking level in the a t r ium, i.e., n in th floor balcony. Consider the fire to be located in the cen te r o f t h e floor o f the a t r ium.

With the fire located in the center o f the atr ium, an axisymmetr ic p lume is expected. First, Equat ion (13) of 3-7.1.2(a) m u s t be applied to de te rmine the f lame height .

Given: Q~ = 3500 B tu / s ec z~ = 0.SSS Q~/5 z 1 0.533 (3500) ~/5 z 1 = 13~9 ft

With the design interface of the smoke layer at 85 ft above floor level, the f lame he igh t is less than the design smoke layer height . Thus , us ing Equat ion (14) of 3-7.1.2(b) to de t e rmine the smoke product ion rate at the he igh t o f the smoke layer interface:

z = 8 5 ft m = 0.022 O~ l/s z ~/s + 0.0042 Q¢ m = 0.022 (gS00) ~/~ (85) ~/s + 0.0042 (~500) m = 564 lb / sec

If the smoke exhaus t rate is equal to the smoke produc t ion rate, the smoke layer de~ th will be stabilized at the design height . Thus , conver t ing the ~ w rate to a volumetr ic flow rate us ing Equat ion ( 1 6 ) . ~ m 2 2 ~ . 5 :

V ~ m / r

V ='~.]~f~)r~'ec or 451,260 scfm

Table ~1 o f Values illustrates the calculation technioue .

Time (sec~

1

10

11

.1.£

14

1...55

16

17

18

2o

Table E-I ~ d Valiies

%i!,,

~ . . J 18.0 :..:: ~:~ 7.9_72 f ?';:-'-:;:;:;:;:;:;.,

~-.:,:.:~,! 946

938

114.9 93o

922

11s.1

112.5

111.9

111.3

110.7

109.5

109.0

108.4

107.9

914

9o6

898

890

882

875

867

86o

852

845

838

Vol fft^3/sec~

229_~

12847

12D2

12619

125O8

12397

12288

12181

12O74

11866

11562

Ll_lfifi

L1__220

11174

646

N F P A 9 2 B ~ M A Y 2 0 0 0 R O P

T i m e {sec)

T a b l e E-1 S a m p l e C a l c u l a t e d V a l u e s ( c o n t i n u e d )

z f f O Mass [.~'~Y~

106.7 824

8_1_7.

104.0 26

27

810

8O4

797

790 104.1

Vol ( f t ^ 3 / s e c ~

11O80

10987

1O895

108O4

1O715

~ 784 10451

77'7 lOZ.O

771

7o5

759

752

746 .~

lol .5

lOl.O

loo.5

i12

3o

32

_3_3_

~4

lO~66

10115

. ~ & , . lOO~Z

f f f i " . . . . . .

.~::".':-"~- "~:~.. -,'-'i~'#~-. ~:..~ ~:-"~"~:'~:?~,..::. '~ '-.'~&~.~.~"

3. De te rmine if the p l u m e will contact all of the walls prior to ~ . ~ : , ..~ii~:'J':~:. reaching the design he igh t no ted in #4 (5 ft above the h ighes t w ' ' ~ i ~ .... walking level). . "" % ~ ~ i ~ . . . . . .

• .+:--- IZ = ~2 2 ~.~..: .~.-" The above calculation in #4 assumes that the smoke p l ume nas ~.:'~!:..::.. ft. l ~ a n - ~ ~ no t widened to contact the walls of the a t r ium prior to reaching the " ~ . . . - , - t v ~ - : : - des ign interface height . This calculation serves as a check. '!~'~: . # ~ _ ~ " : ' [ ~ H tT T ~ / t T + 460 ~1 a/~

Using Equat ion (23) f r om Section 3-8 with an interface he igh t of ":.~i, ..:.:~'~, - ] ~ t ~ k, , t , " ou,,~ L . . . . . t ,~ , 85 ft (z = 85 ft) .,.::#.~g~,~.-'i~::, ":¥;:" V = Z~'~ l t~Z:Z) ( o ) t t v v v - / v J / ~ l v o v + 460)] ~/~

' .?:-::" "::~!!!!!: :~i, V = '~Zo I t / m l n

d -_- 0.5zf~5, ,:::..'.:~iiiiiiii~::. :~!~:.'.i??~:.:.'.-.:.. :'::.:.~: .~:'" (b) For a fire in the a t r ium, de te rmine the opposed airflow _ -.7." _ ~ " " "-';i.":':-'.g.-'::-..,-::';:: . . . . . ~: :oi~i" r eau i red to restrict smoke spread into the t e n a n t space. ct = 4X.b tt "~:~.-'..:-:'..'~" "~:'.-"~::" - -

Thus , the smoke does ~aot contact the w a l l s . , 6 ~ " : t J ~ : ' ~ p~'.$.@::;;~ Given: reaching the des ign interface height . ~.:>...-::':?' "'"~:::'.~::~.:..:~i:: ~ii~:::'~* ~=- 5 ̂ tmO l~tu/s- " ec

z 9 0 f t 4. De te rmine the t empera tu re of the sr~'olt:~i-"::~..e..r after (~" V = 17 [O/z] ~/s c tua t lon '.:m:::::. :.-~: 7 [ 5 01 ~/s

The quality of the smoke conta ined in the s m o k ~ migh t be ~ = " P impor tan t in the context of tenability or damagea .hi'fry studies. Applying Table 3-5:

Given:

O~ = 3500 B tu / sec r = 0.075 lb / f t 3 c = 0.24 Btu/ lb-°F V = 7521 f tS/sec AT = Q J ( p c V )

AT = 3500/[ (0.075) (0.24) (7521) ]

AT = 26°F

5 . O n the t en th floor, a 10 f t wide × 6 ft h igh open ing is des i red f rom the t enan t space into the a t r ium.

(a) For a fire in the t enan t space, de te rmine the opposed airflow requi red to conta in smoke in the t enan t space (assume fire t empera tu re is 1000°F). Using Equat ion (24), 3-10.l:

A p p e n d i x F R e f e r e n c e d P u b l i c a t i o n s

F-1 The following d o c u m e n t s or por t ions t he reo f are re fe renced within this guide for informat ional purposes only and are thus n o t cons idered part of its r e commenda t ions . The edi t ion indicated here for each reference is the cur ren t edit ion as of the date o f the NFPA issuance of this guide.

17-1.1 N F P A Publ i ca t ions . National Fire Protection.Associat ion, 1 Bat terymarch Park, P.O. Box 9101, Quincy, MA 02269-9101.

NFPA 13, Standard for the Installation of Sprinkler Systems, 1996 edition.

NFPA 72, National Fire Alarm Code ®, 1996 edition.

F-I.2 O t h e r Publ i ca t ions .

647

N F P A 92B - - MAY 2 0 0 0 R O P

F-1.2.1 ASHRAE Publication. American Society of Heating, Refrigerating and Air Conditioning Engineers, Inc., 1791 Tullie Circle, N.E., Atlanta, GA 30329-2305.

ASHRAE Handbook of Fundamentals, 1997.

F-1.2.2 ASTM Publications. American Society for Testing and Materials, 100 Barr Harbor Drive, West Conshohocken, PA 19428- 2959.

ASTM E 1321, Standard Test Method for Determining Material Ignition and Flame Spread Properties, 1997.

ASTM E 1354, Standard Test Method For Heat and Visible Smoke Release Rates for Materials and Products Using an Oxygen Consumption Calorimeter, 1997.

F-I.3 A d d i t i o n a l R e f e r e n c e s .

1. Morgan, H.P., Smoke Control Methods in Enclosed Shopping Complexes of One or More Storeys: A Design Summary, BRE, 1~79.

2. NFPA 204M, Guide for Smoke and Heat Venting, National Fire Protection Association, Quincy, MA, 1991.

3. Babrauskas, V., and Krasny, J., Fire Behavior of Upholstered Furniture, NBS Monograph 173, National Bureau of Standards (now National Institute of Standards and Technology), November 1985.

4. Morgan, H.P., and Hansell, G.O., Fire Sizes and Sprinkler Effectiveness in Offices--Implications for Smoke Control Design, Fire

tyJournal, Vol 8, No. 3, 1985, p. 187-198.

5. Law, M., Air Supported Structures:. Fire and Smoke Hazards, Fire Prevention, Vol 148, 1982.

6. Fang, J.B., and Breese, J.N., Fire Development in Residential Basement Rooms, NBSIR 80-2120, National Bureau of Standards, October 1980.

7. Heskestad, G., and Delichatsios, M.A., Environments Detectors--Phase 1 Effect of Fire Size, Ceiling Height and M Volume I - Measurements (NBS-GCR-77-86), Volume I~ NBS-GCR-77-95), National Bureau of Standards ( n q ~ Institute of Standards and Technology), Galthersbufg, Ig

8. Heskestad, G., and Bill, R.G., Jr., M, Responsiveness of Automatic Sprinklers, Fir~ Proceedings of the Second International Publishing Corporation, New York, 198~

9. Heskestad, G., and Delichatsios, M.A., The'~ Flow in Fir~ Seventeenth Symposium (Internatiot Combustion, The Combustion Institute, Pittsbur p.ll13.

1979,

10. Nowler, S.P., Enclosure Environment Characterization Testing for the Base Line Validation of Computer Fire Simulation Codes, NUREG/CR-4681, SAND 86-1296, Sandia National Laboratories, March 1987.

11. Morton, B.IL, Taylor, Sir Geoffrey, and Turner, J.S., Turbulent Gravitational Convection from Maintained and Instantaneous Sources, Proc. Royal Society A, 234, 1-23, 1956.

12. Mulholland, G., Handa, T., Sugawa, O., and Yamamoto, H., Smoke Filling in an Enclosure, Paper 81-HT-8 ,The American Society of Mechanical Engineers (1981).

13. Cooper, L.Y., Harkleroad, M., Quintiere, J., and Rinkinen, W., An Experimental Study of Upper Hot Layer Stratification in Full,Scale Multiroom Fire Scenarios, Paper 81-HT-9, The American Society of Mechanical Engineers (1981).

14. Hagglund, B., Jansson, R., and Nireus, K., Smoke Filling Experiments in a 6 x 6 x 6 Meter Enclosure, FOA Rapport C20585-06, Forsavrets Forskningsanstalt, Sweden, September 1985.

15. Heskestad, G., Engineering Relations for Fire Plumes, SFPE TR 82-8, Boston, Society of Fire Protection Engineers, 1982.

16. Law, M., "A Note on Smoke Plumes from Fires in Multi-Level Shopping Malls," Fire Safety Journal, 10, (1986), p. 197.

17. Morgan, H.P., and Marshall, N.R., Smoke Control Measures in a Covered Two-Story Shopping Malls Having Balconies and Pedestrian Walkways, BRE CP 11/79, Borehamwood, 1979.

18. Thomas, P.H., "On the Upward Movement of Smoke and Related Shopping Mall Problems," Fire Safety Journal, 12, (1987), p.191.

19. Morgan, H.P., "Comments on A Note on Smoke Plumes from Fires in Multi-Level Shopping Malls," Fire Safe~yJournal, 12, 1987, p.83.

20. Modal<, A.T., and Alpert, R.L., Influence of Enclosures on Fire Growth - Volume L" Guide to Test Data, FMRC 0AOR2.BU-8, Factory Mutual Research, Norwood, MA, 1978.

21. Tewarson A., Generation of Heat and Choniical Compounds in Fires, The SFPE Handbook of Fire Protection Engineering, National Fire Protection Association, Quincy, MA, 1988.

22. Mudan, K.S., and Croce, P.A., Fire Hazard Calculations for Large Open H3drocarbon Fires, The SFPE Handbook of Fire Protection EngineeLing, National Fire Protection Association, Quincy, MA, 1 9 ~

. . ~ " 23. Heskes~ ,~ ~ Inflow of Air Required at Wall and Ceiling

Apertures ~.~% . ~ R ~ a p e of Fire Smoke, FMRC J.I. 0Q4E4.RU, Factory ..~ . e ~ Cc)rporation, July 1989.

24" "" ]oe,.~,~L., ~.'dEJ., Evaluating Unsprinklered Fire . ~ d s , ~ R C No. 01~.20 , Norwood, MA, Factory Mutual ~ h ~ o r a t i o n , August 1982, 46 and 48. ~ ::" PP"

25. A ~ Handbook of Fundamentals, American Society of Heating, ~ .~ ' r a t ing , and Air Conditioning Engineers, Atlanta,

..stad, G., and Delichatsios, M.A., "Update: The Initial Flow in Fire," Fire Safety Journal, Vol. 15, 1989, p. 471.

" Newman, J.S., and Hill, J.P., "Assessment of Exposure Fire d to Cable Trays," NP-1675 Research Project 1165-1-1,

Prepared for Electric Power Research Institute by Factory Mutual Research Corporation, January 1981.

28. Sako, S., and Hasemi, Y., "Response Time of Automatic Sprinklers Below a Confined Ceiling," Fire Safety &ience -Proceedings of the Second International S3mposiura, Hemisphere Publishing Corporation, New York, 1989, p. 613.

29. Heskestad, G., and Hill, J.P., "Experimental Fire in Multiroom/Corridor Enclosures," FMRC J.I. 0J2N8.RU, Factory Mutual Research Corporation, October 1985.

30. Mulhoiland, G., Handa, T., Sugawa, O., and Yamamoto, H., "Smoke Filling in an Enclosure," Fire Science and Technclog 3 (1) p. 1, 1981.

31. Hinkley, P.L., Hansell, G.O., Marshall, N.IL, and Harrison, IL, "Experiments at the Multifuncdoneel Training Centrum, Ghent, on the Interaction Between Sprinklers and Smoke Venting," Building Research Establishment Report, 1992.

32. Yamaha, T., and Tanaka, T., "Smoke Control in Large Scale Spaces," Fire Science and Technology, Vol. 5, p. 41, 1985.

33. Walton, W.D., and Notorianni, "A Comparison of Ceiling Jet Temperatures Measured in an Aircraft Hangar Test Fire with Temperatures Predicted by the DETECT and LAVENT Computer Models," NISTIR (Draft), Building and Fire Research Laboratory, National Institute of Standards and Technology, August 1992.

34. Purser, D.A., "Toxicity Assessment of Combustion Products," SFPE Handbook of Fire Protection Engineerin~ Quincy: NFPA, 1988.

35. Quintiere, J.G., Fire Safety Journal, Vol. 15, 1989.

36. Heskestad, G., "Determination of Gas Venting Geometry and Capacity of Air Pollution Control System at Factory Mutual

648

N F P A 92B - - MAY 2 0 0 0 R O P

Research Center," FMRC Ser. No. 20581, Fire Mutual Research Corp., Norwood, MA, November 1972.

37. Quintiere, J.G.,Mc(~ffi'ey, B.J., and Kashwagi, T., Combustion Science and Technology, Vol. 18, 1978.

38. Steckler, K.D., Baum, HAL, and Quintlere, J.G., 21st Symposium (Int.) on Comb~stion, 1986, pp. 145-149.

39. DiNenno, P.J. edt. SFPE Handbook of Fire Protection Engineering, National Fire Protection Association, Quincy, MA 02269, 1988.

40. Drysdale, D.D., "An Introduction of Fire Dynamics," John Wiley & Sons, New York, 1985.

41. Welty, J.R., Wicks, C.E., Wilson, R.E., "Fundamentals of Momentum, Heat and Mass Transfer, 'John Wiley & Sons, New York, 1976.

42. Peacock, R.D., Davis, S., Babrauskas, V., "Data for Room Fire Model Comparisons," Journal of the National Institute of Standards and Technology, Vol. 96, (4) July 1991.

43. Soderbom, J., "Smoke Spread Experiments in Large Rooms. Experimental Results and Numerical Simulations," Statens Provningsanstalt, SR Report 1992:52, Swedish National Testing and Research Institute, Boras, Sweden, 1992.

44. Emmons, H., "The Use of Fire Test Data in Fire Models," The Home Fire Project Technical Report No. 78, Harvard University, Division of Applied Sciences, February 1989.

45. Thomas, P.H., Heselden, J.M., and Law, M., =Fully-Developed Compartment Fires - Two Kinds of Behavior," Fire Research Technical Paper No. 18.

46. Gottuk, D., Roby, R., and Eglo, C., "A Study of Carbon Monoxide and Smoke Yields from Compartment Fires with External Burning; 24th Symposium (International) on Combustion," The Combustion Institute, 1992, pp. 1729-1735.

47. Bullen, M.C., and Thomas, P.A, "Compartment Non-Cellulosic Fuels; 17th Symposium (International Combustion," The Combustion Institute, 1979, pp. ~.

48. Mowrer, F.W., and Williamson, ILB., "Estir Temperatures from Fires along Walls and in Technology, 23,2, May 1987, pp. 133-145. .~"

49. Heskestad, G., "Letter to the May 1991, pp. 174-185.

2,

50. Yarnana, T., and Tanaka, T. "Smoke Con Spaces (Part 2: Smoke Control in Large Scale and Technology, Vol. 5 Nc. 1, 1985, pp. 41-54.

~ e ' Scale Fire Science

51. Personal Communication, G.D. Lougheed, National Research Council of Canada, March 20, 1991.

52. Nowlen, S.P., "Enclosure Environment Characterization Testing for the Base Line Validation of Computer Fire Simulation Codes," NUREG/CR-4681, SAND 86-1296, Sandia National Laboratories, March 1987.

53. Mulholland, G., Handa, T., Sugawa, O., and Yamamoto, H., "Smoke Filling in an Enclosure," Paper 81-HT-8, ASME, 1981.

54. Cooper, L.Y., Harkelroad, M., Quintiere, J. and Rinkinen, W., "An Experimental Study of Upper Hot Layer Stratification in Full- Scale Multiroom Fire Scenarios," Paper 81-HT-9, ASME, 1981.

55. Hagglund, B., Jannson, IL and Nireus, K., "Smoke Filling Experiments in a 6x6x6 Meter Enclosure," FOA Rapport C20585-06, Forsvarets Forkingsanstall, Sweden, September 1985.

56. Milke, J.A., and Mowrer, F.W., "An Algorithm for the Design Analysis of Atrium Smoke Management Systems," FP95-04, Department of Fire Protection Engineering, University of Maryland at College Park, May 1995.

57. Zukoski, E.E., Kubota, T., and Getegen, B. 1980/1981, "Entrainment in Fire Plumes," lqre Safety Journal, Vol. 3, 1980/1981, pp. 10%121.

58. Turner, J.s., Effects in Fluids, Cambridge University Press, Cambridge, Bg~3ancY

59. Spratt, D., and Heselden, A.J.M., "Efficient Extraction of Smoke from a Thin Layer Under a Ceiling," UKJoint Fire Research Organization Fire Research Note No. 1001, 1994.

60. Hansell, G.O., Morgan, H., and Marshall, N.R., "Smoke Flow Experiments in a Model Atrium," BRE Occasional Paper, July 1993.

61. Mulholland, G., Section 1/Chapter 25 "Smoke Production and Properties," SFPE Handbook of Fire Protection Engineering, National Fire Protection Association, Quincy, MA, 1988.

62. Hincklev. P.L. 1995. Smoke and Heat Venting. in SFPE Handbook o(Fire Protection Engineering. National Fire Protection Association. Ouincv. M A v n 3-160-3-173. -

63. CIBSE.Relationshins for smoke control calculations. T¢¢hnical Memoranda q'M19. The Chartered Institution fo

Smoke Control in Atrium Buildings. BR 258. Borehamwood. UK: Building Research Establishment.

Bennetts. I.D.. Culton. M.. Dickerson. M.L., Lewins. 1R.. Poh, K,W,, Poon. S.L.. Ralnh. R.. Lee. A.C.. Beever. P.F.. Conner. R.I.. Hangar. P.I.. 1V(oore. LP.. Ramsay. G.C. and Timms. C~.R. 1997, Simulated ShoDving Centre FireTests. BHPR/SM/R.G/062. Broken Hill ProDrietarv Company Limted. Australia_

Lougheed. G.D.. Mawhinney. I.R. and Q'NeUl. I. 1994. Full-scale Fire-Tests and the Develovment of Design Criteria for Sprinkler Protection of Mobile Shelving Units. Fire T¢chnolo~., ~'olum¢ $0,

Ghosh. B.K. 1997. Fires in Real Scenarios. International Svmvosium on Fire Science and Technolot, v. ISFST 1997. Seoul. Korea. t~. 459-447. -"

Heskestad. G.. Sprinkler/Hot Laver Interaction. NIST-GCR-91-590. National Institute of Standards. Gaithersburg. MD. 1991,

v

Lougheed. G.D. and HadiisoDhocleous. G.V. 1997. Invest~atio~ of AOium Smoke Exhaust Effectiveness. ASHRAE Transactions. Vol. 103. no 1-15.

Lougheed. G.D. HadiisoDocleous. G.V. McCartney. C. and Taber, B.C. 1999. Lar~e-Scale physical Model Studies For An Atrium Smoke Exhaust-Svstem. ASHRAE Transactions. Vol.104. Do.

Hadiisoobocleous. G.V.. Lougheed. G.D. and Can. S, 1999. Numerical Stud~ of the Effectiveness of Atrium Smoke Exhaust Systems~ ASHRAE Tran-sactions.'Vol. 104, pp,

649