ML20024H174

LER 90-002-02:on 900216,determined That Fire Dampers May Not Be Operable Under Expected Ventilation Flow Conditions Due to Design Error.Design Basis Documents updated.W/910517 Ltr

ML20024H174
Person / Time
Site:	Crystal River
Issue date:	05/17/1991
From:	Bandhauer W, Boldt G FLORIDA POWER CORP.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
3F0591-10, 3F591-10, LER-90-002, LER-90-2, NUDOCS 9105290201
Download: ML20024H174 (6)

Text

--

.~

4

[ .

. f * ' ' e ;.

T .;

[,',

V: -

y

+

Florida Power c o neo n aiinN Crystal River Unit 3 Docket No. 50-302 May 17, 1991 3F0591-10 U. S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, D. C. 20555

Subject:

Licensee Event Report (LER) 90-02-02

Dear Sir:

Enclosed is Licensee Event Report (LER) 90-02-02 which is submitted in accordance witii 10 CFR 50.73.

This supplement includes additional information and provides the status of the subject item.

Sincerely, G. L. Boldt Vice President Nuclear Production WLR: mag Enclosure xt: Regional Administrator, Region II NRR Project Manager Senior Resident Inspector 9105290201 910517

[ [k f'DR A Florida Progress Company

.. ADOCV 05000302 PDR

NRC ONM 366 U.S. NUCL E AR RE GUL ATOR Y COMMISSION APPHOvtD OMe NO 3160 0104 j t u rist[$ 4!)O'92 I ttTIVATI D BURDth P(H Rf $PON'il TO COMPLY WT*4 THi$

"^' ' * '

LICENSEE EVENT REPORT (LER) EOv"i . Ts' Ode,'d'f*Jt,is'UM AO rW.L .S"!!!"S O.L"'lTn? c"cJJs"'Ni,R4",# e fANd ""

T* E P AP l HWOR A Rf DUCTION PROJECT 0 150 0104: O* f ICE OF MANAOf ME NT AND BUDGE T W ASHING TON. DC 20's0')

f aCILITY N AME HI ' DOCEIT NUMBE R (21 ' Abt $h CRYSTAL RIVER UNIT 3 0 ls 10 l0 l 0 [3 1 0l2 1 lorl0 15 tii t t <*>

FIRE DAMPERS MAY NOT CLOSE UNDER VENTILATION FLOW CONDITIONS DUE TO FAILURE TO CONSIDER FLOW CONDITIONS IN ORIGINAL DESIGN CRITERIA PER NRC IEN 89 52 EVENT DAf f tSi Lt R NUMetR 16 R E POR T D A f t t h OTHE R 8 ACILITit $ thv0LVED iti Dev ttAR vtAR N DAV ' A D lit V N Awf 5 DOCK E T NJVBt R:5!

MON T +4 [ h* ' [" MON T +4 TEAR N/A 0 l5l0 j o j o; i l

~ -

0l2 1l6 9 0 9l 0 0l0l2 0l 2 0l5 1l 7 9l1 N/A 0 [S[0 l0 1 0; } l opt R M M THl$ REPORT l$ Su6MfittO PURSUANT TO THE Rt OUintMENTE 0910 C0 R { !Oed on. or more ** the teacente: 018 MOOT m 5 20 402i , 20 dosi,3 .0 ,3u2 n# 73,,%

20 406teH1HJ to 341eHit to 736en2H=1 73 71 ts)

H01 0 1 0l0 ro easi.Huw 50 ao.H2,

$0 re.u2H..a

_ ovHi

,, R;so.., Agge, 20 400 ta H1 H ed 50 73teR2Hel 60 73*a42Hvo6H Al Joe A '

20 406tal0 HM X SO 735at'2Hal 50 73.eli2Hvi4HBI {

20 4C1isH1 Hit 50.7 3ia li2 Hm) 60 7Aad2Hss I LICEN54 E CONT ACT FOR THIS Lim his NAME TE LtPHONE NUMet R W. K. BANDHAUER, NUCLEAR OPERATIONS SUPERINTENDENT 910 14 7 l 9151 -I 61418 l6 COMPLETt DNE LINE FOR E ACH COMPONENT F AILURE DEECRISED tN THit REPORT H3i S Y1' EV COMPONENT " 'y C # ** 'C HE*ORTAgLt CALSE TO P S' CAU56 5v 57E Y CoveJNENT ER B K; P D, M; Pj A.1 ;2 ; 4 NO , i ; ;  ; ; ;

I 1 l I i l I I l I ! l l l EUP'LEMENT AL REPORT E XPECTE D Het YONTM CAY VEAR I Sv9M $5=ON

• ^"

} vis no ,es mee emcreo saev:ss:cs ces> & No l } l a,,,nAC,-.,,.,en,-......,.-,..-...-.,-,..

On February 16,1990, at 1132, Crystal River Unit 3 determined that fire dampers may not be operable under expected ventilation flow conditions due to a design error. This conclusion was based on the results of testing and evaluation of plant fire dampers and ventilation flows performed as a result of Information Notice 89-52. Roving fire watch patrol routes were reviewed and revised to assure 100 percent coverage of the fire areas. Continuous fire watches were posted in areas where fire detectors were inoperable. The root cause of this event is the failure of the original design criteria to address the need to close the dampers under ventilation flow conditions. This condition had been identified in 1985 by an internal contractor evaluation but, due to personnel error,.

had not been pursued and resolved. Dampers which are installed in ,

locations with excessive ventilation flows will either be modified to assure closure under design air flow conditions or fans will be turned off during a fire to allow closure of the dampers under reduced air flow conditions. Design basis documents have been updated to reflect the need for fire dampers to close under ventilation flow conditions.

NRC Form 384 i64%

. _ . _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ - _ ~ - .,

~a;, rom M. u . ~un. . out At o , co-.uo.

F EPiHI $ 4130'*J2 LlCENSEE EVENT REPORT (LER) ' 5lC^)),%*JW;[?, T'," "'27d!n /%if7J ' ,*/,"J",l$

TEXT CONTINUAT10N ";"4N,'5."',i'Zij "OM'y^,'L'? '",5 "'$',"f 5

! O f"A','l = J "",O S J.^r c i n % ?s W a 0 ?Ja Of MAN Abt ME NT ANO BUDGE f W ASHING 10N DC 20503 FACILilV hAME tli Docet t i NUMet R (D LE R NUM912163 PAGE131 uaa

" dm W,72 CRYSTAL RIVER UNIT 3 0F 0 l5 l0 l0 l0 l 3l0 l2 9 l0 0 l 0l 2 -

0l2 0 l2 0l 5 IEXT l# rnure spece e regiss( see e.Ateons/ MC form MA's)(1h EVERLRESCRIPTIONJ On February 16,1990, at 1132, Crystal River Ur.ii 3 (CR-3) determined that fire dampers [KP,DMP] may not be operable under expected ventilation flow conditions due to a design error. Several curtain-type fire dampers may not fully close with ventilation air flow.

On August 30, 1989, Florida Power Corporation (FPC) evaluated NRC Information Notice 89-52, " Potential Fire Damper Operational Problems", and determined the concerns identified by this notice may apply to CR-3. On November 10, 1989, FPC completed a review of damper flow rates and identified a representative sample of single section dampers corresponding to approximately 10% of the dampers to drop test under actual flow conditions. On January 26, 1990, CR-3 began drop testing the selected dampers.

On February 6,1990, when three out of five dampers tested failed, an action p' v was developed and a Nonconforming Operations Report was initiated. At this time, CR-3 was in MODE 1(POWER OPERATION), 97% power. The action plan included:

i. Perform a root cause evaluation of the fire damper failures,

11. Assure the hourly fire watch patrol route included all areas associated with the Technical Specification required fire dampers.

iii. Assure a continuous fire watch is posted in areas with inoperable fire detectors (10,28].

iv. Begin repair of the failed fire dampers.

v. Continue testing the selected dampers.

At 1132, on February 16, 1990, the root cause evaluation concluded the damper f ailures were caused by design deficiency. At the time, CR-3 was in MODE 5(COLD SHUIDOWN) with Reactor Coolant System temperature 95 degrees and at atmospheric pressure. CR-3 had been shutdown since February 12 to repair RCV-8, Pressurizer Code Safety / Relief valve [AB,RV]. At 1235, the NRC Operations Center was notified of this event per 10CFR50.72(a)(2)(i). This written report is being made per the requirements of 10CFR50.73(a)(2)(ii)(B) for operation outside the plant design basis.

The fire dampers are designed with several interlocking slats which are retracted in a configuration similar to a raised venetian blind. The fire dampers are actuated by fusible links. When the damper is released, the damper is pulled into position by retract ing springs and may be assisted by gravity. As the damper attempts to close, the increasing air velocity induced pressure can lock up the dampers and prevent them from fully closing. For some dampers, this may cause the closing spring to break. The fire dampers affected are various sizes and manufactured by Air Balance, Inc., Model Nos. N319ALV (vertical) and N319ALH (horizontal), three hour UL-rated.

NRC Poem MA (689!

. _ - _ - _ _ _ . - - - - - - _ _ - _ ~ ~ - - - . . ----~----------.u-- - - - - - - - - - . a.>- - - - - - - . s___~

hk OmW 3esha og Nwt g Am eg outoten, couwikupN

. 8khet$ d'.lWtJ

. LICENSEE EVdNT REPORT (LER) '"l"'l,'l;n*JtM,['," ,"'Oh,";gT,tl ,*'" ',0 TEXT CONTINUATION 'q",','n',,';',Vnd n@ i" ',"^,L'P U;Q'n,"O

~'P'h.'/M'm"#,"p,

,, a.Autuiu sho u.ait napuunto *,',t;=n',

c.c A,.w t O','n' 8 00 411Y h&W4 f t s D(K a ti huest m g, LI A NUU$f R tea F AGE O' CRYSTAL RIVIR UNIT 3 0 l$ l0 j o j o l 3l)}2 9jo -

u p l2 -

ol2 0l3 of 0 j$

u n ~. .. . .~ .c ,- - ,n a, in June 1985, during an evaluation of CR-3 fire protection features by an fPC cor, tractor, the need to verify the ability of dampers, especially multi-section dampers, to close under ventilation flow conditions was identified. A summary of maximum acceptable velocit es through Air Balance type 319 fire dampers was provided by the contractor. As a result of the contractor's finding, a study was performed to determine the ability of multi-section dampers to close. This study concluded several dampers should have larger nine pound springs, rather than the currently installed four pound springs. No action was taken to follow up on these recommendations at that time.

ChMLI The root cause of this event is the failure of the original plant design to consider the need to close the dmapers under ventilation flow conditions. At the time the plant was designed, this was the standard practice since closure under air flow is not required by the NFPA code. When Appendix R was implemented at CR 3 a contractor performed a study to identify deficiencies in the air damper capabilities. The study resulted in several reco.nmendations for improvement of the fire dampers. Due to FPC engineering personnel error, the recominondations in the study were not pursued. This was a violation of engineering procedures which were in effect at the time of the study.

EVENT EVALVATIONi The purpose of these fire dampers is to help assure a fire is limited to a single area. By limiting the spread of the fire and by protecting certain trains of equipment, availability of alternate equipment located in adjacent areas, which may be needed for safe shutdown, is assured.

The design error affects ten of approximately 120 total fire dampers. Several damper desions include an automatic trip of the associated operating fans. Most of the dampers tested to date satisfactorily completed the closure test under ventilation flow conditions. The dampers which have failed are those installed in areas where the ventilation flow is very large. Attached is a list of the l

dampers affecti.d, their location and a general description of the areas protected l by the damper, i

Since 1985, CR-3 has main ^ained an liourly roving fi re watch. The roving fire watch is required to walk an eslebiBhed route each hour and observe for fires.

If a fire is observed, the watch repo ts the fire immediately and extinguishes the fire if possible. The route (s) established since 1985 cover approximately l

90 percent of the plant fire areas. b aMition, the operability of fire i

detectors on at least one side of each fire barrier has been maintained and monitored, hMC Form 3He A f649)

\ - , - - - _ .-. . _ . . . . _ . - _ _ ~

7;.,o,- -. u . .m . . . . .m . , omo . o~

, i.,, . . -

LtCENSEE EVENT REPORT (LER) ',','l%',','L*,"@d',7J AFaM't ,'ijf Ol ,*',".'A' TEXT CONTINUATlON it",7;,,'P,%'K"71 Ar"I',7d'/.*','MT P.' 0'MfA i,T o, u.!.',',W4?"".E..t~ .a vm w *d,

. .O.M.e~

o ouu'oNM2 7Y,'?

f 408411, h. 6 til DWm.t,huwsta(D tia huMDE. il P.0$ IF

"* N CRYSTAL R1VER l' NIT 3 tui ur -. . . %,.w .u w =ac w au ,,i m 0l5l0l0l0l3l0l2 9 [0 _

0l 0l 2 0 l2 0 l4 or 0l5 Considering the safety aspects derived from the roving fire watch and existing fire detection and suppression systems in the areas of concern, the failure of some dampers to close under flow conditions does not create undue risk to the safety of the public.

CORRECTIVE ACTION:

Each fire damper flow rate has been compared to test data which reflects the maximum flow rate under which the damper is expected to successfully close.

Based on this review, dampers which were close to or exceeded the maximum allowable flow rate were drop tested. Additionally, certain dampers identified in the 1995 study were inspected to determine if larger nine pound spring. were installed. The dampers that failed the testing and inspection are identified in the Attachment.

To ensure the identified dampers close during a fire, FPC is pursuing the necessary improvements including:

o multiple closure springs and larger nine pound springs, i e o redesigned spring bracket,

, flow interrupters in the ducts, and e a temperature switch to turn off the associated fans and a higher temperature fusible link to allow the damper tc close under no flow conditions.

I Design basis documents have been updated to reflect the need for fire dampers to close under ventilation flow conditions.

Since 1985, several procedural and organizational changes have occurred which should prevent recurrence of the personnel error associated with the H85 Damper Study. These changes include strengthening engineering procedures by adding requirements for management review and approval of studies, improving and proceduralizing engineering problem reporting, establishing a Design Basis Engineering Group, and the addition of a Fire Protection Engineer to the fire Protection Staff.

PREVIOUS SIMILAR EVENTS 1 There have been three prior events related to fire dampers. Two of these events were related to design errors. This is the first event involving a design failure concerning damper closure under ventilation flow conditions.

i NRC Form 35&A i6497

.p.n, m. ui uvus.a aiouiaioant-o*

,,.,.,,,, m , y, y m ,3

, u nas wn UCENSEE EVENT AEPORT ILER) ' 5,'i*, ^,4'," *dM"J'!M',,'%T;2,' ,T," ',"2 TEXT CONTINUATI0N iU"4';"#,%'.","M MO R's','l# '",' 0,![,"N D,

o m ..ae f'P

','&"J,M"OM.un,a,1.O.,M..,,o,n m wo MWf., om. WV

. .L,L . f y SAM, H . DUCK,1NUMtlR40 gg h vuM8( A !$1 P AGE (3' I

CRYSTAL KlVER UNIT 3 nn a n,n..

0 l6 l0 l0 l0 l 3l 0l2 9 l0 -

0J 0l 2 -

0 l2 0l5 of 0 l5

- . .n,- ..c,n.- ,,on ATTA01tRt(I FIRE DAMPER TAG NUMBEL 9 5CBIEIJM*

FD-223, 224, These multi-section dampers separate fire areas in the 164', 145' 225, 226, and 124' elevations of the Control Complex (NA). These fire areas 238, 239, include the Main Control Room (C0-134-1188) and Cable Spreading Room (CC-134-118A), Operations Office (CC-145-119), Control Complex Ventilation Equipment room (CC-164-121A) and the 124' elevation hallway (CC-124-111). In the event of a fire that spreads between these areas, control of safe shutdown equipment will be transferred to the Remote Shutdown Panel and the dedicated Control Complex HVAC will be used. The dedicated HVAC should not De affected by a fire in this area.

FD-266 This damper is located in the floor of the 119' elevation of the Auxiliary Building (NE]. It separates the Reactor Coolant Pump Seal Injection fiiter Room (AB-953Y) from the 119' elevation Hallway (AB-119-60). A fire could affect the safe shutdown equipment on the 95' elevation and then move upward through this damper to the 119' elevation. Should a design basis fire spread to both of these areas, then Decay Heat (BP] and Nuclear Service Closed Cycle Cooling Water (CC] systems may not be available for shutdown.

FD-271 This damper is located in the floor of the 119' elevation of the Auxiliary Building (NE]. It separates the Miscellaneous Radioactive Waste Rooms (AB-95-3K) from the 119' elevation Central Hallway (AB-119-6J). Should a design basis fire spread to both of these areas, then the Makeup Injection [BG) and Decay Heat (BP]

systems may not be available for shutdown.

FD-273 This damper is located in the floor r f the 119' elevation of the Auxiliary Building (NE]. It separates the 119' elevation Central Hallway (AB-119-6J) from the 95' elevation Central Hallway (AB 3G). Should a design basis fire spread to both of these areas, then Makeup Injection and Decay Heat systems may not be available for shutdown.

FD 278 This damper is located in the wall between the intermediate and Auxiliary (NE] Buildings on the 95' elevation. This damper separates the 95' elevation North Hallway Nuclear Sample Room (A8-95-38) from the Intermediate Building Fenetration and fan area (IB-95-200C). Should a design basis fire spread to both of these areas, at least one train of safe shutdown equipment will still be available.

• Evaluation is based on the Crystal River Unit 3 Updated fire Hazards Analysis, Revision 2, November 1989. Mitigating factors for a fire in the affected areas is discussed in " Event Evaluation".

..e. =4 .. .

~. _ - _ _ _ _ _ _ _ . -