ML20134B723

Summary of 961218 Meeting W/Westinghouse in Rockville,Md Re Issues Associated W/Fire PRA for AP600.List of Attendees & Handouts Encl

ML20134B723
Person / Time
Site:	05200003
Issue date:	01/29/1997
From:	Joseph Sebrosky NRC (Affiliation Not Assigned)
To:	NRC (Affiliation Not Assigned)
References
NUDOCS 9701310045
Download: ML20134B723 (20)

Text

.

January 29, 1997

-.g

,A

Wsstinghause Electric Corporation 1

FACILITY:

AP600 i.

SUBJECT:

SUMARY OF DECEMBER 18, 1996, MEETING TO DISCUSS ISSUES RELATED TO j.

THE FIRE PROBABILISTIC RISK ASSESSMENT (PRA) FOR THE AP600 i

j The Nuclear Regulatory Commission (NRC) staff and representatives of Westing-house Electric Corporation held a meeting on December 18, 1996, at Westinghouse's office in Rockville, Maryland, to discuss issues associated l

i with the fire PRA for the AP600. Attachment 1 is a list of meeting j

attendees. Attachment 2 is a copy of handouts provided by Westinghouse.

The meeting began with Westinghouse providing an overview of how they per-formed their fire PRA including the assumptions and fire scenarios that were used. The fire PRA requests for additional information (RAI) were then i

discussed. The RAls were sent to Westinghouse in letters dated September 18, j

1996, and December 2, 1996. The September 18, 1996, letter concerned the at-power portion of the fire PRA while the December 2,1996, letter concen-trated on the shutdown portion of;the fire PRA. Westinghouse provided draft responses to the at-power portion ~of the fire PRA questions in a December 9, 1996, letter.

Each draft response in the December 9,-1996, l'etter was discussed. The staff found some of the responses acceptable, while others were determined to need more information.

In some cases-the< staff needed to look at the response in

.more detail before it determined'if the response was acceptable. Attachment 3 i

was provided to the staff during the discussion of RAI 720.345. The diagram was used to explain how transfer to the Remote Shutdown Workstation from the Main Control Room would be accomplished in the. event of a fire.

original signed by:

Joseph M. Sebrosky, Project Manager Standardization Project Directorate Division of Reactor Program Management Office of Nuclear Reactor Regulation Docket No.52-003 1

Attachments: As stated cc w/ attachments:

i See next page

%y DISTRIBUTION w/ attachments:

! Docket File,

PDST R/F JSebrosky PUBLIC TKenyon DTJackson BHuffman i

DISTRIBUTION w/o attachments:

FMiraglia/AThadani, 0-12 G18 RZimmerman, 0-12 G18 BSheron, 0-12 G18 jIO.3, TMartin DMatthews TQuay ACRS (11)

JMoore, 0-15 B18 WDean, 0-17 G21 MGareri, 0-8 H3

' NSaltos, 0-10 E4 MPohida, 0-10 E4 JFlack, 0-10 E4 n 10 0 07 JH ]mes, 0-8 D1

HLi,

-8 H3 DOCUMENT NAME: A: FIRE PRA. SUM n

h,e. e.n. mi.

\\t. u.a.ae h. es bem *T/G y wMhout attachmenuencloewe T = Copy with attachment /encloswo "N* = No copy 0FFICE PM:PDST:DRPM BC:SPSB:bE$A D:PDST:DRPM l

l NAME JSebrosky,Mff JFla'qk U /

TQuay m W DATE 01/ 2.7/97 //

01/s W 01/ M/97 "FICIAL RECORD COPY 9701310045 970129 PDR ADOCK 05200003 F

PDR

5

{

Westinghouse Electric Corporation Docket No.52-003 l

)

cc: Mr. Nicholas J. Liparulo, Manager Mr. Frank A. Ross Nuclear Safety and Regulatory Analysis U.S. Department of Energy, NE-42 Nuclear and Advanced Technology Division Office of LWR Safety and Technology j

Westinghouse Electric Corporation 19901 Germantown Road P.O. Box 355 Germantown, MD 20874 i

Pittsburgh, PA 15230 Mr. Ronald Simard, Director i

Mr. B. A. McIntyre Advanced Reactor Program j

Advanced Plant Safety & Licensing Nuclear Energy Institute Westinghouse Electric Corporation 1776 Eye Street, N.W.

I Energy Systems Business Unit Suite 300 i

Box 355 Washington, DC 20006-3706 Pittsburgh, PA 15230 i

Ms. Lynn Connor Mr. John C. Butler Doc-Search Associates i

Advanced Plant Safety & Licensing Post Office Box 34 1

Westinghouse Electric Corporation Cabin John, MD 20818

{

Energy Systems Business Unit i

Box 355 Mr. James E. Quinn, Projects Manager Pittsburgh, PA 15230 LMR and SBWR Programs GE Nuclear Energy Mr. M. D. Beaumont 175 Curtner Avenue, M/C 165 Nuclear and Advanced Technology Division San Jose, CA 95125 i

Westinghouse Electric Corporation i

One Montrose Metro Mr. Robert H. Buchholz 11921 Rockville Pike GE Nuclear Energy Suite 350 175 Curtner Avenue, MC-781 1

Rockville, MD 20852 San Jose, CA 95125 t

Mr. Sterling Franks Barton Z. Cowan, Esq.

U.S. Department of Energy Eckert Seamans Cherin & Mellott 1

NE-50 600 Grant Street 42nd Floor

{

19901 Germantown Road Pittsburgh, PA 15219 l

Germantown, MD 20874 Mr. Ed Rodwell, Manager 4

i Mr. S. M. Modro PWR Design Certification Nuclear Systems Analysis Technologies Electric Power Research Institute Lockheed Idaho Technologies Company 3412 Hillview Avenue Post Office Box 1625 Palo Alto, CA 94303 Idaho Falls, ID 83415 l

Mr. Charles Thompson, Nuclear Engineer AP600 Certification l

NE-50 19901 Germantown Road

{

Germantown, MD 20874 i

1 i

i I

2

=.

~

WESTINGHOUSE AP600 FIRE PRA MEETING ATTENDEES December 18, 1996 1

M8HE ORGANIZATION JIM WINTERS WESTINGHOUSE BARRY SLOANE WESTINGHOUSE KEN DEUTSCH WESTINGHOUSE i

CINDY HAAG WESTINGHOUSE i

)

AMIR AFZALI WESTINGHOUSE CONSULTANT i

i HULBERT LI NRR/DRCH/HICB NATHAN SIU INEL MARIO GARERI NRR/DRCH/HICB NICK SALT 05 NRR/DSSA/SPSB MARIE P0HIDA NRR/DSSA/SPSB JOHN FLACK NRR/DSSA/SPSB JEFF HOLMES (PART TIME)

NRR/DSSA/SPSB J0E SEBROSKY NRR/DRPM/PDST j

1.

1 4

i i

Attachment I

1).

Westinghouse Electric Corporation i

i i

i 4

l i

AP600 INTERNAL RRE l

RISK ANALYSIS

)

i l

i 4

1 4

5 d

e i

i l

I l

j December 18,1996 i

i

}

1 i

)

i

i l

AGENDA i

Summary of Fire PRA Assumptions Fire scenarios i

Discussion of RAls i

Power conditions Shutdown conditions Summary / Conclusion

i AP600 Internal Fire Analysis Basic Approach Perform a qualitative assessment of the internal fires impact on systems required for normal and safe shutdown of the plant Quantify the risk posed by the postulated impact in terms of core damage frequency Analyze the results for sensibility and insights l

w

4 i

Major Steps in Qualitative Assessment l

I Analysis of fire area vs. plant trip / safe shutdown equipment damage was performed.

i

)

For fire areas with no credible fire propagation j

mechanism, the analysis was performed assuming j

that all equipment and cables in the exposing fire area were damaged.

For fire areas where fire propagation from them j

was determined to be possible, the analysis was j

performed assuming that all equipment and cables i

in both the exposing and the exposed fire areas j

were damaged.

4 i

l

}

t

Major Steps in Qualitative Assessment A fire area was screened out from further analysis based on the following criteria:

If a fire originating in that fire area was not expected to create a demand for safe shutdown under normal plant operating conditions, and if none of the PRA-credited equipment were considered damaged.

If the fire was expected to cause a plant shutdown due to technical specification requirements but was not postulated to impact operability of any PRA-l credited shutdown systems.

For those areas that did not screer' out, a qualitative assessment of the consequences of the fire was performed. As a part of this qualitative evaluation, the consequences of different fire-induced cable failure modes (open or short) were considered.

i l

l l

1

_...._._m._.

.....--_..__..m.._.--_-__..

i i

- i Dec====e Type: Analyses File Project: AP600 Quebeative Analysis Fue Neunber: RZ39.FP.01.1 Revision No.: O Prepared By: A. Afzali Beviewed By: D. Hua Date: 06/27/95 TAEEE 1C - APSee FIRE AREAfZONE QUAUTATIVE EVALUATION l

Fhe Assa Neanber:1291 AF 02 Fhe Asea Ce41- _ Divtsson B Basedes Fire Zone Number Fire Zone Desenphon Combushble Loadng (BTU /sq. ft.): 69,286 Detechon Srnoke kimed Suppression Hose Station

1. Does arealzone contain equipment and/or cable required to achieve hot shutdown per PRA? (Yes/No) i YES. Division B Batteries

2. Is there a requirement for plant shutdown given tre damage for Eres contained wittwo the area / zone? (Yes/No)

/'A'

'f/Shg 1

YES See Assumphons kMWM Won / / z_

3. Is Are propagabon credible for Gres initiated in this arealzone? (Yes/No)

YES. Fire propagation is assumed to be possible for tres originating in Are areas with intemal combushble loadings of greater than 20,000

[

BTU /sq. ft. Table 1E presents a Est of potentially at risk Are areas. The worst case possible Are propagation scenano would be a lire spread to the division B DC equipment room (Are area 1222 AF 01), since spreading to 1201 AF 03 is not conssdered possible.

4. Is there a requirement for plant shutdown given Are propagation from this area / zona? (Yes/No)

Since a Hre in this area is assumed to result in a plant trip, this question is not relevant.

I

5. Can this area be quatta5vely screened out? (Yes/No)

NO.

A-79 t

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

~

~ i I

l 4

Decoment Type: Analyas File Progett: AP600 Quebeneve Analysis File Neueber: RZ39.FP.01.1 Revaise No.: o Prepered By: A. Afrali Reviewed By: D. Heaneke Date:06/27/95 TABLE 1C - APtes FIRE AREAL 2ONE QUAUTATIVE EVALUATION Fire Ases Number.1291 AF 92 Fhe Asea Descdymon: Division B Basedes Fire Zone Number Fire Zone Desenphon Combustible Loading (BTL#sq. ft.): 69,286 Detechon Smoke IIxod Suppression Hose Station

6. Postulated Are scenado? (Apptcable/Not Apptcable)

-For tres conened wittun the area, the B divisson of power and controlis assumed lost with no other systems or equepment impacted Fire i

damage state (FDS) 1AB4 is designated to represent the damage caused by the above desenbod tre scenario.

-For fires propagalm0 out of this area, the followmg scenarios are postulated:

l Fire Scenario 1-The impact of fires covered under this Are scenano wig be bounded by the loss of the equipment discussed for FDS 1AB4. Fire damage state (FDS) 1AB4 is also designated to represent the damage caused by the above described Nre scenario.

l Fire Scenario 2-The impact of Ares covered under this Are scenario will be bounded by the loss of the egiapment discussed for FDS 1AB4 plus spurious openmg of the ADS valves due to the Are induced dernage to division B related cabEng ulitzed to support ADS valves. Fire damage state (FDS) 1 ABS is desegnated to represent the damage caused by this fire scenano i

I l

t i

I 4

A-80

[

i 1

i Major Steps in Quantitative Assessment j

For each fire area surviving the qualitative screening, core damage frequency due to fire damage, coincident with fire unrelated unavailability of redundant / alternate safe shutdown equipment was evaluated.

4 l

CDF was evaluated based on the assumption that all fires damaged all potential targets in one step.

An initiating event was assigned to each scenario based on the information collected during the qualitative analysis, together with potential damage to the PRA-credited safe shutdown systems.

l Fire Propagation Fire propagation to more than one fire area was considered, but sequential fire propagation (from a fire area to an unconnected fire area via an intermediatory fire area) or simultaneous fire propagation to more than j

one fire area were judged not to be credible.

l Fire Propagation Possibility -

Fire propagation across a fire area boundary

{

without sealed openings (e.g., walls with doors, vent louvers, cable penetrations, etc.) was judged not to be credible.

Fire propagation across a barrier was judged not to be credible if the deterministic fire protection assessment had credited the barrier as being capable of preventing fire propagation, unless:

i.

The barrier's integrity had been compromised by a barrier. element failure (e.g., penetration seal failure)

AND ii.

If the fire suppression activities in the exposing location were to fail to contain the fire within the area.

4 Fire Propagation l

Fire Propagation Probability -

Probability of fire propagation was calculated based on the failure probability of the fire barrier and fire suppression system in the exposing fire area.

i.

Fire suppression was not credited to limit fire damage within an area 4

ii.

Manual fire suppression was not credited for preventing propagation to other fire areas.

iii.

Automatic fire suppression system in the exposed fire area was not credited to prevent fire propagation to that area or limit the consequences of the fire.

i Fire Propagation l

i Fire propagation across the containment inter-zone fire i

barriers was not considered to be credible on the basis that

{

i.

Per deterministic fire analysis, each fire zone is physically separated from the other fire zones by structural barriers, or in one case by large distance, j

such that fire propagation across a fire zone boundary can be dismissed.

AND j

ii.

The containment inter-zone barriers are designed j

not to contain sealed openings (e.g. doors or penetration seals).

i a

i

4 Fire-Induced Spurious Actuation l

Probability of a specific fire-induced fault occurring was I

calculated based on the nature of the fault.

i Based on NUREG/CR-2258 evaluation, the best estimate conditional probability of a hot short event was estimated as 0.06. Note that for large LOCA at least two hot shorts are required. Thus, the conditional probability of a fire-induced large LOCA was estimated to be 3.6E-3.

1

I Fire Analysis Conservatism j

Conservatism used in the AP600 fire analysis include:

The fire frequency was estimated based on the fire events which have occurred in the existing plants. The AP600 design is such that the number of potential ignition sources is significantly less than that in a i

conventional plant (less pumps, less MOVs, less MCCs/switchgears, etc.)

A fire originating from any ignition source in an area is assumed to disable all equipment located in the fire area. A review of the historical evidence indicates that most fires are localized fires with limited severity. In most fire PRAs, a fire severity factor is usually used to adjust the fire frequency to a more realistic value.

Manual fire suppression is not credited to limited the extent of damage in an area nor to prevent fire 1

propagation to an adjoining area. Again historical i

evidence indicates that the majority of suppressed fires j

were manually suppressed with little or no additional damage.

l i

l 1

i i

i

i Fire Analysis Conservatism i

Automatic fire suppression system in the exposed fire area was not credited to prevent fire propagation to the area or limit the consequences of the fire.

During the quantification process, the nonsafety-related l

systems (e.g. main feedwater, normal RHR, offsite l

power, etc.) are not credited. However, fires in only a few AP600 fire areas would be capable of disabling all the nonsafety-related systems.

1 l'

Design Features Design features that minimize risk from fire include:

l The AP600 fire protection design provides, to the extent possible, for separation of the alternate safety-related shutdown component and cabling using three hour rated i

fire barriers without any penetrations.

The high reliability of the shutdown systems as i

determined by the internal events model and passive nature of these system, making them less susceptible to j

fire-induced failure mechanism i

(e.g., loss of power).

l e

J Summary of AP600 Fire Analysis Results AT-POWER Plant 3.5E-07 /yr Control Room 5.7E-11 /yr Containment 3.0E-07 /yr Total 6.5E-07 / yr SAFE SHUTDOWN Plant 2.5E-08 /yr Control Room 9.9E-09 /yr Containment 3.0E-10 /yr Total 3.5E-08 / yr MID-LOOP Plant 2.6E-07 /yr Contro; Room 7.8E-10 /yr Containment 5.1E-08 /yr Total 3.1E-07 / yr

~

I I

I I

PMS PMS PMS PMS DIVISION A DIVISION B DIVISION C DIVISION D MCR RSV MCR RSV MCR RSV MCR RSV MUX MUX MUX MUX MUX MUX MUX MUX l

l l

~

l I

1 l

l

~

l l

l l

l l

I l

lpassh=EE:d==========h-----J

____[______j l

i llll L_____

-_ _ _ _L _ _ _ _ _ _

. _ 9,l L_ _ _ _ _ _ _

lIl j i

ii i I

il l llI;jll s'i l llll Div a xrcR sviTCn jf l

[Il DIv C XFER sVITCH DIV A XFER sVITCH

{

j oiv o xrcR sviTCn aa MAIN CONTROL TRANSFER SWITCH REMOTE SHUTDOWN ROOM PANELS WORKS T AT ION

"'