ML14058A077

Slides on Oconee Ssf Flood Barrier Breach SDP Lessons Learned

ML14058A077
Person / Time
Site:	Oconee
Issue date:	02/04/2014
From:	Circle J A, Vail J A NRC/NRR/DRA/APOB
To:
Shared Package
ML14055A421	List: ML090570117 ML090680737 ML091620669 ML092020480 ML092940769 ML100280954 ML101750619 ML101890803 ML102990064 ML103410042 ML12268A404 ML14055A410 ML14055A411 ML14055A474 ML14058A015 ML14058A016 ML14058A017 ML14058A018 ML14058A019 ML14058A020 ML14058A021 ML14058A022 ML14058A023 ML14058A024 ML14058A025 ML14058A026 ML14058A027 ML14058A028 ML14058A029 ML14058A030 ML14058A031 ML14058A032 ML14058A033 ML14058A034 ML14058A035 ML14058A036 ML14058A037 ML14058A038 ML14058A039 ML14058A040 ML14058A042 ML14058A043 ML14058A044 ML14058A045 ML14058A046 ML14058A047 ML14058A048 ML14058A049 ML14058A050 ML14058A051 ... further results
References
FOIA/PA-2012-0325
Download: ML14058A077 (21)

Text

Oconee SSF Flood BarrierBreach SDP-LessonsLearnedJeff Circle, DRA/APOB.

James Vail, DRAAPOB.NHRUe0 -Sen 'ive/Inte oHf o t'njji0USS.NRCIflNTE STATES NUTCLEAR REGIIATORY COMMAESOProtectinS People and the Environment Purpose* To inform DRA personnel of the lessonslearned from the disposition of theOconee Standby Shutdown Facility (SSF)wall SDP finding., To help encourage and foster aquestioning attitude by staff on regulatory and enforcement processes.

NR &ernetZmats 0 !nU.SNRCMU.iED 5TT.F. B L.

NO'!! .s Protewng People and the Environment Discussion Topics# Background of finding.e Technical lessons learned.s Process lessons learned.s Actions taken.e Considered future actions.NRRUse lySeivInto Infma! ni0i2US.NRCLiNfED STATES NUGLEA REDGLU.MRYCOMMISSION

%.Protecting People ad Ohe Environment Importance of the SSF at Oconee# The SSF contains the only means to shut allthree units down following a station blackoutinduced by catastrophic flood, fire, or otherexternal events.@ The Oconee site does not have emergency diesel generators.

-On-site emergency ac power is provided by twohydro-electric generators at the Keowee dam.s Catastrophic flooding to the site can render allthe switchyards and Keowee dam unavailable thus blacking out all three units.-Under these conditions, the additional failure of theSSF function results in core damage.NRRcials Myqnsi We0U@S1NRCUNIMTED SATESnVCLEAR REGLLUMRYOMMSSION Protecting People and the Environment Aerial View Showing The Relationship of theOconee Site Environment and Lake Keowee..... * ...- '..............

Use n .SeItiveN R R In Infort!-r.

n0IJOS NRCU Nn S TAE LCL.LU REGIRl.KERYCOMSO P rotecting People and the Environment Aerial View of the Oconee Sitecial Us ly -Se itiveA ter IInf (~ma!'tn'v 0USONRCIMNTED STATES NUCtEBR EGULqATRY COMMISSION Protecting People and4the Environment The Flood Barrier FindingV, .Licensee opened an access coveruncovering a previously cut hole inthe wall on August 13, 2003.-Should have done a 10CFR50.65

, (a)(4) assessment immediately.

-Should have done a 1OCFR50.59

.* evaluation after 90 days.* Licensee opportunities to identify7.issue-June 2,2005 NRC inspectors notified the licensee of condition.

Licensee issued PIP (condition report in their corrective action*system).

Corrective action not, taken.-,August 3, 2005 NRC inspectors questioned lack of corrective actionFive.foot flood wall and licensee issued a further PIP.protecting the doorway.

• Opening sealed on August 3, 2005.Opening in outside wall with access plateon inside of SSF building.

OfiiUse nly Sen tive fl -U.S.N RCN R R nte Ifrm ' in UNMlf 5TATES hI.CI LAR REGURflORY COMM.ISSION 49I4i i FW ii Protecting Peoplfe and the Enviroment Interior and Exterior Walls of SSFNRRQUS.NRClNMTDlSTATES

?I1CLEAR REGIJUMT

~COMM5SON Protecting People and the Environment The Flood ScenarioRuptiire of JocasseeLake Jocassee-300-fNine Timesi. _ ., fDamBoOa-Wes Creek .cpA3Oconee Site.a" ., .* ..#'*.oSIxOMiUse fly -ensi lye-InternaillInformatibl 0US.NRCI#PTED S eApeS R a1 d cthqEnmiRYOMMeN Protecting People and the Environment Aerial View of Jocassee Dam- Upstream of theOconee SiteNRRO a ial Us Only sitieInnterna nno mato0,U.SNRCUNrtED) STATES MICITAR ISGU1ATORY 0MMOSMSoN Protectin~g People anid the Enzironment SSF Catastrophic Flood Heights GivenRupture of Jocassee DamLicensee study in 1980s established a resultant floodheight of approximately 5 feet.-Justification for erection of the 5-ft wall around the doorway.-Study is unavailable.

-Flood height almost matches that of overtopping of Keoweedam.* Licensee commissioned additional study by FERC in1992.-Established a higher resultant flood height range of 12-16 feet.-Licensee still used the 5-ft height for their IPEEE submittal developing a 20% split fraction apportioning the Jocassee Damrupture frequencies to account for these higher FERC floodheights.fici Us nly SensiUvRC inenlfo iv -)U.SNRC..Cl W OW §i ýProtecitng People a, d the Environment Background Summary, August 17, 2006- SERP meeting assessed as preliminary WHITE based on ablended qualitative and quantitative risk-informed approach (pre-IMC 0609 App M).* August 31, 2006 -Choice letter sent to licensee.

, October 5, 2006 -Licensee provided written response to choice letter and waivedregulatory conference.

, Nov. 22, 2006 -Final significance determination issued. WHITE based on qualitative erosion of defense-in-depth, but includes quantitative CDF based on apportioning flood frequency to flood height.* December 20, 2006 -Licensee appeals the final significance determination.

Requests NRC to accept incomplete, un-docketed new information on Jocasseedam seismic failure..Appeal panel convened on January 9, 2007.* Licensee contractor seismic fragility analysis of Jocassee issued January 29, 2007..March 1, 2007 -Appeal panel upholds WHITE finding..May 3, 2007 -Licensee requests reassessment of final significance determination.

.June, 2007 -Assembled a team to review new information.

Flooding expert reviewof data on random dam failure.* June 28, 2007 -Follow up telecom with Licensee on dam failure questions andcomments.

• July 17, 2007 -Licensee response to analysis questions by email..September 5, 2007 -Final SERP split vote forces upper management resolution.

• November 20, 2007 -Final determination letter released to licensee.

NOffi al se "Only siN eProtediang People and ke Environment Jocassee Dam Random FailureFrequency Licensee developed random dam failure frequency forIPEEE submittal based on rockfill dam failures perpopulation of earthen,

rockfill, and composite dams.-Severely underestimated frequency by an order of magnitude at1.4E-5 per year.s As part of the second appeal, DRA/APOB investigated the derivation of this frequency.

(b)(7)(F)

NRRU.S.NRCLFNTTED STATES5 MIiLLAR REGt'UTORT COMMISSION Protectinig People and the Environment Dam and Dike Failure# Dam failure is well documented and can be characterized by type ofdam.# Of the 79,777 dams in the US, 72% are embankment type and 28% are concrete.

1. Nineteenth century dams would fail at 5% in the first five years afterconstruction but would settle out to a 1 to 4% additional failure by20 years of life.s This failure rate was reduced to 2% in the first 5 years for damsbuilt after 1930.a By 1960, dam failure rates were less than 0.01 % due to betterengineering.

@ Whatever the era, half of all dams that ever fail, do so in the firstfive years. This high infant mortality is often due to piping in the soilaround the dam or underneath it. Even concrete dams are notimmune. However, dam construction dropped dramatically after1980 so that nearly all dams are older than 5 years.lnalnf rma n U.S.NRCN R R -0Nmo STATESNIRUAR REGLI ATORCProtecting People and the Environment Technical Lessons LearnedInvestigation of Dam Failure Frequencies http://crunch.tec~army.mil/nid/webpaqes/nid.cfm http://npdp.stanfordedu/index.html All Arch DamsAll Buttress DamsAll Concrete DamsAll Earth DamsAll Gravity DamsAll Masonry DamsAll Multi-Arch DamsAll Rockfill DamsAll Stone DamsAll Timber Crib DamsTotalButtress Dams Over 50 FeetArch Dams Over 50 FeetConcrete Dams Over 50 FeetEarth Dams Over 50 FeetGravity Dams Over 50 FeetMasonry Dams Over 50 FeetMulti-Arch Dams Over 50 FeetRockfill Dams Over 50 feetTotalFailures22103662850723425Failures02056700469Dam.years 91019819110227224040312279821692240558721136565362605987Dam-years 18765667192151448101954219877719900213184Mean2.1E-041.9E-049.3E-051.6E-042.3E-042.2E-041.5E-041.3E-041.7E-043.6E-041.6E-04Mean2,OE-042.8E-048.2E-053.8E-043.2E-042.OE-042.4E-042.1E-042.4E-04NRRLhi 'alUseO90

-ensit' eNU.S.NRCIWO STDIAIES NIT1AIX RiWAUTRY COMIWSSION ProtectiNg People 4111 the Environment Licensee Evaluations s Only two IPEEEs addressed dam failure floods quantitatively

-FtCalhoun and Diablo Canyon. Everyone else only addressed probable maximum precipitation and screened out dam failure aslow probability.

s Unfortunately, there were few dam failure data sources aroundback then, so many plants used the estimate published inNUREG/CR-5042.

The data source for the estimate inNUREG/CR-5042 was the Oconee PRA -NSAC/60.

The estimateof 2.5E-05/dam-year in NSAC/60 was done in error by an order ofmagnitude and it propagated throughout the industry.

e

References:

-NUREG/CR-5042, "Evaluation of External Hazards to Nuclear Power Plants in the UnitedStates.-NSAC/60, "Oconee PRA"Na Uely-e ll e US.,NRCSI rmat" n 0~ inau~gjp~iProtecting People anid the Environment SPAR Considerations for External FloodDams as far up or downstream as 300 miles should be considered for bothflood and loss of heat sink. It is noteworthy that all forms of dams have afailure rate between 1 E-4 and 4E-4, even for blue sky events.Determining flood levels, however, is a complex matter. The USACOE hassoftware named HEC RAS that when combined with GIS geographical datawill model river flow and flooding in great detail.Weather based floods remain in the deterministic world because the inputconditions are always from the same source as was used in the originalplant design basis. Besides, the growth of the maximum precipitation onlyincreases about 20% when a 100 year interval is compared to a 1000 yearinterval.

With only 100 years of data available in many locations, projecting beyond a I E-3/yr event is very uncertain anyway.N R R 0n nlfom o IN S1TAtSNUIA RE .TR£OMSHIC-8 liýProteding~People and th'e nviunmati Process Lessons Learned@ Process did not account for situations wherelicensee did not communicate to the regionaloffice and waived participation in a regulatory conference.

s Process did not accept new information prior tofinal determination.

s Without a regulatory conference, communication was impacted to program officeand regional management of potential greater-than-green SDP findings.

sly Sensi ,e USNRCl ,.. UNITED STATES NUCL[AR ItEGiATORn COMMISSION i~m. b People ,,nd the Envir.onment Actions TakenI IMC0609 in draft revision to account for:-Licensees may present new information provided thatthey informed the agency either during a regulatory conference or in writing before the final significance determination is made.-New information has to be submitted within the 30-day appeal period.-Using best available information to determine finalsignificance in a timely manner.-NRR concurrence with regional management onaccepting an appeal.Ral Use I -Sensi imrtm .N RC UPR Intrc Pnfor atioand 7 T h Eni RRnCMISN Actions Taken (cont'd)e Training conducted to regional SeniorReactor Analysts (SRAs) in recentcounterparts meeting.@ Met with NSIR to inform DHS on potential dam and other outside infrastructure vulnerability.

NRRfi~ Use 0 y -e~ns' ive,U.S NRCINrwrTITOSAE 1:C1 FARREGIRAToRY COMMI1SSION Protecting People and the Environment Considered Future Actions@ NRR Program Office Director and/orDeputy Director should be informed of allgreater-than-green findings.

1. Exploration of licensing and potential backfit requirement for protection againstmost likely flood.s Generic communication to licensees onflooding.

NRR 0 I U.S,NRC