ML14055A410
| ML14055A410 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 02/04/2014 |
| From: | Office of Information Services |
| To: | |
| Shared Package | |
| ML14055A421 | List:
|
| References | |
| FOIA/PA-2012-0325 | |
| Download: ML14055A410 (341) | |
Text
Oconee Nuclear Station Standby Shutdown Facility (SSF)
Flood Barrier Breach Violation informal reassessment USNRC UtD TATESNtCLEAI R iMLtAmR I
Pn~tecting People and the Environment
Why is the SSF wall is needed?
Two flood studies show the Oconee site vulnerable to flooding from a Jocassee dam failure
- 1982 flood study showed the need to protect from flooding up to 4.71 feet on the SSF
- 1992 FERC study showed that flooding could occur at 12 to 16 feet.
Oconee decided to protect from "the most likely flood scenarios" by building a 5-foot wall in 1988
- wall will protect against 80% of floods based on engineering judgment - no technical basis The probability that a flood will occur due to a failure of the Jocassee Dam contains large uncertainty.
I UNITED STATES N'MCLEAR PGULmTORY COMMIS*ION Protecting People and the Environment
I.
SSF Wall Relative Heights of Flood Barriers and Penetration SSF FLOOD ELEVATION REFERENCE DRAWING EL 801' 9 (5.75 ft)
WE.BOf (W.5 A)I
~TOP ssFFOTo W RU W 8OTTO.M (7$.CF
¶ oE MOTO
~SSF R=OO.
ELY Th!ý4!_Q Fl..
YAAD (I~~l IELFVýW gO 7Wf 11 EL 801T (5.06')
E....
L 8
.0.0... )
EL 8001Y.5 (4.71')
EL 801' (5.0')
RliV. W'-. f"* A.
.4
.4
'.4.4 6**,
4 4
1*.4 a6.0a4 0
a a
0 4
A' 1'
A J
A1 6
d 4 A a
m.m.
- m r i-~iI-*i't-ii ___ i-Il ___
il-Il I-"tiI~-1Tl-~iii 3
U.SNRC UNITE STATES MXUtAR REMtATURY COMUSSION Protecing People and tke Enuironmen
Nature of Violation/Performance Deficiency and SDP
- Open penetration access cover in the SSF exterior wall for approximately 2 years.
- No 50.65(a)(4) evaluation done by licensee.
- Opening below 5-ft max. flood height identified by licensee in IPEEE.
Susceptible to site flooding from rupture of Jocassee Dam 11-miles upriver.
Region Il SRA performed analysis based on information from the licensee and apportioned flood height based on split fractions to arrive at a quantitative WHITE finding.
Jocassee Dam Failure Mode Increase in CDF (per Reactor.Year)
Random 1.84 x 106 Seismic 1.44 x 10-6 Total 3.28 x 10-6 4
UNTIED STATES NU.CLEAR REGULA70RY COWMISSION Proteclng Peopleanmd thie Environrnext
Summary Timeline August 17, 2006 -SERP meeting assessed as preliminary WHITE based primarily on qualitative aspects (pre-MC 0609 App M)
- August 31, 2006-Choice letter sent to licensee.
- October 5, 2006 - OCO provides written response choice letter to waived regulatory conference.
- Nov. 22, 2006 -FSD issued. WHITE based primarily on qualitative erosion of defense-in-depth,but includes quantitative CDF.
- December 20, 2006 -OCO appeals the FSD. Requests NRC to accept incomplete, un-docketed, Jocassee fragility study
- January 9, 2007 - Appeal panel convened
- February 5, 2007 - OCO sends completed seismic fragility analysis of Jocassee to NRC.
- March 1, 2007 - Appeal panel upholds White finding
- May 3, 2007 - OCO requests "reassessment of FSD"
- June 28, 2007 -Follow up telecom with OCO after seismic fragility analysis was evaluated.
- July 17, 2007-OCO response to analysis questions by email.
- June 22, 2007 - Reassessment of FSD assigned to RII
- August 27, 2007 - OCO 95002 inspection - Emergency AC White MSPI, U3 sump debris, SSF flood barrer breech white finding
- September 20, 2007 - RII reassessment results due to Oconee (90 days from June 22)
C 2US.NRC UNrr STATES NUCLEAR RECILATORY C04MMISSON Protehueg Peopfe ad the Eaviroomeni
Reassessment Team Evaluation
- Seismic fragility review did not endorse licensee's conclusions completely
- NRC agreed that the seismic frequencies were sufficiently "low" to be insignificant. Therefore, only random dam failures were considered.
- Epistemic uncertainty in Jocassee rupture frequency
- Licensee and NRC cannot resolve resultant flood height fractions resulting from below opening to 5-ft.
t Assume that mean dam rupture frequency producing a 4.71-ft flood identical to a mean frequency producing a 5-ft flood;
- in other words there is no probability distribution for floods of various heights, floods at any height on the 5.foot wall are equally likely
- Using licensee's minimal IPEEE SSF failure cutsets with OCO revised flooding frequency results in ACDF of 8.22 x 10.6 per year.
6 yUSNRC UNfED STATES NICL*AR RLGULTORY COMMISSION Protecting People and *th Enzironmet
Jocassee Dam Random Failure
- Licensee computed random failure frequency based on operating industry history of failures for rockfill dams of over 50-ft in height
@ Discrepancy exists on counting the denominator of total operating dam-years.
- Licensee included operating years of other dam types in demoninator, but did not count their corresponding failures in the numerator.
s NRC performed independent random dam failure analysis for rockfill dams >5 yrs old and >50 feet in height
- Random frequency could be higher which might offset any uncertainty in the resultant flood height difference between the bottom of the opening and the 5-ft height of the wall.
Resultant ACDF could range from WHITE to RED in severity, SU.S.NRC ProEDcSTATEU PeColeR RKEI.RY COMMISSION
Risk-informed Regulation and Defense in Depth
, Used as a way to account for uncertainties in equipment and human performance.
- The independence of barriers have been degraded.
, Licensee has an exclusive reliance upon the SSF to prevent core damage without redundancy or diversity of mitigation.
- Any functional degradation of the SSF flood barrier from these initiating events directly increased the failure probability of the SSF and therefore, increased the likelihood of core damage.
Using a blended qualitative and quantitative perspective, the NRC's final Significance Determination remains low to moderate (White)".
UNITED STATES N.1CUAR lEGUIATORY COMMISSION Protecting People 4nd the Environment
Judgment needed
, Does a 3.5.inch change in flood wall height that last for longer than a year result in an increase in core damage frequency greater than 1 E-6?
, Should the NRC conduct additional inspection to determine why a licensee did not assess the maintenance risk of a small breach in the only barrier between flooding and core damage and allow it to exist for two years?
%2jU.S.NRC uirrED STATES NIJCILER REGULATORY COW~SSION frtecring Peopl* and the Environmmnt