ML110140845

Letter to Board from NextEra Energy Seabrook, LLC, Enclosing Document Containing New Information Potentially Relevant and Material to Issues Raised in Friends/Necs Contention 4

ML110140845
Person / Time
Site:	Seabrook
Issue date:	01/14/2011
From:	Hamrick S NextEra Energy Seabrook
To:	Kennedy M, Paul Ryerson, Richard Wardwell Atomic Safety and Licensing Board Panel
SECY RAS
References
RAS 19428, 50-443-LR, ASLBP 10-906-02-LR-BD01
Download: ML110140845 (126)

Text

NextEra Energy Seabrook, LLC 801 Pennsylvania Avenue, N.W., Suite 220, Washington, D.C. 20004-2604 Steven Hamrick (202) 349-3496 steven.hamrick@fpl.com January 14, 2011 Paul S. Ryerson, Esq., Chair Atomic Safety and Licensing Board Panel Mail Stop T-3 F23 U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Dr. Michael Kennedy Atomic Safety and Licensing Board Panel Mail Stop T-3 F23 U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Dr. Richard E. Wardwell Atomic Safety and Licensing Board Panel Mail Stop T-3 F23 U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 In the Matter of NextEra Energy Seabrook, LLC (Seabrook Station)

Docket No. 50-443-LR; ASLBP No. 10-906-02-LR

Dear Administrative Judges:

This proceeding involves NextEra Energy Seabrook, LLCs (NextEra) application submitted by [[letter::L-2010-101, Comment (1) of Larry Nicholson, on Behalf of Florida Power & Light Co., on Proposed NUREG-0800, Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants; Branch Technical Position (BTP) 7-19, Guidance for .|letter dated May 25, 2010]], for a renewed license for Seabrook Station, Unit 1. On October 21, 2010, Friends of the Coast and the New England Coalition (Friends/NEC) filed a Petition now before the Board, which contains four proposed contentions. This letter encloses a document containing new information potentially relevant and material to issues raised in Friends/NECs Contention 4.

A party to an NRC proceeding is obligated to keep the Board informed of relevant and material new information. Sacramento Municipal Utility District (Rancho Seco Nuclear Generating Station), CLI-93-5, 37 NRC 168, 170 (1993) (citing Duke Power Co. (William B. McGuire Nuclear Station, Units 1 and 2), ALAB-143, 6 AEC 623, 62 (1973)). The scope of this obligation extends to matters that could affect the course of the litigation, such as a change in the license application or an event that would moot or resolve some issue. Metropolitan Edison

January 14, 2011 Page 2 Co. (Three Mile Island Nuclear Station, Unit 1), ALAB-774, 19 NRC 1350, 1358 n.6 (1984).

In accordance with this obligation, NextEra encloses for the Boards consideration its letter dated January 13, 2011 (SAMA RAI Response) by which NextEra responded to a number of Requests for Additional Information (RAIs) propounded by the NRC Staff (SAMA RAI).1 The SAMA RAI Response includes new information that is relevant and material to issues raised in Friends/NEC Contention 4, as detailed below.

Friends/NECs Contention 4 challenges the Severe Accident Mitigation Alternatives (SAMA)

Analysis in NextEras Environmental Report. One of several proffered bases for this contention is Friends/NECs argument that NextEras SAMA Analysis failed to properly account for the sea breeze phenomenon. See Friends/NEC Petition at 49-50, 57. Specifically, Friends/NEC argue that NextEras model ignores the presences [sic] of sea breeze circulations which dramatically alter air flow patterns. Id. at 49. According to Friends/NEC, [s]ea breeze pulls the plume down towards the land surface increasing dose to the population. Id. See also id. at 57 (sea breeze turbulence causes the plume to be drawn down to ground level).

The NRC Staff addressed this issue in their recent SAMA RAI. Specifically, RAI 4g stated:

The ER provides no discussion of the effects of sea-breeze circulation on radionuclide deposition and whether this sea-breeze effect was factored into the MACCS2 calculations. Clarify whether sea-breeze effects were considered in the SAMA evaluation and, if not, provide an assessment of the sea-breeze effect on the results of the SAMA evaluation.

SAMA RAI at 6.

NextEra responded to the NRC Staffs request as follows:

Sea-breeze effects are considered in the SAMA evaluation to the extent they are included in the onsite meteorological data (used in the MACCS2 calculations). In response to this RAI, further consideration of sea-breeze effects at the site, as discussed below, are based on the quantitative description of sea-breeze and onshore gradient flow from the Seabrook UFSAR, Rev.12, Section 2.3, Meteorology, pages 32-37.

The two major mechanisms by which sea-breezes could affect the Level 3 MACCS2 calculations are the formation of a thermal internal boundary layer (TIBL) and a mixing front between the sea-breeze and inland winds. During the summer when the land surface is warmer than the air over the sea, air heated from contact 1 See Enclosure 1 - Letter from P. Freeman, NextEra Energy Seabrook, LLC to NRC Document Control Desk, Seabrook Station; Response to Request for Additional Information; NextEra Energy Seabrook License Renewal Application, January 13, 201. See also Letter from M. Wentzel, NRC to P. Freeman, NextEra Energy Seabrook, LLC, Request for Additional Information for the Review of the Seabrook Station License Renewal Application-SAMA Review, Nov. 16, 2010 (ADAMS Accession No. ML103090215).

January 14, 2011 Page 3 with the land rises and is undercut by cooler denser breezes from the sea. The boundary between these two distinct air layers caused by this difference in air temperature (or density) is the thermal internal boundary layer (TIBL). When these onshore breezes meet inland wind fields, which differ in magnitude and direction from the sea breeze, those differences result in something akin to a weather front with increased turbulence (mixing) along the mixing front between the two wind systems.

Mixing Front - The mixing front results in increased plume mixing and dispersion which would, in turn, result in lower population dose. The UFSAR simulates increased mixing (decreased atmospheric stability) below the TIBL by resetting site measured stability classes E through G (most stable classes) to D (less stable) during times of TIBL occurrence. This stability simulation was not implemented for the SAMA evaluation, thus resulting in a conservative evaluation (less dispersion, greater doses).

Thermal Internal Boundary Layer (TIBL) - The minimum TIBL height is noted in the UFSAR reference as 93 meters (above ground level). The Level 3 release height (top of containment) is 54.6 meters. Thus, the severe accident release plumes would be trapped beneath the TIBL, resulting in limited vertical dispersion and increased ground concentrations (doses). A TIBL was present during 527 hours0.0061 days <br />0.146 hours <br />8.713624e-4 weeks <br />2.005235e-4 months <br /> for the period April 1979 - March 1980 (the period analyzed in the UFSAR). 638 hours0.00738 days <br />0.177 hours <br />0.00105 weeks <br />2.42759e-4 months <br /> (7% of the year) of TIBL formation was found during 2005 (the met year used for the SAMA analysis) by applying the UFSARs criteria (based on time of year, time of day, wind speed, wind direction and solar radiation) for TIBL formation.

Using additional MACCS2 runs, the effect of TIBL formation was bounded by assuming a summer mixing layer (both morning and afternoon) of 100 meters, the minimum allowable by MACCS2.

Applying this restrictive lid to mixing during the entire summer (2190 hours0.0253 days <br />0.608 hours <br />0.00362 weeks <br />8.33295e-4 months <br />) bounds the 638 hours0.00738 days <br />0.177 hours <br />0.00105 weeks <br />2.42759e-4 months <br /> of TIBL formation found for 2005. The resulting population dose and offsite economic cost risks were found to increase by 4% and 7% respectively compared to the baseline SAMA case. The sensitivity of the lid height was investigated by specifying a 110 meter height. This shows that a decrease of 10 meters (from 110 to 100 meters) increases the dose and offsite cost risks by 0.2% and 0.5%. A decrease from 100 meters to 93 meters would be expected to show a similar change.

Because the release plumes are trapped beneath the TIBL, no special effect (other than the trapping itself, which is reflected in the MACCS2 runs as described above) is expected on radionuclide deposition.

January 14, 2011 Page 4 Given the conservatisms in the above analysis (no accounting for increased mixing with sea-breeze, assuming 25% of annual hours result in TIBL formation vs. 7% of annual hours demonstrated for 2005) and the conservatisms built into the baseline analysis (see RAI response 4c and especially the assumption of perpetual rainfall in the 40-50 mile ring), the potential effects of sea-breezes do not change the SAMA analysis.

SAMA RAI Response at 74-75.

NextEras response to the NRC Staffs RAI is material and relevant to an issue currently before this Board, namely, whether NextEras SAMA Analysis considers the sea breeze and its potential to dramatically alter air flow patterns and pull[] the plume down towards the land surface, increasing dose to the population. Petition at 49. Consistent with the Appeal Boards decision in Three Mile Island, NextEra provides this information to the Board because it has the potential to moot or resolve this pending issue. ALAB-774, 19 NRC at 1358 n.6.

Sincerely,

/Signed electronically by Steven Hamrick/

Steven Hamrick Counsel for NextEra Energy Seabrook, LLC Enclosure - NextEra Letter SBK-L-11001 to the NRC dated January 13, 2011

UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION Before the Atomic Safety and Licensing Board In the Matter of

)

)

NextEra Energy Seabrook, LLC

)

Docket No.

50-443-LR

)

(Seabrook Station)

)

)

ASLBP No. 10-906-02-LR (Operating License Renewal)

)

CERTIFICATE OF SERVICE I hereby certify that copies of the foregoing Board notification were provided to the Electronic Information Exchange for service to those individuals listed below and others on the service list in this proceeding, this 14th day of January, 2010.

Administrative Judge Paul S. Ryerson, Esq., Chair Atomic Safety and Licensing Board Mail Stop T-3 F23 U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Email: psr1@nrc.gov Administrative Judge Dr. Michael Kennedy Atomic Safety and Licensing Board Mail Stop T-3 F23 U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Email: michael.kennedy@nrc.gov Administrative Judge Dr. Richard E. Wardwell Atomic Safety and Licensing Board Mail Stop T-3 F23 U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Email: richard.wardwell@nrc.gov Secretary Attn: Rulemakings and Adjudications Staff Mail Stop O-16 C1 U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 hearingdocket@nrc.gov

2 Office of Commission Appellate Adjudication Mail Stop O-16 C1 U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 E-mail: OCAAMAIL@nrc.gov Mary Spencer, Esq.

Maxwell C. Smith, Esq.

Emily L. Monteith, Esq.

Megan Wright, Esq.

Office of the General Counsel Mail Stop O-15 D21 U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 E-mail: mary.baty@nrc.gov Raymond Shadis New England Coalition Post Office Box 98 Edgecomb, Maine 04556 E-mail: shadis@prexar.com Kurt Ehrenberg New Hampshire Sierra Club 40 N. Main Street Concord, NH 03301 E-mail: Kurt.Ehrenberg@sierraclub.org Paul Gunter, Reactor Oversight Project Beyond Nuclear 6930 Carroll Avenue, Suite 400 Takoma Park, MD 20912 E-mail: paul@beyondnuclear.org Doug Bogen Executive Director Seacoast Anti-Pollution League PO Box 1136 Portsmouth, NH 03802 E-mail: bogen@metrocast.net

/Signed electronically by Steven Hamrick/

Steven Hamrick U.S. Nuclear Regulatory Commission Attention: Document Control Desk One White Flint North 11555 Rockville Pike Rockville, MD 20852 Seabrook Station NEXTera" EN~~~

January 13, 2011 SBK-L-11001 '

Docket No. 50-443 Response to Request for Additional Infolmation NextEra Energy Seabrook License Renewal Application

References:

L NextEra Energy Seabrook, LLC letter SBK-L-I0077, "Seabrook Station Application for Renewed Operating License," May 25, 2010, (Accession Number MLlO 1590099)

2. NRC Lette("Request for Additional Infonnation for the Review of the Seabrook Station License Renewal Application~SAMA Review (TAC NO. ME3959) November 16,2010 (Accession Number ML103090215)

In Reference 1, NextEra Energy Seabrook, LLC (NextEra) submitted an application for a renewed facility operating license for Seabrook Station Unit 1 in accordance with the Code of Federal Regulations, Title 10, Parts 50, 51, and 54, In Reference 2, the NRC requested additional infonnationin order to complete its review of the License Renewal Application. The Enclosure contains the NextEra Energy Seabrook response to the NRC request for additional information dated November 16, 2010, There are no new or revised commitments made in this submittal.

The License Renewal Application, Appendix E, page F-6 contains a list of acronyms used in these responses. Ifthere are any questions or additional infOlmation is needed, please contactMt*. Richard R.Cliche, License Renewal Project Manager, at (603) 773-7003.

NextEra Energy Seabrook, LLC, P,O, Box 300, Lafayette Road, Seabrook, NH 03874

United States Nuclear Regulatory Commission SBK-L-llOOl / Page 2 If you have any questions regarding trus cOlTespondence, please contact Mr. Michael O'Keefe, Licensing Manager, at (603) 773-7745.

Sincerely, NextEra Energy Seabrook, LLC.

Paul Freeman Site Vice President Enclosure cc:

W.M.Dean, NRC Region I Administrator G. E. Miller, W. J. Raymond, R. A. Plasse Jr.,

M. Wentzel, NRC Project Manager, Project Directorate 1-2 NRC Resident Inspector NRC Project Manager, License Renewal NRC Project Manager, License Renewal Mr. Clrristopher M. Pope Director Homeland Security and Emergency Management New Hampsrure Department of Safety Division of Homeland Security and Emergency Management Bureau of Emergency Management 33 Hazen Drive Concord, NH 03305 John Gianusso, Jr., Nuclear Preparedness Manager

. The Commonwealth of Massachusetts Emergency Management Agency 400 Worcester Road Framingham, MA 01702-5399

United States Nuclear Regulatory Commission SBK-L-llOOl/Page3 NEXTera" ENER~~

~

SEABROOK I, PaulO. Freeman, Site Vice President of Next Era Energy Seabrook, LLC hereby affirm that the infOlmation and statements contained within are based on facts and circwnstances which are hue and accurate to the best of my knowledge and belief Sworn and Subscribed Before me this

~!,--3~_d.ay of,-_..!,--,"'-.::.'-'=-'l"'--I'i~-" 2011

/

PaulO. Freeman Site Vice President

Enclosure to SBK-L-ll001 NextEra Energy Seabrook Responses to NRC Request for Additional Information Regarding Severe Accident Mitigation Alternatives Analysis

United States Nuclear Regulatory Commission SBK-L:IIOOI/Enclosure RAJ Section 1 SAMARAIla

1)

Provide the following information regarding the Levell Probabilistic Safety Assessment (PSA) used for the Severe Accident Mitigation Altematives (SAMA) analysis:

a. Environmental Report (ER) Section F.3 states that Probabilistic Risk Assessment (PRA) model SSPSS-2006 is the model-of-record used to support the SAMA evaluation. Identify any changes to the plant (physical and procedUJal modifications) since 2006 that could have a significant impact on the results ofthe SAMA analyses, and provide a qualitative assessment of their impact on the PRA and on the results of the SAMA evaluation.

NextEra Energy Seabrook Response to SAMA RAI la There have been no major plant hardware changes or procedUJal modifications implemented at Seabrook Station that would have a significant impact on the results of the SAMA analyses since PRA model SSPSS-2006 was issued. In 2009, the PRA model SSPSS-2006 underwent a periodic update to incorporate minor plant changes including the latest plant-specific and generic infOlmation as shown below.

Plant Changes Plant specific data collection & update Emergency Core Cooling System (ECCS) Sump switchover timing Containment Building Sump strainer modification Supplementary Emergency Power System (SEPS) diesel alignment to Emergency Bus E6 Condensate Storage Tank to Demineralized Water Storage tank (CSTIDWST) connection Procedure changes as they impact plant operating state (POS) definition, operator action models, and accident sequences Operator review / input into operator action models Model Changes A number of modeling and documentation changes were made to improve the quality and usefulness of the PRA. The most significant changes include:

Incorporated electric power convolution model Expanded SO model to include condenser cooling, circulating water and condenser steam dump Page 1 of 116

United States Nuclear Regulatory Commission SBK-L-IIOOI/Enclosure Updated generic initiator frequency data using latest NUREG sources Updated generic component failure rate data using latest NUREG sources Updated generic data for LOOP & LOOP recovery using latest NUREGsources Based on the updated 2009 PRAmodel (SSPSS-2009), the Seabrook Station baseline at-power core damage frequency decreased approximately 18.7 percent, from 1.44E-05/yr to 1.17E-OS/yr. Also, there was no significant shift in the relative impOltance of initiating events or components. Based on this, it is judged that plant changes incorporated into the latest SSPSS-2009 PRA model would not have a significant impact on the overall SAMA results. The SSPSS-2009 results suggest that the SAMA benefits are reduced, thus providing additional margin between the SAMA benefits and associated costs.

Preliminary Results ofSSPSS-2010 Model The 2009 PRA model is currently in the process of being updated. This SSPSS-201 0 model is scheduled to be issued in 20 II. The most significant change to the SSPSS-20 10 model is an upgrade to the intemal flooding model to meet the latest IF requirements of the ASME PRA Standard and Regulatory Guide 1.200. Insights from the upgraded IF model indicate that Control Building flooding scenarios from postulated pipe breaks in the fire protection 6" and 4" diameter standpipe dominate the risk of intemal flooding. A modification has been proposed to reduce the risk of Control Building flooding by installing a globe valve or flow limiting orifice upstream in the fire protection system. The globe valve or orifice would be sized to effectively limit the maximum postulated break flow, yet not impact the design function of downstream hose stations during nOlIDal rue fighting activities. A Phase 2 PRA case study was perfonned to assess the potential cost-benefit of ins tali at ion of the proposed modification. The PRA case (NOCBFLD) assumes that the Control Building fire protection flooding initiators are eliminated.

The following results were estimated:

Reduction in CDF:

~25%

Reduction in LERF:

- 0.6%

Nominal Cost Benefit:

~$16IK Best Estimate Discount Rate: ~$143K Extended Period Cost Benefit: -$266K Upper Bound Cost Benefit:

~$23IK

% Reduction in OECR:

11%

Cost of SAMA: The engineering, hardware and implementation cost of this design change to install a flow limiting device is estimated at approximately $200K.

Based on this case study, the modification to install the flow limiting device to reduce the risk of Control Building flooding is judged as potentially cost-beneficial. Tllis SAMA is not aging-related and is identified as SAMA #192.

Page 2 of 116

United States Nuclear Regulatory Corrunission SBK-L-I1001/Enclosure SAMARAI1b

1)

Provide the following information regarding the Levell Probabilistic Safety Assessment (PSA) nsed for the Severe Accident Mitigation Altematives (SAMA) analysis:.

b. ER Section F.3 explains that the SAMA evaluation is determined from severe accident risk based on Levelland 2 PRA models for intemal and extemal events, including intemal floods, intemal fires, external floods, and seismic events. A table in Section F.3.1.1.2 shows the PRA model history from 1983 to 2006 for intemal and external full power events and indicates that it was a single PRA model. It is not clear from the table when internal flooding, external flooding, fire, and seismic modeling components were incorporated into the model or what were their individual contributions to the total core damage frequency (CDF) and large early release fi'equency (LERF).

Also, it is not clear from the description of model changes provided on pages F-20 through F-28 what the most significant updates between models were. Relative to these issues, provide the following:

I) The CDF and LERF contributions from internal and extemal events, including flooding, fire, and seismic hazard categories, for each PRA model update.

2) Indicate when internal flooding, external flooding, fire, and seismic modeling components were incorporated into the model.

3) Of the changes identified for each model update, identifY those changes that had the most impact on changing CDF and LERF.

NextEra Energy Seabrook Response to SAMA RAI 1b1 The CDF & LERF contributions from internal & extemal events for each PRA model update are provided in the table below. This table follows the historical designation from II'E and JPEEE where "internal" events include internal floods and "external" events include external floods, fn'es, and seismic hazards.

Page 3 of 116

United States Nuclear Regulatory Commission SBK-L-llOOI/Enclosure 2009 2006 2005 2004 2002 2001 2000 1999 1996 1993 1990 (1PEEE) 1989 (IPE) 1983 1.2E-5 59%

1.5E-5 70%

1.4E-5 68%

3.0E-5 56%

4.5E-5 56%

4.8E-5 59%

4.6E-5 58%

4.6E-5 58%

4.3E-5 49%

8.0E-5 55%

l.lE-4 55%

1.4E-4 68%

2.3E-4 80%

nla ~ LERF results not available 41%

8.lE-8 30%

1.2E-7 32%

l.lE-7 44%

I.OE-7 44%

6.8E-8 41%

5.lE-8 42%

5.lE-8 42%

5.IE-8 51%

3.7E-8 45%

1.6E-8 45%

2.2E-7 32%

nta 20%

nla NextEra Energy Seabrook Response to SAMA RAllb2 85%

15%

91%

9%

90%

10%

89%

11%

92%

8%

92%

8%

92%

8%

92%

8%

91%

9%

nta nta nta n/a nta nta nta ilIa All hazards, including internal flooding, external flooding, fire, and seismic, were included in the Seabrook Station PRA fi'orn the initial 1983 PRA and evelY subsequent update to the PRA model.

Page 4 of 116

United States Nuclear Regulatory Commission SBK-L-llOO I I Enclosure NextEra Energy Seabrook Response to SAMA RAI lb3 The following model changes had the most impact on CDP & LERF for each update from 1993 to 2009:

Basis fOr Change in CDF /i'om 1993 to 2009 Update 1993 to 1996 Update:

The CDP decreased by almost a factor of2. The most significant change quantitatively is in the modeling ofPCC, specifically the c.ornmon cause modeling of the PCC single train initiators with the opposite PCC train failure. Because these sequences dominated risk in the SSPSS-1993 model, a more realistic modeling made a significant reduction in CDP and in transient-initiator contribution to CDF. The ATWS contribution increased due to the change from 18-month to 24-month fuel cycle. Also, plant specific data tends to be better (lower mean value) than generic data distributions, which contributes to the reduction in failure rate for a number of risk important systems.

1996 to 1999 Update:

Several significant changes were made to the model. The LOCA initiator fi'equencies were updated, which generally reduced their importance. The ATWS model was updated to account for more current failure rates (reactor trip breakers, etc) and to account for the return to an 18-month fuel cycle. These changes reduced the intemal events contribution to ATWS by over two orders of magnitude. The RCP seal LOCA model and related electric power recovery models were revised to be explicit in the event tree model (rather than calculated offline). The EDG mission time for weather-related LOSPs (and other similar initiators) was changed from 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. Also, the definition of "inte111al" verses "external" events was changed so that the summary of results for "intemal" events includes weather-related LOSP (LOSPW) and internal floods. The common cause factors and mission time for system initiators (SWS and PCC) were modified based on a Peer Review comment. Operator dependencies were reviewed and changes were made to event tree rules and HEP quantification.

1999 to 2000 Update:

The primary change for the 2000 Update was the change from DOS-based RlSKMAN 9.2 (model SB99PR) to windows-based RlSKMAN 3.0 (model SB2000). This change in software allowed lower truncation limits in solving fault trees, which resulted in some slight increase in the results for SWS and PCC systems. There was no change in the CDF results.

Page 5 ofll6

United States Nuclear Regulatory Commission SBK-L-llOOI / Enclosure 2000 to 2001 Update:

The primary changes for the 2001 Update were minor changes to system initiator models. The change in CDF was not significant.

200 I to 2002 Update:

The primary changes for the 2002 Update were related to integrating the shutdown and low power risk models into an all-modes model. Changes to the full power model CDP results were not significant.

2002 to 2004 Update:

A number of changes were made in the 2004 Update, both for modeling and documentation. The most significant changes with regard to quantitative results include updates in the BRA analysis using the EPRI BRA tool; a revised fire PRA, and the addition of the SEPS diesel generator. The change in CDF was not significant.

2004 to 2005 Update:

A number of changes were made in the 2005 Update, both for modeling and documentation. The most significant changes include revision to success criteria and operator timing, revised seismic PRA, and SEPS design and modeling changes. The change in CDP was not significant.

2005 to 2006 Update:

A number of changes were made in the 2006 Update focused on the Mode 4, 5 and 6 shutdown model.

The change in CDP was not significant.

2006 to 2009 Update:

As mentioned in the response to SAMA RAIla, a number of changes were madc in the 2009 Update, both for modeling and documentation. The most significant changes include data updates (plant specific data and generic data distributions), updates to the electric power model (convolution, revised generic LOOP initiator and recovelY data), and revisions to operator action modeling. These changes resulted in a small improvement to the CDF..

Page 6 ofl16

United States Nuclear RegulatOlY Commission SBK-L-llOOII Enclosure Basis (Or Change in LERF Results trom 1993 to 2009 1993 to 1996 Update:

The LERF increased due to the change in definition of LERF to include steam leak from SGTR (early core melt), consistent with the WOG guidelines. The frequencies of the SERF (smaU-early) and INTACT categories decreased due to the decrease in CDF. The relative contribution of INTACT decreased due to the modeling changes that reduced the importance of loss of PCC initiators. The LATE category increased due to a change in how seismic sequences are modeled for failures of small containment penetrations. The seismic SBO sequences now include credit for an operator to manually close the RCP seal return line MOV. This moved the sequence frequency fi'om SERF to LATE.

1996 to 1999 Update:

The changes in the Level 2 results are due to the impact of changes to the Levell model. For example, the LERF contribution from SGTR was increased based on modeling of operator dependencies.

1999 to 2000 Update:

The primaty change for the 2000 Update was the change from DOS-based RISKMAN 9.2 (model SB99PR) to windows-based RISKMAN 3.0 (model SB2000). 111ese changes are described in the Levell update summaty, above. There was no change in the LERF results.

2000 to 2001 Update:

No changes were made to the Level 2 model. The changes in the Level 2 results are due to the impact of changes to the Level 1 model. There was no change in the LERF results.

2001 to 2002 Update:

No changes were made to the Level 2 modeL The changes in the Level 2 results are due to the impact of changes to the Level I model. The change in LERF was not significant.

Page 7 of1l6

United States Nuclear Regulatory Connnissioll SBK-L-IIOOl! Enclosure 2002 to 2004 Update:

Most of the changes in the Level 2 results are due to the impact of changes to the Levell model. The LERF model was expanded by the addition of a consequential SGTR model to SLB (steam line break) and ATWS sequences. The change in LERF was not significant.

2004 to 2005 Update:

Significant changes were made to update the Level 2 analysis to the CUlTent state of knowledge, including modeling of SAMG actions within the CET. The change in LERF was not significant.

2005 to 2006 Update:

No changes were made to the Level 2 model. The changes in the Level 2 results are due to the impact of changes to the Level I model.

2006 to 2009 Update:

The most significant changes that impact the Level 2 model include data updates (plant specific data and generic data distributions) and revisions to operator action modeling. These changes resulted in a small improvement in LERF consistent with CDF.

SAMARAllc I)

Provide the following infolTUation regarding the Levell Probabilistic Safety Assessment (PSA) used for the Severe Accident Mitigation Alternatives (SAMA) analysis:

c. The table in SectionF.3.1.1.2 that shows the PRA model history from 1983 to 2006 provides the CDF and LERF. The ratio ofLERFto CDF <<1%) is atypically small. Explain why LERF is so low compared to CDF:

NextEra Energy Seabrook Response to SAMA RAI 1c Seabrook Station has a vety large and extremely strong containment building in comparison to the design capacity (52 psig) and to other containment designs. The containment structure has a median failure pressure of 187 psia (dty) and 210 psia (wet). (Note, the tenns "dty" and "wet" refer to the status of R WST injection; the "dty" state results ifthe RWST is not injected). The most likely overpressure failure mechanism is a leak-before-break failure mode involving several small (0.5 i~ leak area) mechanical penetrations. As a result of the high containment ultimate strength, there are no conceivable severe accident progression scenarios that result in a catastrophic containment overpressure failure early in the Page 8 of 116

United States Nuclear Regulatory Commission SBK-L-llOOl/Enclosure accident sequence. The containment would eventually fail as a result of overpressure given a core melt accident without recovery of containment cooling. However, this type of failure would occur late (well more than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />) after core melt and thus does not contribute to LERF.

As a result of the strength of Seabrook Station's containment, the impoliant events that contribute the most to LERF are containment bypass and containment isolation failure. The conditional probability of these events occUlTing given core melt is approximately 0.007.

SAMARAJld

1)

Provide the following information regarding the Levell Probabilistic Safety Assessment (PSA) used for the Severe Accident Mitigation Alternatives (SAMA) analysis:

d. ER Section F.3.3 identifies two peer reviews that have been perfOlmed on the PRA: a 1999 Westinghouse Owner's Group celiification peer review and a 2005 focused peer review against the American Society of Mecham cal Engineers (ASME) standard and presents all the CategOlY A and B facts and observations along with their associated resolutions. However the scope of those peer reviews was not described and other reviews (e.g., internal reviews) were not identified.

Provide the following:

1) A summalY of the scope ofthe 2005 focused peer review against the ASME standard and the 1999 peer review including whether Levell, Level 2, internal flooding, external flooding, fire, or seismic event modeling was reviewed.

2) A summalY of the scope of any other PRA model reviews, a discussion of how each finding was resolved, and an assessment of the impact of all unresolved findings on the SAMA evaluation.

NextEra Energy Seabrool. Response to SAMA RAJ Idl The scope of 1999 peer review was a full review of all internal events and internal flood, including all teclmical elements. This included Levell and Level 2 restricted to LERF. The 2005 peer review was a focused scope, including internal events limited to Levell accident sequences, success criteria, post-initiating event HRA, and configuration control. Neither the 1999 or 2005 peer reviews included external flooding, fire or seismic hazards. A Peer Review was conducted in late 2009 focusing exclusively on internal flooding.

NextEra Energy Seabrook Response to SAMA RAJ)d2 A number of other internal reviews and vendor-assisted reviews have been performed on specific model updates. Comments from these reviews along with plant changes and potential model enhancements are tracked through a model change database to assure the comments are addressed in the periodic update Page 9 ofl16

United States Nuclear Regulatory Commission SBK-L-l1001/Enclosure process. Umesolved PAA comments from these reviews primarily reflect model completeness and documentation issues and are not significant to the results and conclusions of the PRA. As a result, umesolved PRA comments are judged not to have a significant impact on the SAMA evaluation.

SAMARAIle I)

Provide the following information regarding the Levell Probabilistic Safety Assessment (PSA) used for the Severe Accident Mitigation Alternatives (SAMA) analysis:

e. Describe the quality control process for the PRA, including the process of monitoring potential plant changes, tracking items that may lead to model changes, making model changes (including frequency for model updates), documenting changes, software quality control, independent reviews, and qualification ofPRA staff.

NextEra Energy Seabrool{ Response to SAMA RAI le Seabrook Station PRA Group instructions define the process of maintaining and updating the Seabrook Station PRA model. The process is consistent with the requirements of the ASMEfANS PRA Standard and ensures that the PRA accurately reflects the current Seabrook Station plant design, operation and performance, and that the PRA remains consistent with current risk technology and modeling. A general description ofthe configuration control process is as follows:

(a) Monitor PRA inputs for new information. This includes monitoring changes to Seabrook Station plant design and operation, monitoring Seabrook Station and industry operating experience, and changes in PRA technology and modeling.

(b) Record applicable new information. Applicable new information that has the potential to impact the PRA model is recorded in the Model Change Database (MCDB). These MCDB entries form the content of the next PRA revision. Until close-out, these records are pending changes against the PRA model of record.

(c) Assess the significance of new information. The significance of the new information is reviewed with regard to its impact on the PRA model, including cumulative impacts from pending changes.

This process identifies the need for a prompt focused PRA revision verses periodic PRA revision, and the need for PRA upgrade (with Peer Review) verses PRA maintenance.

(d) PerfOlm the PRA revision. The PRA is revised to evaluate the new information and incOlporate the model changes identified in the MCDB as appropriate. Control ofPRA revisions is provided in PRA Group Instructions. A "periodic" revision to the PRA model is peliOlmed at least once every three cycles (~4.5 years) to address open items in the MCDB as well as incorporate any changes in plant design and operations; and to reflect operating experience.

Each model change documented in the MCDB requires an independent teclmical review. The review of each model change is documented in the "Disposition of Change" field within the MCDB. The plllpose of the independent review is to verify that the model change was perfonued correctly and adequately reflects the plant or data change. The review may consist of a point-by-point check or an audit of calculations, analysis and documentation. Note that for PRA changes Page 10 of 116

United States Nuclear RegulatOly Commission SBK-L-llOOll Enclosure judged to be PRA "upgrades" (new methodology or significant change in scope or capability) a formal peer review would be required in addition to the independent tecluiical review.

The PRA model documentation is updated as applicable for each update. The Seabrook Station PRA documentation consists of three levels (or tiers). Tier"l is an Executive Summary, a high level report appropriate for plant management. Tier 2 is the comprehensive documentation of the model at a level adequate for an external reviewer to understand the basis for the risk from Seabrook Station. Tier 2 consists of the detailed systems notebooks, data notebooks, and RISKMAN model file reports for event tree rules and master frequency file. Tier 3 consists of spreadsheets, data bases, and other detailed calculations and repOlis as well as the RISKMAN computer model itself. This level is adequate for an external reviewer to be able to reproduce any of the risk results. Tier 2 and 3 comprise the controlled risk model.

(e) Control of computer codes and models. Control of computer codes and models is provided in PRA Group Instructions. These instructions provide guidance for maintaining the computer codes that form the basis of the Seabrook StationPRA and risk-infOimed applications for both vendor-provided software and in-house software. The PRA computer codes are controlled and maintained to meet requirements of the NextEra Energy corporate Software Quality Assurance Program including: classification of PRA software, identification of associated SQA requirements, and control of computer code configuration.

Seabrook Station PRA staff members have many years of plant engineering, operations and PRA experience. PRA qualification is performed as part of the Engineering SUppOlt Personnel Training Program (ESPT) for the duty area of Risk Management Engineer! Analyst Engineering.

SAMARA! If

1)

Provide the following infOimation regarding the Levell Probabilistic Safety Assessment (PSA) used for the Severe Accident Mitigation Altematives (SAMA) analysis:

f.

ER Table F.3.1.1.1-2 presents the top basic events by Risk Reduction W Olth (RR W) for the Level 1 PRA. While the contributions ofinitiating events to CDF are provided in Table F.3.1.1.1-1, no initillting events appear in the list for RR W in Table F.3.l.1.1-2. CJariJY if initiating events were included in the determination of the RRW listing and, if initiating event were not included, "

determine their RR W values and identiJY and evaluate SAMAs to address these events.

NextEra Energy Seabl'ook Response to SAMA RAI 1f The determination of basic event Risk Reduction Worth (RRW) importance presented in Table F.3.1.1.I-2 includes consideration of initiating events. Support system failures of either one or both trains are specifically modeled in the PRA and thus accounted for in the basic event importance.

It is also noted that, as presented in Appendix F.A of the SAMA repOli, a number of SAMA PRA "case studies" were perfOlmed in support of Phase II screening. Many of these SAMA case studies specifically Page 11 of 116

United States Nuclear Regulatory Conunission SBK-L-1I0011 Enclosure detennined the risk reduction when assuming complete elimination of cettain initiating events. The results of these case studies, in terms of reduction in CDF and reduction in offsite dose, are identified for applicable SAMAs in Table F.7-1, Seabrook Station Phase II SAMA Analysis.

SAMARAIlg I)

Provide the following infonnation regarding the Levell Probabilistic Safety Assessment (PSA) used for the Severe Accident Mitigation Alternatives (SAMA) analysis:

g. ER Section F.3.1.1.1 states that "The event tree quantification was calculated using a tmncation cut-off frequency ofl.OE-14." It is not clear whether this value indicates the tmncation level for the Level 1 PRA model. ClarifY what tmncation level was used for the Level 1 PRA model results used for the SAMA evaluation.

NextEra Energy Seabrook Response to SAMA RAI 19 The initiating event tmncation level used when performing the event tree sequence quantification for SAMA cases is lE-14. This value determines the CDF limit used in quantifYing event tree sequences. At Seabrook, Station this tmncation level is used as a standard tmncation level to assure that results for low frequency initiators are fully accounted.

The event tree top event nodes are quantified separately using Binary Decision Diagram (BDD) methodology, which provides an exact solution. The BDD method quantifies split fractions without solving for cutsets and, thus, tmncation is not an issue for any split fraction value.

The BDD quantification method is not used for sUPPOtt system initiators modeled by fault tree logic. For these initiators, cutsets are used in the quantification process along with selected tmncationlevel. The cutset sol ver within the Systems module in RISKMAN has two cutoff parameters - frequency cutoff and cutset order cutoff. Most initiating event systems are solved at zero frequency cutoff and 12th order cutset order (maximum value). However, several initiator-related fault trees require frequency cutoffs to limit the number of cutsets, based on RISKi\\1AN split fraction quantification limitations. These include:

Loss of Primary Component Cooling Water Train A and B (solved at lE-12, 12th order)

Loss of Service Water Train A and B (solved at IE-08, 8th order)

Loss of Offsite Power (plant-centered) (solved at lE-12, 12th order)

Loss ofRCP Seal Cooling (solved at lE-12, 12th order).

Of these, only Service Water is sensitive with regard to truncation limit. However, the related Service Water single train split fraction values are more than two orders of magnitude above the Service Water top event truncation limit, which is judged sufficient to assure meaningful results.

Page 12 of 116

United States Nuclear Regulatory Commission SBK-L-ll 00 1 I Enclosure SAMARAIlh

1)

Provide the following infOlmation regarding the Levell Probabilistic Safety Assessment (PSA) used for the Severe Accident Mitigation Altematives (SAMA) analysis:

h. Section FJ.1.1.1 explains that "The fault tree method of quantification is binary decision diagram quantification which provides an exact solution for split fractions." We understand binary decision diagram quantification to be used to evaluate Event Trees to pass along dependencies in associated fault trees. Is this what is meant?

NextEra Energy Seabrook Response to SAMA RAI Ih RISKMAN uses the large event tree (Event Tree Module), small fault tree (Systems Module) modeling approach. Fault tree linking to account for support system dependencies is not perfonned. Instead of fault tree linking, a support system dependency analysis is used to identify all dependent support system impacts on each top event - system, train or function.

Top event split -fractions are defined (within the Systems Module) based on the dependent support system impacts. As mentioned in the response to SAMA RAI 19, each top event node is quantified separate from the event tree quantification using Binary Decision Diagram (BDD) methodology. The top event split-fraction values are maintained in a database file (Master Frequency File), which is common to both the systems and event tree modules.

The Event Tree Module quantifies the event trees. Split-fraction logic lUles are used to apply the proper system/train split-fraction (obtained from the Master Frequency File) for a given sequence of events during quantification of the event trees. Thus, it is both the development ofthe system/train dependent split-fractions and the event tree logic rules that account for system dependencies, not the BDD method.

For example, if Train A of AC power is failed in the SUppOlt system tree, split fraction logic rules would guarantee failure of system/train top events in the frontline system tree that rely on Train A AC power.

SAMARAlli I)

Provide the following infOlmation regarding the Levell Probabilistic Safety Assessment (PSA) used for the Severe Accident Mitigation Alternatives (SAMA) analysis:

i.

Table FJ.1.1.1-1 listed two switchgear (SWGR) room fire frequencies as 1.0E-3/yr, which would seem low unless these were specifically for localized fires involving only the buses cited. Are only the cited buses involved with these events?

Page 13 of 116

United States Nuclear Regulatory Commission SBK-L-1I001/Enclosure NextEra Energy Seabrook Response to SAMA RAJ 1i The fire initiating event frequencies for FSGAE5 and FSGBE6 are based on the following calculations from the Seabrook PRA:

FSGAE5, Fire in SWGR Room A, Loss of Essential 4KV Bus E5: frequency = l.1E-03/yr. Fire initiation frequency for this scenario is based on the total ignition frequency for Bus E5 (21 cubicles) and other electrical cabinets (170 cabinets) located in the "A" switchgear room. These cabinets are modeled as follows:

Bus E5 cabinets are a fixed combustible and fire within the bus is assumed to stay inside the bus, no severity factor or suppression is credited, and the probability of bus damage is equal to one.

Therefore, the total fire frequency for Bus E5 is equal to (21 cabinets x 4.6E-5/yr per cabinet) =

9.7E-4/year.

Other electrical cabinets located in the switchgear room are also a fixed combustible and have a potential to raise the room temperature to 120°F, and to jeopardize operation of the various electrical components. 120°F is the maximum operating temperatme of the thermal overloads of the MCC cabinets, based on Motor Control Cabinet Specification. Based on screening results, at least 380,000 BTDs are needed to get to this temperatme in the room. Transient fire sources in this room are not significant and assumed not to have the required heat content. The ignition frequency of these cabinets is equal to (170 cabinets x 4.6E-05/yr) = 7.8E-3/year. A severity ractor of 0.2 is used based on cabinet fires in SGWR. The results from the heat-up analysis show that it would take approximately 20 minutes to reach 120°F (half of the cabinet buming time).

Thus, manual suppression is credited at 0.1 because the total fire brigade response time for SWGR-A is about 8 minutes. Therefore, the fire frequency of other cabinets impacting Bus E5 is 7.8E-03/yr x 0.2 x 0.1 = ~ 1.6E-04/yr.

Total fire frequency for impact to Bus E5 = 9.7E-04/yr + 1.6E-04/yr = I.IE-03/yr FSGBE6, Fire in SWGRRoom B, Loss ofEssential4KV Bus E6: frequency = 1.0E-03/yr. Fire ignition frequency for this scenario is based on the total ignition frequency for Bus E6 (21 cubicles) and other electrical cabinets (86 cabinets) located in the "B" switchgear room. These cabinets are modeled as follows:

Bus E6 cabinets are a fixed combustible and fire within the bus is assumed to stay inside the bus, no severity factor or suppression is credited, and the probability of bus damage is equal to one.

Therefore, the total fire frequency for Bus E6 is equal to (21 cabinets x 4.6E-5/yr per cabinet) =

9.7E-4/year.

Other electrical cabinets located in the switchgear room are also a fixed combustible and have a potential to raise the room temperature to 120°F, and to jeopardize operatiOli of the various electrical components. 120°F is the maximum operating temperature of the thermal overloads of the MCC cabinets, based on Motor Control Cabinet Specification. Based on screening results, at least 300,000 BTUs are needed to get to this temperature in the SWGR-B room. Transient fire sources in this room are not significant and assumed not to have the required heat content. The ignition frequency of these cabinets is equal to (86 cabinets x 4.6E-05/yr) = 3.9E-3fyear. A Page 14 of 116

United States Nuclear Regulatory Commission SBK-L-ll 00 1/ Enclosure severity factor of 0.2 is used based on cabinet fires in SGWR. The results from the heat-up analysis show that it would take approximately 18 minutes to reach 120°F (half of the cabinet burning time). Thus, manual suppression is credited at 0.1 because the total fire brigade response time for SWGR-A is about 13 minutes. Therefore, the fire frequency of other cabinets impacting Bus E6 is 3.9E-03/yr x 0.2 x 0.1 = ~ 7.8E-05/yr.

Total fire frequency for impact to Bus E6 = 9.7E-04/yr + 7.9E-05/yr = 1.OE-03/yr Page 15 of lJ6

United States Nuclear Regulatory Commission SBK-L-llOOI/Enclosure RAJ Section 2 SAMARAI2a

2)

Provide the following information relative to the Level 2 analysis:

a. ER Section F.3.2.1 explains that "inputs to 'the Level 2 analysis are the core damage sequences,"

and that these sequences are "considered in groups of accident sequences that exhibit similar thermal-hydraulic behavior." The ER does not identifY or discuss the use of Plant Damage States (PDSs). Describe the grouping of Level I accident sequences that provide the input to Level 2.

hlc!'ude in that discussion identification of those groups (e.g. PDSs), the attributes that define that group, and the CDF associated with each group.

NextEI'a EncI'gy Seabrook Response to SAMA RAl2a The quantification of the Level I & Level 2 models employs the linked event tree method and does not use the plant damage state (PDS) method. With the event tree linking method, the Levell event trees are linked directly to the containment event tree (CET). Therefore all Levell sequences are evaluated by the CET and it is not necessary to summarize and group similar sequences into Levell plant damage states before input to the CET.

SAMARAI2b

2)

Provide the following infOlmation relative to the Level 2 analysis:

b. ER Section F.3.2.l explains that "mapping of sequences between the Level I model and the release categories is governed by the CET [containment event tree 1" and that "containmeilt analysis covers all conceivable failure modes of the containment, including pre-existing leaks, containment bypass sequences, external events impacting the stlUcture, and internal loads that have the potential to fail the containment early (shortly after core melt) or late (many hours after the melt)." No example or actual CET is presented. Present or describe the CETs. Discuss the selection of the top events, how the branch point probabilities are determined, and the number of CETs developed for each of the four PRA model aspects (i.e., internal events, internal fire, seismic, external flooding).

NextEra Energy Seabrook Response to SAMA RAI2b The Level 2 model employs a single containment event tree (CET). The single CET is used for all PRA model aspects of internal events, internal fire and external events including seismic events.

The CET has 37 top events, which includes: 10 hardware-related, 13 human action-related and 13 phenomena-related top events along with a single mapping top event. MAAP analyses, supplemented by Page 16 of 116

United States Nuclear Regulatory Commission SBK-L-11001 fEnclosure specialized phenomena assessments are used to detennine which phenomena-based CET top events are to be included in the CET stmcture. Except for the first top event (MAPX), the CET is a branch-everywhere tree. As a result, there are a large number of potential sequences.

CET branch point split fraction numerical values are detelmined differently based on the type of top event. Hardware-related top event numerical values are determined and assigned based on systems analysis and develop.h:ient of associated system fault tree logic and associated split fractions. Human action-related top event numerical values are determined and assigned based on human reliability analysis (HRA) methodologies and. development of associated human elTor probabilities (HEP). Phenomena-related top event numerical values are determined and assigned based on review of relevant physical phenomenon for Ii given group of accident sequences (MAAP analyses or supplemental phenomena assessments).

The CET top event success criteria is defined and split fraction logic mles are used to apply the COlTect top event split fraction values during CET quantification. For example, the top event for the operator action to depressurize the RCS (top event XODPI) would not be applicable to large and medium LOCA events because the RCS is already depressurized as a result of the initiating event. Therefore, the CET mle logic would assign success (or bypass) to this operator action top event.

A description of each CET top event is provided below.

CET Mapping Top Event Top event MAPX - Sequence Mapping This is the first top event in the tree. This top event has three branches: LEVELl, LEVEL2 and SUCCESS. MAPX is a "mapping" top event designed to separate sequences into SUCCESS (stable, non-melt sequences) and failed (melt sequences). The SUCCESS branch sequences bypass the remainder of the top events in the CET. All melt sequences are mapped to both LEVELl and LEVEL2 branches. The LEVELl branch also bypasses the remainder of the CET top events. The LEVEL2 branch is mapped through the CET. This allows both Levell results (MELT) and Level 2 results (LERF, etc.) to be calculated at the same time.

CET Hardware Top Events Top Events CBSIA, CBSIB - Containment Building Spray Injection These top events model the injection mode of the Containment Building Spray (CBS) pump trains. CBS consists of two independent trains, A and B. Success of these top events requires that the associated RWST valves open, motive and control power are available, and all support systems (including a start signal) are available. CBS protects the containment from ovelpressure and scrubs the containment atmosphere of radionuc!ides. These functions also reduce the chance of containment failure should core melt occur and reduce the severity of release if the containment does fail.

Page 17 of 116

United States Nuclear Regulatory Connnission SBK-L-IIOOI JEnclosure Top Events CBSRA, CBSRB - Containment Building Spray Recirculation These top events model the recirculation mode of the CBS pump trains. Success of these top events requires that the associated CBS pump lUns, room cooling and pump cooling are available, sump valves open, with CBS injection path open. For MELT sequences, the success of these top events is useful for contaimnent fission product sClUbbing and heat removal functions.

Top Events CBSXA, CBSXB - Containment Building Spray Cooling These top events model the availability of the CBS heat exchangers (CBS-E-16A, CBS-E-16B), including opening of the Primary Component Cooling Water isolation MOVs (CC-V137, CC-V136) to the associated CBS heat exchangers. These exchangers provide cooling to the containment spray water recirculated from the contaimnent sump prior to entering the spray nozzles. The CBS pump trains must be successful for the heat exchangers to function.

Top Event CIL - Containment Isolation Large This top event models the automatic isolation of the two, 8" diameter containment on-line purge (COP) lines, and referred to as the "large" containment isolation lines. Success of this top event is defined as proper actuation and operation ( closure) of at least one valve in each of the COP supply and exhaust lines.

This top event also includes the potential for large unidentified pre-existing openings (leakage) in containment that are not automatically isolated.

Top Event CIS - Containment Isolation Small This top event models the automatic isolation of the small containment isolation lines. "Small" is defined as any line which has an inside diameter less than or equal to three inches; this definition incorporates all containment isolation lines except the COP lines (top event ClL). All contaimnent isolation lines contain isolation valves which are reqnired to automatically close on receipt of a containment isolation signal.

Success of this event is defined as the proper actuation of at least one of the two in-series valves in each line. Top events CIL and CIS are mutually exclusive.

Top Event CILK - Containment Leakage This top event models containment leakage within Technical Specification limits. For sequences with successfhl contaimnent isolation, this top event distinguishes two leakage cases - "nominal" small leakage below the Technical Specification limit and "excessive" leakage above the Technical Specification liroit but well below the size of a small penetration.

Top Event SGTI - Stearn Generator Tube Integrity This top event models the potential for SG tuhes to be intact at the time of core melt. This top event includes consideration of both SGTR initiating event sequences and overpressure sequences with the potential for a consequential SGTR.

Page 18 of1l6

United States Nuclear RegulatOly Commission SBK-L-11001! Enclosure eET Human Action Top Events Top Event OPSIG - Operator Action to Initiate "P" Signal This top event models failure of the operator to manually initiate a "P" signal prior to core melt. This is modeled for sequences that require CBS actuation, i.e., containment pressure exceeding 18 psig.

Top Events OCI1, OCI2 - Operator Action to Isolate Containment These top events model the operator actions needed to manually complete containment isolation. All containment isolation lines contain isolation valves which are required to close automatically on a containment isolation signal ('T' signal or 'P' signal). However, if these isolation signals failed to perform their function, the operator is instructed to verify containment isolation. Top event OCII models failure of the operator to manually initiate a containment isolation actuation. Top event OCI2 models failure of a local action to isolate the RCP seal return MOV for a station blackout sequence.

Top Events XODPI, XODP2 - Operator Action to Depressurize RCS These top events represent an operator action to depressurize the RCS using the Pressurizer Operated Relief Valves (PORV). The Seabrook Station PORVs are DC powered with an AC powered, normally opened, block valve upstream. Each PORV is powered from a different train of DC electric power, and opening of both PORV s is assumed required to accomplish depressurization. Top event XODPI models failure to depressurize prior to hot leg creep rupture. The act of depressurizing will reduce the potential for direct containment heating (DCH) and will essentially eliminate the potential for thermally induced steam generator tube rupture (TISGTR). Top event XODP2 models failure of the same action, but on a longer time frame, Le., prior to core relocation. Success of this action prevents a high pressure melt ejection (HPMI) event.

Top Events XRACE, XRACL - RecovelY of AC Power These top events model failure to recover AC power using the same recovery curve used to calculate recovelY of off site power (ROSP) in the Levell Transient (TRANS) tree. These operator actions model different time phases - early (prior to core relocation) and late (after vessel failure but before containment failure). Success of these recovery actions allows recovery of equipment to restore RCS and/or containment injection, as modeled by subsequent action.

Top Event XOECC - Recovmy of ECCS Injection This top event models failure to restore RCS injection via ECCS pumps after core melt but before core relocation. This action requires previous success of recovery of AC power (XRACE) to restore pump flow or ReS depressurization (XODPl) to allow RHR injection at low pressure.

Top Event XOEFW - Restore Feedwater Flow to Steam Generators This top event models failure to restore feed flow to the faulted SG given a SGTR. This action must be started prior to significant release (i.e., before core exit thermal couples reach ~1800°F).

Page 19 of 116

United States Nuclear Regulatory Commissiou SBK-L-llOOI/Euclosure Top Events XOINE, XOINL - Operator Action to Inject Water into Containment These top events model failure of the operator action to inject water into containment. Injection should occur either early (prior to core relocation) or late (after vessel failure but before containment failme).

Success ofthe early action supports ex-vessel cooling and assures that the RPV remains intact. The late action, along with late recovery of AC power (XRACL) and success of sump recirculation, allows for long term cooling of containment and assures that the containment remains intact.

Top Event XOSMP - Operator Action to Transfer Injection to Recirculation This top event models failure of the actions required to transfer ECCS or CBS from injection mode to sump recirculation mode. This action requires the sump to be functional and successful injection of the RWST inventOlY.

Top Event XOVNT - Operator Action to Vent Containment This top event models failme of the operator action to vent the containment to the environment when containment pressure exceeds 130 psia.

CET Phenomena Top Events The CET structure groups phenomena-related top events into three phases of the severe accident progression. These phases include:

EARLY - after core damage but before failure of the reactor pressme vessel (RPV)

RPV Failure - RPV failure and phenomena immediately following, to debris quench or to dry-out LATE - long term behavior, fi'om RPV failure to containment failure.

CET Top Events during lite "Earlv" Phase Top Event XHLI - Hot Leg Remains Intact (No HLCR)

This top event represents the potential failure of the RCS hot leg (the pressurizer surge line) by thermal creep rupture prior to RPV failure. Seabrook Station's hot legs are fabricated from stainless steel. Success in the preceding top event (XODP=S) precludes the possibility of hot leg failure since the differential pressure is sufficiently reduced.

Top Event XDCIV - Core Debris Cooled In-Vessel This top event represents those sequences where the accident is terminated with the core debris still in the vessel. Possible cooling mechanisms include: (a) recovery of in-vessel core cooling after the stmt of core damage; (b) heat transfer frolll the core debris through the lower head to a water-filled reactor cavity.

Page 20 ofll6

United States Nuclear Regulatory Commission SBK-L-11001/Enclosure Top Event XSGTI - SG Tubes Remain Intact (Thennally Induced Steam Generator Tube Rupture -

TISGTR)

This top event represents the possibility that the steam generator tubes experience such high temperatures that they fail from thermal creep rupture prior to hot leg or vessel failing. This top event is guaranteed successful (i.e., no TISGTR) given low RCS pressure - which can occur due to the accident sequence, success ofXODPl or XODP2, or failure ofXHLI (HLCR occurs). In addition, the secondary side must be dry (i.e., secondary cooling failure) or the tubes will never reach the required temperature.

Top Event XNH2E - No Early Hydrogen Bum (Prior to Reactor Pressure Vessel Failure)

This top event represents the possibility of a hydrogen bum prior to vessel failure. This has the potential to cause early containment failure. However, the bum also consumes hydrogen which decreases the probability of a later bum when the containment pressure might be higher.

Top Event XCONE - Containment Remains Intact Early This top event represents the possibility that an early containment failure may result from either the initial primary system blow-down forces or an early hydrogen bum. A detailed evaluation of the accident loads and containment strength shows that there are no sequences where the containment fails structurally during this early phase.

CET Top Events during the "RPV Failure" P/f((se Top Event XRPV - Reactor Pressure Vessel Remains Intact This top event represents the probability that the RPV remains intact following core melt. With successful in-vessel core debris cooling (XDCIV) or successful injection of the RWST into containment (XOINE), the RPV is highly likely to remain intact.

Top Event XNHPM - No High Pressure Melt Ejection This top event represents the possibility of a Direct Containment Heating (DCH) event occuning during the vessel blow-down. This top event serves to track when a DCH event can occur, based on previous top events. The probability of containment failure due to the pressure rise from DCH is included in Top Event XCONV. Evaluation of the pressure increase resulting from aDCH event fO,und the peak pressure to be within the capacity of the containment. This event is precluded if the RPV pressure at failure is low, the operator depressurizes using the PORV s (XODP 1 =S, XODP2=S), the hot leg fails (XHLI= F), or the vessel remains intact (XRPV=S).

Top Event XNH2V - No Hydrogen Bum at Vessel Failure This top event represents the possibility of a hydrogen bum occurring as the vessel blow-down occurs or as the debris is quenched in the cavity. Hydrogen previously contained in the oxygen depleted vessel is now available in the contaimnent where oxygen is present. The bum may occur at the vessel failure location or a global burn may occur slightly later as the hydrogen generated in the cavity is mixed with the containment atmosphere. A previous bum before vessel breach reduces the probability of this bum since the hydrogen concentrations will be lower.

Page 21 of 116

United States Nuclear Regulatory Commission SBK-L-llOOI / Enclosure Top Event XCONV - Containment Failure at Vessel Failure TIns top event represents the possibility that the contaimnent will fail simultaneously or slightly after the vessel blow-down. The additional possible pressure loads considered include those arising from the vessel blow-down, debris quench including potential steam explosion, DCH, and hydrogen bums. Not all of the possible loads may occur in a given sequence, based on previous CET top events and plant/containment conditions.

Top Event XDCXV - Core Debris Cooled Ex-Vessel This top event models the cooling of debris dispersed from the cavity at vessel failure. Tjle debris is unlikely to be cooled if the RWST is not injected.

The Seabrook Station lower compartment configuration includes a 30" high lip around the openings into the cavity. Approximately two-thirds of the RWST contents must be injected before overflowing into the cavity. There is little uncertainty regarding the coolability of debris if water is introduced before significant core-concrete interaction has begun.

CET Top Evellts durillg the "Late" Phase Top Event XSUMP - Recirc Sump Functional Post Core Melt This top event models the potential that the contaimnent recirculation sump is functional during severe accident conditions. If high pressure melt ejection occurs, it is likely that the sump functionality is challenged.

Top Event XNH2L - Late Hydrogen Bum This top event represents the possibility that the containment will experience a late hydrogen bum. The combustible gas concentration will be impacted by the extent of debris quenching and core-concrete attack. The combustible gas mixture will be primarily hydrogen since little carbon monoxide will be produced from the core attack ofthe basaltic concrete used in the Seabrook Station containment stlUcture.

No hydrogen bum would occur if the debris is cooled ex-vessel.

Top Event XCONL-Late, Large Contaimnent Failure This top event represents the possibility that the contaiIunent stlUcture may fail in the long term. Failure may be a result of a slow pressurization from an uncooled debris bed, lack of contaimnent heat removal (even if sprays operate), or a rapid pressure spike from a hydrogen bum. A detailed analysis of the loads present and the containment stlUctural strength, including the impact of a velY hot containment atmosphere, was performed to evaluate this top event.

This top event also represents the possibility that the contaimnent may fail from basemat melt-through.

This event is dependent on the outcome of Top Event XDCXV, i.e., whether the debris is cooled in cavity since a quenched debris bed will not attack the basemat. An unquenched debris bed does not necessarily penetrate the basemat since the debris may be sufficiently diluted by inclusion of the ablated concrete into the debris pool that the penetration stops.

Page 22 of 116

United States Nuclear Regulatory Commission SDK-IA 10011 Enclosure SAMARAI2c

2)

Provide the following information relative to the Level 2 analysis:

c. ER Section F.3.2.l explains that the "CET is linked directly with the Levell event trees to generate the frequencies of each release categOlY bin." Explain how release categOlY bin frequencies are determined, beginning with Level 1 accident sequence grouping and CET sequence results.

NextEI'a Energy Seabrook Response to SAMA RAI 2c Also refer to NextEra Energy Seabrook response to SAMA RAI 2a. The quantification of the Levell &

Level 2 models employs the linked event tree method, which directly links all Levell event trees to the containment event tree (CET). As a result, all Levell sequences are evaluated by the CET. The event trees used in the "at power" Levell and Level 2 Seabrook Station PRA are shown below in the Seabrook Station PRA Event Tree Table.

Seabrook Station PRA Event Tree Table Status Tree Status tree is the initial tree to define the plant operating states Hazard Tree Hazard tree models plant response to seismic and control room fire initiators Support Tree I Support tree I models availabiliiy of AC and DC power systems SUppOlt Tree 2 Support tree 2 models availabiliiy of cooling and control systems Systems Tree Systems tree models availability ofEFW and ECCS systems ATWSTree LOCA Tree SGTR Tree SLB ATWS tree models plant response to a transient without reactor trip LOCA tree models plant response to medium and large LOCA inside containment SGTR tree models response of the operator and plant safeiy systems to a SG tube rupture event SLB tree models plant ' response to a secondaty side break (feedwater and main steam)

Page 23 of 116

United States Nuclear Regulatory Commission SBK*L*1l001/Enclosure Support/System TRANS VSEQ Containment CET Transient tree models plant response to loss offeedwater, loss ofRCP seal cooling and small LOCA VSEQ tree models the response of the operator and plant safety systems to a inter-system LOCA from the RCS to the.RHR system CET models the containment response tCl core damage Each tree in the Support/System group is linked together beginning with the "Status" tree, The output from the "Systems" tree is linked directly to the frontline tree for the initiating event being evaluated, The ontput ofthe frontline tree is linked directly to the input of the CET, Each initiating event is run through all the applicable event trees including the CET to evaluate the Levell core damage frequency results and the Level 2 release category (RC) frequency results, The total frequency of each RC bin is the sum of all Level 2 sequences assigned to that bin, SAMARAI2d

2)

Provide the following infonnation relative to the Level 2 analysis:

d, ER Section F.3.2.l explains that representative Level 1 sequences are used to evaluate the thennal-hydraulic response of the core and containment in order to determine whether certain phenomena would be expected to occur and that Modular Accident Analysis Program (MAAP) 4.0.5 was used to investigate severe accident progression for the Level 2 analysis. Section F.3.2.1 does not identifY or discuss the representative MAAP cases selected for each release categOlY.

Describe the MAAP cases selected to represent each release category and the basis for their selection, and discuss how scenarios of less than dominant frequency but larger potential consequences were considered in the selection of each MAAP case.

NextEra Energy Seabrook Response to SAMA RAI 2d The Level 2 analysis was updated in 2005, The basis for the Level 2 update is documented in a Westinghouse report WCAP-16600-P Revision 0 (June 2006) and summarized in the Seabrook Station PRA. This update included the following release categories, as the interface between the Level 2 model and the Level 3 model.

Page 24 of 116

United States Nuclear Regulatory Commission SBK-L-1100l / Enclosure

. ~

~

~

~'R~je,ase

~I)escript~~n

.i -

Category *...*

LElA LARGE, EARLY - SGTR-initiated (or pressure-induced Mapped to LE I tube ruptures) core melt with NO feedflow to faulted SG (failure ofEFWfSUFP or operator does not restore flow).

LEIB LARGE, EARLY - Thennally-induced SGTR.

Mapped to LEl LE2 LARGE, EARLY - Interfacing LOCA with RHR pipe rupture (V sequence).

LE3 LARGE. EARLY - Failure of large containment penetration to isolate (COP valves).

LE4 LARGE, EARLY - Early containment failure due to Contributes 0.0.

direct containment heating or steam explosion.

SEI SMALL, EARLY - Early SGTR-initiated core melt with feed to the faulted SG.

SE2 SMALL, EARLY - Interfacing LOCA through RHR pump seals (submerged release).

SE3A SMALL, EARLY - Small containment penetration leak Mapped to SE3 that may progress to large late failure.

SE3B SMALL, EARLY - Failure of large containment Mapped to SE3 penetration to isolate, with spray injection / scrubbed release.

SE4 SMALL, EARLY - Small early leak that may progress to Contributes 0.0 large late failure.

LL3 LARGE, LATE - Vented containment.

LL4 LARGE, LATE - Long tenn containment overpressure failure.

LL5 LARGE, LATE - Basemat melt-through.

SL4 SMALL, LATE - Small containment failure.

Mapped to LL3 INTACT!

INTACT - Containment intact with less than Tech Spec Mapped to INTACT allowed leakage.

INTACT2 INTACT ~ Containment intact with greater than TecH.

Mapped to INTACT Spec allowed leakage.

Page 25 of 116

United States Nuclear Regulatory Commission SBK-L-IIOOII Enclosure Note that in a few cases, a single source term group is defined for two release categories. This is a case where the release categories occur due to different phenomena but the consequence is essentially the same. For example, source term group LE-I represents SGTR-type releases; release categories LE-la and LE-lb represent tube ruptures prior to core melt (LE-la) or following core melt, induced by the high temperature gas (LE-l b). Release categOly is LE-la is much more likely to occur than LE-l b, so the source term is based on LE-l a type sequences. The Level 2 release category bins were combined to simplifY and broaden the Level 2 binning as follows:

1. Combine LE-l A and LE-lB into one bin LE-l Bin LE-IA captured sequences associated with STGR initiating events with no feedwater to the mptured SG; the LE-IA sequences are ~ lE-07/yr. LE-lB captured sequences involving thennally-induced SGTR (SGTR as a result of accident progression); the LE-lB sequences are ~ lE-ll1yr. Given the velY low frequency ofLE-lB, both bins are combined into single large/early bin with source tenn based on LE-IA.

2. LE-4 contributes 0.0 to frequency Bin LE-4 was used to capture containment failure resulting from direct containment heating or steam explosion. Given the reactor/cavity design at Seabrook, DCH is not a credible containment challenge because the peak pressure associated with DCH is within the capability of containment. The LE-4 bin sequence total is 0.0, thus LE-4 does not contribute.

J. Combine SE-3A and SE-3B into one bin SE-3 Bins SE-3A and SE-3B are both containment penetration failures and were used to discriminate between a small penetration failure with no source teun scmbbing (Bin 3A) and a large penetration failure with the possibility of successful scrubbing with containment building spray (Bin 3B). The 3A bin sequence total is~IE-06 and the 3B bin sequence total is~2E-l 1. These bins are combined and source telm associated with SE-3A is used.

4. SE-4 contributes 0.0 to frequency Bin 8E-4 was used to capture sequences associated with a small early containment leak that could progress into a large, late containment failure. The 8E-4 bin sequence total is 0.0; thus, 8E-4 does not contribute.

5. Combine LL-4A, LL-4B and 8L-4 into one bin LL-4 Bins 4A and 4B are used to discriminate the timing and potential large source terms between long telm containment failure when dlY (4A) (oase mat failure) and containment failure when wet (4B)

(overpressure failure). The source term and timing from MAAP case lOla (representing 4B) are worse than from MAAP case 10 1 e (representing 4A); thus source term fi*oml 01 a are used in the combined bin LL-4 as the representative source term. Bin 8L-4 is used to capture sequences that result in similar late containment failures but with small release. This bin is added to LL-4 to simplifY the bins. TIlls provides a conservative result to LL-4.

Page 26 of 116

United States Nuclear Regulatory Commission SBK-L-1I001 / Enclosure

6. Combine INTACTl and INTACT2 into one bin INTACT.

INTA CTl and INT A CT2 are used to discriminate between successful contairunent isolation with nominal leakage - less than Tech. Spec. leakage (INTACTl) and successful contairunent isolation but with excessive leakage - greater than Tech. Spec. leakage but less than the leakage associated with a small penetr~tion (INTACT2). The INTACT2 frequency is approximately two orders of magnitude less than INTACTl, but assumes a source tenll 10 times greater than INTACT!. INTACT 1 and INT ACT2 were aualyzed separately to account for each. For reporting purposes they were combined to INTACT.

The following table shows the map between release categories and source telm IDs.

SAMA Sotirce Description Related Release Terll1 Group.. i Categorie~

. LE-!

Early Large Coniainment Bypass - Tube Rupture LE-Ia, LE-Ib LE-2 Early Large ContailUnent Bypass - ISLOCA LE-2 LE-3 Unisolated Containment (Large)

LE-3 SE-I Early Small Containment Bypass SE-I SE-2 Early Small Containment Bypass SE-2 SE-3 Unisolated Containment (Small)

SE-3a, SE-3b LL-3 Containment Venting LL-3 LL-4 Late Containment Failure - Overpressurization by LL-4a, LL-4b Noncondensible Gases or Steam LL-5 Basemat Failure LL-5 INTACT Nominal!Excessive Contaimnent Leakage INTACT-I, INTACT-2 A series ofMAAP cases are used to SUppOlt Level 2 success criteria and timing. These cases were lUn using MAAP Version 4.0.5 and a Seabrook Station-specific parameter file. The MAAP cases described below were used to assess the source term releases for categories LEI, SE-l, LL-3, LL-4 and INTACT.

Refer to RAI 4e for discussion of other release categories.

The 2009 PRA model is currently in the process of being updated and is scheduled to be issued in2011.

The scope of the PRA update will include a reassessment of source tenns along with update of other PRA elements, e.g., update to pre-initiator (latent) human actions. Based on preliminmy update results, including more current source telms, 110 changes to the overall SAMA conclusions are expected. An update to the SAMA evaluation will be provided in the next ER annual report.

Page 27 of 116

United States Nuclear Regulatory Commission SBK-L-llOOl/Enclosure LE-I - MAAP Case 103d: LE-I captures reieases involving SOTR sequences. SO tube rupture events were evaluated to judge whether the fission product releases are early or late and whether covering the tube rupture location with water by initiating EFW to the ruptured SO may provide scrubbing to reduce the fission product releases. Seven MAAP case scenarios were evaluated. MAAP Case I 03d was used to define the event timing and release assuming all EFW failed and no benefit of opening PORVs to de-pressurize RCS. Tins case represents the situation where a contaimnent bypass is via a failed SO tube and the RSC is not depressurized to containment. The source tenn from a thelmally-induced SOTR was not determined in the MAAP analyses because no TI-SOTRs were predicted. However, the source term would not be expected to be significantly different since the release from the core goes straight to the enviromnent in both cases with very little 0ppOliunity for retention. Core overheating occurs at ~3 hours with vessel failure at ~4.7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> as opposed to over 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> for cases with EFW. The core heatup plateau is at ~5 hours with release of ~.3% ofthe CsI core inventory.

SE-I - MAAP Case 103g: SE-I captures small, early release involving contaimnent bypass. Case 103g is similar to case 103d and considers the impact of pressurizer PORV opening. Opening ofPORVs delays, but does not prevent RPV failure since there is no cooling of the debris by either internal or external cooling. The MAAP analyses show that the source telm is significantly reduced if the SO can be reflooded to provide scrubbing of releases. The volatile fission product release starts at ~3 hours as the core overheats and plateaus at about 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />, after the PORVs are open, with -0.5% of the core inventOlY of CsI. This release fraction is almost an order of magnitude lower than the case without the PORV opening.

LL MAAP Case 107b: LL-3 captures large, late releases involving contaimnent venting during medium or large LOCA sequences. Containment venting is represented in five MAAP cases. MAAP Case 107b is the case chosen to define the realistic timing and release and is judged as the median source term case from the 107 series cases. In this case, the contaimnent vent pressure setpoint is reached at 35 hours4.050926e-4 days <br />0.00972 hours <br />5.787037e-5 weeks <br />1.33175e-5 months <br />, which is slightly longer than other cases with assumed auto actuation of contaimnent spray. This case is a dry cavity sequence - no injection of RWST water and significant CCI. Two separate contaimnent ventings are performed in the 100 hour0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> simulation time. The CsI release is ~0.28% of the core inventOlY at 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br />.

LL MAAP Case lOla: LL-4 captures late contaimnent failure due to overpressurization by noncondensible gases or steam. Seven MAAP case scenarios were evaluated. MAAP Case 101 a is the case used to define the realistic timing and release. A small LOCA event with limited ECCS and EFW are assumed. In this case, RPV failure occurs at ~2.9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br />. The SI pmnp injects a small amount of RWST inventory at ~0.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> as the primalY pressure drops. ECC is assumed to be insufficient to prevent core melt and vessel failure. The RWST is assumed completely injected to contaimnent after RPV failure by the Sl pumps. Contaimnent overpressure failure occurs at 37 hours4.282407e-4 days <br />0.0103 hours <br />6.117725e-5 weeks <br />1.40785e-5 months <br /> due to heat transfer from the top of the core debris to the overlying pool. The pool eventually dries out after contaimnent failure and CCl is re-initiated. Also in this case, the CsI fi'action of the core inventory released is ~35%.

This is due to the presence of two openings in the RCS after vessel failure, which pelmits natural circulation through the RCS and the efficient transfer to the contaimnent of fission products.

INTACT - MAAP Case 102k:

INTACT captures releases involving sequences with an intact contaimnent with nominal Tech. Spec. leakage (INTACTl) and excessive leakage (INTACT2). Ten MAAP case scenarios were evaluated using transient -loss offeedwater events. INTACT! - MAAP Case 102k is the case chosen to define the realistic timing and release for nominal contaimnent leakage.

Page 28 of 11 6

United States Nuclear Regulatory Commission SBK-L-llOOl/ Enclosure ill this case, contaimnent spray (CBS) was started when core exit thermal couples were at 11 OOF and the RPV was externally submerged at -2.8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />. The RPV failed at 4.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> due to ejection of an in-core tube. CBS recirculation was effective in preventing containment pressurization to the failure pressure and scrubbing the fission product release. Ex-vessel cooling was not effective due to the high RCS pressure.

JNTACTI - excessive contaimnent leakage is assumed to be ten times the nominal Tech. Spec. leakage, which translates to 10 times the release from cases with nominal Tech. Spec. leakage.

SAMARAI2e

2)

Provide the following infotmation relative to the Level 2 analysis:

e. ER Table F.3.2.1-2 presents the top basic events by RRW for the Leve12 PRA basic events that contribute to a large early release frequency (LERF). While the contributions of initiating events to CDF are provided in TableF.3.1.1.l-1, no initiating events appear in the list for RRW in Table F.3.2.l-2. Relative to Table F.3.2.l-2, address why there are no initiating events in this list.

ClarifY ifinitiating events were included in the determination of the RR W listing and, ifinitiating events were not included, determine their RR W values and identifY and evaluate SAMAs to address these events.

NextEra Energy Seabrook Response to SAMA RAJ 2e The determination of basic event RR W impotlance presented in Table F.3.2.l-2 includes consideration of initiating events. It is also noted that, as presented in Appendix F.A of the SAMA report, a number of SAMA PRA "case studies" were perfotmed in support of Phase II screening. Many of these SAMA case studies specifically determined the risk reduction when assuming complete elimination of cellain initiating events. The results of these case studies, in terms of reduction in CDF and reduction in offsite dose, are identified for applicable SAMAs in Table F.7-1, Seabrook Station Phase II SAMA Analysis.

SAMARAI2f

2)

Provide the following information relative to the Level 2 analysis:

f.

Relative to Table FJ.2.1-2, provide a listing of the risk impollant basic events contributing to the other release categories (e.g., LL3, SE3) that contribute 90% of the population dose-risk. ldentifY and evaluate SAMAs to address these events.

Page 29 of 116

United States Nuclear RegulatOly Commission SBK-L-IIOOII Enclosure NextEra Energy Seabrook Response to SAMA RAJ 2f The top ranked release categories that contribute approximately 90% to the population dose-risk include release categories SE3 (40.2%), LL3 (27.8%), LEI (9.4%), SEI (7.5%) and LL4 (6%). The top 15 basic events that contribute to each of these release categories is provided below along with a SAMA disposition. Based on assessment of these basic events, one new SAMA candidate, related to contaimnent isolation valve basic event CSV 167.FO, is identified as "potentially cost-beneficial". CS-V-167 is an MOV that *provides the outboard contaimnent isolation function of the RCP seal water retmn line. A Phase 2 PRA case study (V167AC) was performed to assess the potential cost-benefit of the concept to eliminate the valve's AC power dependence via possible installation of an AOV. The following results were estimated:

Nominal Cost Benefit:

~$I92K Best Estimate Discount Rate: -$171K Extended Period Cost Benefit: ~$240K Upper Bound Cost Benefit: -$365K

% Reduction in OECR:

-44%

Cost of SAMA: The engineering, hardware and implementation cost of this change is estimated at approximately $250K to $300K.

Based on this case study, the concept to modify the isolation design of the seal water return line is judged as potentially cost-beneficial. This SAMA is not aging-related and is identified as SAMA # 193.

Page 30 of 116

United States Nuclear Regulatory Commission SBK-L-IIOOI/Enclosure 1 ~!W1 'm~C: :9' ;i'..,

.........,., *. )i'.*...*..*..*. ***i.; "

~ \\.. /. " r:,;tij,*,i;2.,...*. *.*f;,~~ii~id'E~enfl"(

11n~til?~~,i;. Eve~t

,'1 ~ ~,, _

,

RC-".

, ;::,:: 'Y

.:Specific::"

\\,(ji~ii;::;T... "

< RRW '

1. 1 DG-IA fails to run for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> 1.096 DGDGIAFR3

1. 2 DG-IB fails to run for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> 1.076

. DGDG IB.FR3

1. 3 Small pre-existing unidentified containment leakage 1.044 ZZ.ClS,PRE,EXIST

1. 4 Operator fails to close SEPS breaker from MCB, 1.028 HH.OSEP2Q.FA given Sl Signal Top Event OSEP

1. 5 Turbine Driven l"!mp FW-P-37A fails to run 1.028 FWP37A.FR

1. 6 Top Event CIS, Pen. X-37 isolation valve CS-V-167 1.025 CSVI67.FO

"Y;r;sid" >...... '.

. ' ;'!' : '<i';:;

.. a

~o.~".,.'./'. "

';"" :"', '

~"AsS~ciatef,

,:"~

::,j ',":' '~-'~-~ -

.1 ~'!o-I!r';~;

SAMA,"

PRA.C~se ':

As'sessment

~Disp~sjtion

':' ?:':'~

1. 9, #10, #14, NOSBO

$155K (nominal benefit)

Not cost beneficial based on

1. 155

$295K CUB benefit) assumed elimination of all OECRRed: -14%

EDG failures Min, Cost: >$500K

1. 9, #10, #14, NOSBO

$155K (nominal,benefit)

Not cost beneficial based on

1. 155

$295K (UB benefit) assumed elimination of all OECRRed: -14%

EDGfailures Min. Cost: >$500K Hardware or ClSPRE

$IIK (nominal benefit)

Not cost beneficial based on procedural

$20.3K CUB benefit) assumed elimination of small change to OECR Red: -2%

pre-existing leakage failure eliminate or Min. Cost: >$50 to contribution reduce lOOK likelihood of small pre-existing unidentified leakage Hardware OSEPALL

$32.5K (nominal benefit)

Not cost beneficial based on change for

$62K CUB benefit) assumed elimination of all auto closure OECR Red: -3%

operator actions needed to ofSEPS Min. Cost: >$IOOK close SEPS breaker breaker to eliminate operator action

1. 163 TDAFW

$IOOK (nominal bcnefit)

Not cost beneficial based on S190KCUB benefit) assumed elimination of all OECRRed: -7%

TDEFW pump failures Min. Cost: >S250K Hardware Vl67AC

$198K (nominal benefit)

Potentially cost beneficial change to

$376K CUB benefit) based on assumed eliminate OECR Red: -46%

replacement ofMOV with MOVAC Min. Cost: >$300K FCAOV power dependence Page 31 of 116

United States Nuclear Regulatory Commission SBK-L-llOOl/Enclosure

.**.*!>ii,

';ro~:t

"",.... " i:.-'. "..

'-' f;:~;.;d~:' (:-: -.

i>~~~iPtiO~ C,? 'f

~:

c

,,~: ;' Basic-Event 1:,:,'0':;'." "

1. 7 Operator fails to close SEPS breaker from MCB HFLOSEPIQ,FA given seismic event

1. 8 I-SEPS-DG-2-A falls to run within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> SEPSDG2AFR3

1. 9 I-SEPS-DG-2-B fails to run within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> SEPSDG2B,FR3

1. 10 Operator fails to close containment isolation valve HFLOC12Q.FL CSV-167 locally given medium seismic event

1. 11 DG-IA Engine Driven Lube Oil Pump fails to run DGP115AFS

1. 12 DG-IA fails to starr on demand DGDGIAFS

_ *.,,,c ___

-,RC-"'>

.' AssoCIated '"

.- SpecifiC'.:,,.,sAMA RRW*

1.016 Hardware change for auto closure ofSEPS breaker to eliminate operator action 1.014

1. 9, #14, Elimination of all potential forSEPS failure 1.014

1. 9, #14, Elimination of all potential for SEPS failure LOIO Provide a hardware modification (additional signals or remote capability) to automatically close V-167.

LOIO Provide hardware modification to improve lube oil pump reliability_

1.009

1. 9, #10, #14,

1. 155 Page 32 of1l6

SEl-" " ",

,,',' Z2I

':"'-tt/~

--;-:-'.:~:r.-~--

_:i.X*,J _,:::.,_:'t-*.~~:~)*

"PRACase Assessment '

..-;' Disposition.:-'--* -'

OSEPALL

$32.5K (nominal benefit)

Not cost beneficial based on

$62K (UB benefit) assumed elimination of all OECRRed: -3%

operator actions needed to Min, Cost: >$IOOK close SEPS breaker SEPES S39,8K (nominal benefit)

Not cost beneficial based on

$75,8K (UB benefit) assumed elimination of all OECR Red: -2%

SEPS failures Min. Cost: >$IOOK SEPES

$39_8K (nominal benefit)

Not cost beneficial based on

$75,8K (UB benefit) assumed elimination of all OECRRed: -2%

SEPS failures Min_ Cost: >$IOOK Oel2S

$2.6K (nominal benefit)

Not cost beneficial based on

$5K (lIB benefit) assumed complete OECRRed: -1%

elimination of human action Min_ Cost: >$1 OOK DGP115AIB

$9K (nominal benefit)

Not cost b~neficia1 based on Sl7K (UB benefit) assumed eHmination of OECR Red: -2%

DGPl15AIB basic event Min_ Cost: >$IOOK RRW SE3, LL3, SEl, LL4 failure contribution NOSBO Sl55K (nominal benefit)

Not cost beneficial based on

$295K (UB benefit) assumed elimination of all OECR Red: -14%

EDG failures Min. Cost: >$SOOK

United States Nuclear Regulatory Conunission

. SBK-L-IIOOli Enclosure

1. 13 DGDGIAFRI

1. 14 DG-IA Engine Driven Lube Oil Pump fails to run DGPI15B.FS

1. 15 DG-IB fails to start on demand DGDGIB.FS

1. 9, #10, #14,

1. 155 1.008 Provide hardware modification to improve lube oil pump reliabili 1.007

1. 9, #10, #14,

1. 155 Page 33 of 116

$155K (nominal benefit)

$295K (UB qenefit) assumed elimination of all OECRRed: -14%

EDGfailures Min. Cost: >$500K DGPIl5AIB

$9K (nominal benefit)

Not cost beneficial based on

$ I 7K (UB benefit) assumed elimination of OECRRed: -2%

DGP II 5AIB basic event Min. Cost: >$ lOOK RRW SE3, LL3, SEI, LL4 failure contribution NOSBO

$155K (nominal benefit)

Not cost beneficiai based on

$295K (UB benefit) assumed elimination of all OECRRed: -14%

EDG failures Min. Cost: >$500K

United States Nuclear Regulatory Commission SBK-L-llOOI / Enclosure

1. 2 4KV Bus E6 fault (IE)

XX.EDESWG6.FX

1. 3 PCC Train B Temperature Element CC-TE-2271 CCTE227I.FZ transmits false low

1. 4 PCC Train A Temperature Element CC-TE-2171 CCTE2171.FZ transmits false low

1. 5 Turbine Driven Pump FW-P-37A fails to run FWP37A.FR

1. 6 Loss of Offsite Power subsequent to Plant Trip ZZSYl.FX Improve Bus E5 reliability, eliminate potential for bus fault 1.113 Improve Bus E6 reliability, eliminate all potential for bus fault 1.075 ImprovePCC TE reliability, eliminate potential for temp. element failure 1.074 lmprovePCC TE reliability, eliminate potential for temp. element fuilure 1.064

1. 163 1.058

1. 13, #156,

1. 160 Page 34 of1l6 assumed elimination of all risk associated with Bus fault, case E6S representative E6S

$39K (nominal benefit)

Not cost beneficial based on

$74.1K (DB benefit) assumed elimination of all OECR Red: -2%

risk associated with Bus fault Min. Cost: > lOOK PCCTS

$29K (nominal benefit)

Not cost beneficial based on

$55.3K (DB benefit) assumed elimination of all OECR Red: -3%

risk associated with TE-2171 Min. Cost: > lOOK and TE-227 1 failures PCCTS

$29K (nominal benefit)

Not cost b~neficial based on

$55.3K (DB benefit) assumed elimination of all OECR Red: -3%

risk aSsociated with TE-217\\

Min. Cost: > lOOK and TE-2271 failures IDAFW

$IOOK (nominal benefit)

Not cost beneficial based on

$190K (DB benefit) assumed elImination of all OECRRed: -7%

IDEFW pump fuilures Min. Cost: >$250K NOLOSP

$335K (nominal benefit)

Not cost beneficial based on S638K (DB benefit) assumed elimination of all OECR Red: -43%

risk associated with loss of Min. Cost: >lM offsite power

United States Nuclear Regulatory Commission SBK-L-l 1001/ Enclosure

ru~!m;mi<'.:' ;,, '. 'i.....

'.': >'.........*."C'i:1!'.,..,-"-..2{/...... < ~;'i(l"si~E~~ntS~ont;ibtti~i td R.2'i~3.

.i.." *...

52

~

1~~)J~!~asic~ve~t

':' ~'"

. ~L*

'c*" *.

," ""*.RC-.

I

,I "'PRA Ca~e";'"

, :':..SpeCific

.. Associated Assessment

• rA*:;

RRW,..

SAMA

~.

1. 7 Operator aligns containment sump recirculation after 1.056 Provide a XOSMPS

$21K.(nominal benefit)

HlLXOSMPI.FA core melt hardware

$39.7K (VB benefit) modification OECR Red: -5%

for auto-Min. Cost: >$100K control.

eliminate operator action to align sump after core melt

1. 8 Train B HX E-17B Excessive Leakage During 1.052 ImprovePCC PCCLS

$21.8K (nominal benefit)

CCEl7B.GL Operation HtEx

$41.6K (UB benefit) reliability, OECR Red: -3%

eliminate Min. Cost: > lOOK potential for heat exchanger leaka..,&e

1. 9 Loss of Offsite Power subsequent to LOCA initiator 1.052

1. 13, #156, NOLOSP

$335K (nominal benefit)

ZZ.SY2.FX

1. 160

$638K (lJB benefit)

OECR Red: -43%

Min. Cost: > 1M

1. 10 Train A HX E-17 A Excessive Leakage During 1.051 Improve PCC PceLS

$21.8K (nominal benefit)

CCE17AGL Operation HtEx

$41.6K (UB benefit) reliability, OECR Red: -3%

eliminate Min. Cost: > lOOK potential for heat exchanger leakage

1. 11 DC Power Panel lIB fails to operate 1.048 Improve Bus E6S

$39K (nominal benefit)

EDESWGI1B.FX Top Event DCP, CCF Group BUSFL EI1B

$74.1K (lJB benefit) reliability, OECR Red: -2%

eliminate bus Min. Cost: > lOOK failure Page 35 of 116

&:"" ;~'xi '..

-=-. ~

'~*i.*.

Dispos"ifion Not cost beneficial based on assumed' elimination of human action Not cost beneficial based on assumed elirnination of all risk associated with heat exchanger E 17 A and E 17B leakage Not cost beneficial based on assumed elimination of all risk associated with loss of offsite power Not cost beneficial based on assumed elimination of all risk associated with heat exchanger E17A and EI7B leakage Not cost beneficial based on assumed elimination of SWG II AlB basic event, case E6S representative

United States Nuclear Regulatory Commission SBK-L-llOOl / Enclosure

1. 13 SWV5.FO

1. 14 DGDGIAFR3

1. 15 EDESWG6.FX SW Secondary Isolation MOV SW-V-5 fails to close 1.038 On demand DG-1A fails to run for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> 1.034 4KV Bus E6 Fault 1.031 reliability, eliminate bus failure Improve SWV-5 reliability, eliminate valve failure

1. 9, #10, #14,

1. 155 Improve Bus E6 reliability, eliminate potential for bus fault Page 36 of 116 SWV5 NOSBO E6S

$8K (nominal benefit)

$16K (UB benefit)

OECRRed: -1%

Min. Cost: >$lOOK

$I55K (nominal benefit)

$295K (UB benefit)

OECR Red: -14%

Min. Cost: >$500K

$39K (nominal benefit)

S74.1K (DB benefit)

OECRRed: -2%

Min. Cost: >1 OOK Not cost beneficial based on assumed elimination of SWG lINE basic event, case E6S representative Not cost beneficial based on assumed elimination of SWV5 basic event RRW LL3, LL4, SEI failure contribution Not cost beneficial based on assumed elimination of all EDG failures Not cost beneficial based on assumed elimination of all risk associated with Bus fault

United States Nuclear Regulatory Commission SBK-L-ll001! Enclosure

1. 1 FWP37A.FR

1. 2 HH.xOEFWI.FA

1. 3 ZZ.SY2.FX

1. 4 FWVI56.FC

1. 5 FWVI63.FC

1. 6 FWP37B.FS Turbine Driven Pump FW-P-37A fails to run 1.445 Operator establishes feed flow to faulted SG 1.197 Loss of Off site Power subsequent to LOCA initiator 1.187 SUFP to EFW HeaderMOVFW-V-156 fails to open 1.124 on demand SUFP to EFW Header MOV FW-V-163 fails to open 1.124 on demand EFW motor-driven pump FW-P-37-B fails to start on 1.096 demand

1. 163 Hardware for automatic feed

flow, eliminate potential for operator failure to feed SG

1. 13, #156,

1. 160 Improve reliability of

SUFP, eliminate potential for SUFP failures Improve reliability of

SUFP, eliminate potential for SUFP failures Hardware change to eliminate or reduce mechanical failu;es of MDEFW pump.

Page 37 of 116 IDAFW XOEFW NOLOSP SUFPS SUFPS MEFWS

$IOOK (nominal benefit)

Not cost beneficial based on

$190K (UB benefit) assumed elimination of all OECR Red: -7%

IDEFW pump failures Min. Cost: >$250K

$4.lK (nominal benefit)

Not cost beneficial based on

$7.8K (UB benefit)

OECR Red: -1%

Min. Cost: >IOOK

$335K (nominal benefit)

$638K (UB benefit)

OECR Red: --43%

Min. Cost: >1/'>1

$42.4K (nominal benefit)

$81K CUB benefit)

OECR Red: --4%

Min. Cost: > lOOK

$42.4K (nominal benefit)

$81K CUB benefit)

OECR Red: --4%

Min. Cost: > lOOK

$38.5K (nominal benefit)

$73.4 CUB benefit)

OECR Red: -3%

Min. Cost: >IOOK assumed elimination of operator failure to feed SG Not cost beneficial based on assumed elimination of all risk associated with loss of offsite power Not cost beneficial based on assumed elimination of all risk associated with Bus fault Not cost beneficial based on assumed elimination of all risk associated with Bus fault Not cost beneficial based on assumed elimination of all risk associated with MD EFW pump failures

United States Nuclear Regulatol)' Commission SBK-L-llOOl I Enclosure

-' *:" '~" ;W"B~si~.EveDt.

.' "-; ' ~

HH.OTSI3.F A given successful ceoldown and depressurization for SGTR.

fujls to I 1.083 HR.ORWMZI.FA ZZ.2PORV.NOCRl ATWS - Unfavorable Exposure Time (VET)

Probabili1y, given 2 PORVs & 3 SVs available, wlo Control rod insertion (CRI) 1.065 1.056 improve reliability of SGTR control, eliminate or reduce operator failure to terminate S1

1. 163 Implement hardware change for automatic ECC control, eliminate or reduce operator failure to min.

ECCflow Improve hardware!

procedures to reduce or eliminate BE exposure probabili1y, improve eRl Page 38 of 116 J

~,.

d' I.

"'.',, 1

'RAo*Case' ::.<

DisDosition -'.;c- 'co

. ~~>:;:

1\\:'* '

OTSIS

$28.3K (nominal benefit)

$53.8K (UB benefit) assumed elimination of the OECRRed: -5%

human failure risk including Min. Cost: >$IOOK actions OTSI3, OTSI4 and OTSI5 for SI termination duringSGTR TDAFW

$ lOOK (nominal benefit)

Not coSt beneficial based on

$190K (UB benefit) asswned elimination of all OECRRed: -7%

TDEFW pump failures ORWS

$31.~K (nominal benefit)

Not cost beneficial based on

$60.6K (UB benefit) assumed elimination of the OECR Red: -5%

human failure risk Min. Cost: >$IOOK UET

$3K (nominal benefit)

Not cost beneficial based on

$6K (UB benefit) assumed elimination ofUET OECRRed: <1%

basic event RRW LEI, LL4 Min. Cost: >$50K probability contribution

United States Nuclear Regulatory Commission SBK-L-llOOI / Enclosure

1. 11 HROSUFP2.FA

1. 12 FWPI13.FS

1. 13 FWP37B,FR

1. 14 FWPCV4326.FC

• RC-

• . Assod~ted.* ' :'...*. ",j:.... *.

"Description Specific" SAMA.,

. PRA'C.se."

'. Assessment RRW '

Operator fails to start SUFP given SI initiator Startup Feed Pump FW-PII3 fails to start on demand EFW motor-driven pump FW-P-37-B fails to run I-FW-PCV-4326 SUFP recirculation fails to open on demand 1.054 Provide auto-OSUFPS

$6,6K (nominal benefit) startofSUFP, SI2.7K(UBbenefit) 1.053 1.044 1.040 eliminate potential for operator failure to start SUFP

]mprove SUFP reliability, eliminate potential for SUFP mechanical failures Hardware change to improve pump reliability, eliminate or reduce mechanical failures of MDEFW pumP Hardware change to improve SUFP reliability, eliminate or reduce potential for SUFP/valve failures Page390f1l6 SUFPS MEFWS SUFPS OECRRed: <1%

Min. Cost: > lOOK S42,4K (nominal benefit)

$81K (UB benefit)

OECR Red: --4%

Min. Cost: > l OOK

$38.5K (nominal benefit)

$73,4 (UB benefit)

OECR Red: -3%

Min. Cost: >1 OOK S42AK (nominal benefit)

$81K (UB benefit)

OECR Red: -4%

Min.Cost:>lOOK

. ' ""i:'[,I:'

DispOSition Not cost beneficial based on assumed elimination of all risk associated with manual action to start the SUFP Not cost beneficial based on assumed elimination of all risk associated with Bus fault Not cost beneficial based on assumed elimination of all risk associated with MD EFW pump failures Not cost beneficial based on assumed elimination of all risk associated with Bus fault

United States Nuclear Regulatory Commission SBK-L-1I001 1 Enclosure

!i'j'~~\\~irl;..~~~t recirculatlon MOV FWV347.0P open (flow diversion) change to improve reliability, eliminate or reduce mechanical failures of MDEFW Page 40 of 116

$38.5K (nominal benefit)

$73.4 CUB benefit)

OECR Red: -3%

Min. Cost: >1 OOK assumed elimination of all risk associated with MD EFW pump failures

United States Nuclear Regulatory Commission SBK-L-llOOI/Enclosure

1. 2 HH.ORWMZl.FA

1. 3 ZZ.SY2.FX

1. 4 HH.ODDSGI.FA Operator minimizes BeeS flow w/recirculation failure Loss of OffSite Power subsequent to LOCA initiator Operator fails to diagnose SG Tube Rupture Event 1.570 1.465 1.096 automatic

control, eliminate or reduce operator failure to tenninate sr Implement hardware change for automatic ECC control, eliminate or reduce operator failure to min.

ECCflow

1. 13, #156,

1. 160 Implement hardware change to eliminate or reduce operator failure to terminate SI Page 41 of 116 ORWS NOLOSP OTS1S

$31.8K (nominal benefit)

S60.6K (VB benefit)

OECR Red: -5%

Min. Cost: >$100K

$335K (nominal benefit)

$638K (VB benefit)

OECR Red: -43%

Min. Cost: > 1M

$28.3K (nominal benefit)

$53.8K (VB benefit)

OECR Red: -5%

Min. Cost: >$IOOK assumed elimination of the human failure risk including actions OTSI3, OTS14 and

. OTSI5 for SI termination duringSGTR Not cost beneficial based on assmned elimination of the human failure risk Not cost beneficial based on assumed elimination of all risk associated with loss of affsite power OTS1S is representative case forHH.ODDSGI.FA. Not cost beneficial based on assumed elimination of the human action to diagnose SGTRevent

United States Nuclear Regulatory Commission SBK-L-lJOOI/Enclosure

1. 5 Operator depressurizes RCS to Stop Break Flow to HH.OSGRD I.FA Ruptured SG (SGTR)

1. 6 Operator fails to CooldownlDepressurize RCS to HH.ORWCD1.FA Minimize Leak wlRecirculation Failure (VSEQ)

1. 7 Turbine Driven Pump FW-P-37A fails to run FWF37A.FR

1. 8 Operator fails to initiate makeup to RWST given HH.ORWINI.FA LOCA wi recirculation failure

1. 9 Operator fails to terminate EFW flow to ruptured SG HH.OTEFW3.FA

-SGTR Implement hardware change to improve reliabiIi[yof SGTR control.

eliminate operator action to de ressurize 1.072 Implement hardware change to improve reliability, eliminate operator action to de ressurize 1.049

1. 163 1.048 Hardware change to provide auto-makeup to

RWST, eliminate operator action 1.038 Hardware change to eliminate operator action to depressurize Page 42 of 116 Not cost beneficial based On assumed elimination *of operator action to depressurize ReS ORWCDS 54.4 K (nominal benefit)

Not cost beneficial based on

$8.3K CUB benefit) assumed elimination of OECRRed: -1%

operator action to cooldown Min. Cost: >$IOOK and depressurize RCS TDAFW

$IOOK (nominal benefit)

Not cost beneficial based on

$190K CUB benefit) assumed elimination of all OECR Red: -7%

IDEFW pump failures Min. Cost: >$250K ORWS

$31.8K (nominal benefit)

ORWS is representative case S60.6K (VB benefit) for HH.ORWINl.FA. Not OECR Red: -5%

cost beneficia! based on Min. Cost: >$IOOK assumed elimination of the human action to perform RWSTmakeup OSGRDS

$4.5 K (nominal benefit)

OSGRDS is representative

$8.5K CUB benefit) case for HH.OTEF3.FA. Not OECRRed: -1%

cost beneficial based on Min. Cost: >$lOOK assumed elimination of operator action to terminate EFW flow for SGTR

United States Nuclear Regulatory Commission SBK-L-llOOl / Enclosure

1. 10 Operator fails to maintain s1able plant conditions wi HH.ORWLTl.FA long tenn makeup

1. 11 Operator fails to cooldown RCS to allow isolation of HH.OSGRCI.FA ruptured SG (SGTR)

1. 12 Operator fails to cooJJdepressurize RCS for RHR HH.ORHCD8.FA SID cooling for SGTR wi failure of OSGRD

1. 13 Containment Building Spray Pump P-9B discharge CBSVI7.FC MOV CBS-V-J7faiJ to open on demand Hardware change for automatic control or eliminate operator action to maintain stable conditions 1.032 Hardware change for automatic control or eliminate operator action to cooldovm RCS 1.030 Hardware change for automatic conuolor eliminate operator action to cooldown RCS 1.025 Hardware cbange to improve valve reliability, elIminate CBS discharge MOV failures Pag.e 43 of 116 ORWS ORWS is representative case for HH.ORWLTI.F A. Not cost beneficial based on Min. Cost: >$100K assumed elimination of the human failure risk OSGRDS

$4.5 K (nominal benefit)

OSGRDS is representative

$8.5K (UB benefit) case for HH.OSGRC1.F A.

OECRRed: -1%

Not cost beneficial. based on Min. Cost: >$100K assumed elimination of operator action to cooldown RCS ORHCDS

$12.3 K (nominal Not cost beneficial based on benefit) $23.5K (UB assumed elimination of all benefit)

ORHCD related operator OECR Red: -2%

actions to cooldown ReS Min. Cost: >$100K CBSDVS

$<IK (nominal benefit)

Not cost ben~ficial based on

$<IK (UB benefit) assumed elimination of both OECRRed: <1%

CBS-V-lJ and CBS-V-17 Min. Cost: >$100K failure to open on demand

United States Nuclear Regulatory Commission SBK-L-IIOOI I Enclosure

1. !5 HH.OSEP2.FA-Containment Building Spray Pump P-9A discharge MOV CBS-V-!! fail to open on demand Operator fails to close SEPS breaker from the MCB given SI signal 1.0!9 Hardware cbange to improve valve reliability, eliminate CBS discharge MOV failures Hardware change for auto closure ofSEPS breaker to eliminate operator action Page 44 of 116 OSEPALL

$32.5K (nominal benefit)

$62K (UB benefit)

OECRRed: -3%

Min. Cost: >$IOOK assumed elimination of both CBS-V-ll and CBS-V-!?

failure to open on demand Not cost beneficial based on assumed elimination of all operator actions needed to close SEPS breaker

United States Nuclear Regulatory Commission SBK-L-ll 00 I I Enclosure

1. 1 HH.XOVNTJ.FA Hardware change for automatic venting

control, eliminate need toperronn late containment ventin

1. 2 HH.XOSMPI.FA Operator aligns containment sump recirculation after 1.53 I Provide a hardware modification for auto-

control, eliminate operator action to align sump after core melt.

1. 3 FWP37A.FR

1. 4 DGDGIA.FR3

1. 5 XX.EDESWGS.FX core melt Turbine Driven Pump FW-P-37 A fails to run DG-\\A fails to run for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> 4KV BUS ES fault (IE) 1.134

1. 163 1.088

1. 9, #10, #14,

1. 155 1.080 Improve Bus E5 reliability, eliminate all potential for bus fault Page 45 ofU6 XOSMPS TDAFW NOSBO E6S

$2 J K (nominal benefit)

$39.7K (UB benefit)

OECR Red: -5%

Min. Cost: >$IOOK

$IOOK (nominal benefit)

$190K (UB benefit)

OECRRed: -7%

Min. Cost: >$250K

$I55K (nominal benefit)

$295K (UB benefit)

OECRRed: -14%

Min. Cost: >$500K

$39K (nominallienefit)

$74.IK (UBbenefit)

OECR Red: -2%

Min. Cost: > lOOK Not cost beneficial based on assumed elimination of operator action to vent containment Not cost beneficial based On assumed elimination of human action Not cost beneficial based on assumed elimination of all TDEFW pump failures Not cost beneficial based on assumed elimination of all EDG failures Not cost beneficial based on assumed elimination of all risk associated with Bus fault, case E6S representative

United States Nuclear Regulatory Commission SBK-L-llOOI/EncJosure DGDGIB.FR3

1. 7 4KV BUS E6 fault (IE)

XX.EDESWG6.FX

1. 8 Loss of Off site Power subsequent to plant trip zz.SYI.FX

1. 9 4KV BUS E6 fault EDESWG6.FX

1. 10 I-SEPS-DG-2-A fails to run within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> SEPSDG2A.FRJ

1. 11 I-SEPS-DG-2-B fails to run within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> SEPSDG2B.FRJ 1.068 Improve Bus E6 reliability, eliminate all potential for bus fault 1.049

1. 13, #156,

1. 160 1.042 Improve Bus E6 reliability, eliminate all potential for bus fault 1.038

1. 9, #14, Improve reliability of

SEPSDG, eliminate potential for SEPS failure 1.038

1. 9, #14, I~prove reliability of

SEPSDG, eliminate potential for SEPS failure Page 46 of 116

$155K (nominal benefit)

$295K CUB benefit)

OECRRed: -14%

Min. Cost: >$500K E6S

$39K (nominal benefit)

Not cost beneficial based on

$74.IK CUB benefit) assumed elimination of all risk OECR Red: -2%

associated with Bus fault Min. Cost: > lOOK NOLOSP

$335K (nominal. benefit)

Not cost beneficial based on

$638K CUB benefit) assumed elimination of all OECR Red: -43%

risk associated with loss of Min. Cost: > 1M off site pov:,cr E6S

$39K (nominal benefit)

Not cost beneficial based on

$74.1K CUB benefit) assumed elimination of all risk OECR Red: -2%

associated with ~us fault Min. Cost: > lOOK SEPES

$39.8K (nominal benefit)

Not cost beneficial based on

$75.8K CUB benefit) assumed elimination of all OECR Red: -2%

SEPS fajlures Min. Cost: >$IOOK SEPES

$39.8K (nominal benefit)

Not cost beneficial based on

$75.8K CUB benefit) assumed elimination of all OECR Red: -2%

SEPS failures Min. Cost: >$100K

United States Nuclear Regulatory Commission SBK-L-11001/Enclosure

1. 12 CCTE2271.FZ

1. 13 PCC Train A Temperature Element CC-TE-2171 CCTE217I.FZ transmits false low

1. 14 Operator fails to 'start containment injection early HH.OINE3.FA withoutAC power (gravity drain ofRWST)

1. 15 4KV BUS E5 fault EDESWG5.FX ImprovePCC TE reliability, eliminate potential for temp. element failure 1.036 ImprovePCC TE reliabil ity, eliminate potential for temp. element failure 1.033 Hardware change for automatic initiation of containment injection gravity drain, eliminate operator action 1.030 Improve Bus E5 reliability, eliminate all potential for bus limit Page 47 of 116 Not cost beneficial based on assumed elimination of all risk associated with TE-2171 and TE-2271 failures PCCTS

$29K (nominal benefit)

Not cost beneficial based on

$55.3K CUB benefit) assumed elimination' of all risk OECRRed: -3%

associated with TE-2171 and Min. Cost: >100K TE-227f failures XOINES

$4.1 K (nominal benefit)

Not cost beneficial based on S7.7K CUB benefit) assumed elimination of all OECRRed: -1%

operator actions to initiate Min. Cost: >$IOOK*

containment injection (XOINEI, XOINE2, XOINE3)

E6S

$39K (nominal benefit)

Not cost beneficial based on

$74.IK CUB benefit) assumed elimination of all risk OECRRed: -2%

associated with Bus fault, case Min. Cost: > lOOK E6S representative

United States Nuclear Regulatory Commission SBK-L-l100 1 / Enclosure SAMARAI2g

2)

Provide the following infOlmation relative to the Level 2 analysis:

g. Table F.3.4.3-2 provides the accident categolY frequencies and release fractions. The release times indicate a range for the bulk of noble gas and cesium (Cs) release. ClarifY the meaning of the two release time ranges provided for each of the release categories. In addition, for each MAAP case, provide the following for noble gas, Cs, and Iodine releases: (1) time after SCRAM or when a general emergency (GE) is declared, (2) total duration of the release modeled, and (3) release fraction and start/end of release for each plume release (if multiple plume releases are modeled).

NextEra Energy Seabrook Response to SAMA RAI 2g The two release time ranges are the stmi and end times (hours from SCRAM) of the noble gas and Cs releases. For the Seabrook original analysis accident categories (LE-2, LE-3, SE-2, SE-3, and LL-5), the time ranges represent the entire release plume. For example, the entire LE-2 release of noble gas and Cs is between 2.5 and 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> from scram. For the MAAP accident categories, the time range represents most of the entire release (see following tables).

The following tables give the noble gas, iodine, and cesium release fractions, start time of release plume (measured from SCRAM), and duration of release plume (the release plume end time would be the stmi time + duration) for each of the MAAP cases (all modeled with multiple release plumes). The tables also give the accident categOlY general emergency declaration time. In the following tables, all times are in seconds (in keeping with MACCS2 input specifications).

Category General Emergency Declaration, sec fi'om SCRAM 9846 NIA NIA NIA Plume Duration, sec 2984 2912 1156 69502 Plume Start Time, sec from SCRAM 9846 12830 15743 16898 Noble Gas Release Fraction 2.44E-02 5.86E-Ol 8.80E-02 1.00E-03 Iodine Release Fraction 2.38E-03 2.65E-02 9.00E-04 1.00E-04 Cesium Release Fraction 2.26E-03 2.37E-02 7.17E-04 8.34E-06 Page 48 of 116

United States Nuclear Regulatory Commission SBK*L*1l001/Enclosure Category General Emergency Declaration, sec from SCRAM Plume Duration, sec _

Plume St.tt Time, sec from SCRAM Noble Gas Release Fraction Iodine Release Fraction Cesium Release Fraction Category General Emergency Declaration, sec from SCRAM Plume Duration, sec Plume Start Time, sec from SCRAM Noble Gas Release Fraction Iodine Release Fraction Cesium Release Fraction Category General Emergency Declaration, sec from SCRAM Plume Duration, sec Plume Start Time, sec from SCRAM Noble Gas Release Fraction Iodine Release Fraction Cesium Release Fraction 3600 1800 9000 9,OOE*OJ 7,OOE*OJ 5.00E*OI 10800 7200 14400 2.00E*OI 4.00E*03 4.00E*03 9846 2635 9846 1.26E*02 9.47E*07 8.68E*07 NIA 14400 21600 3.00E*OI 5.00E*03 5.00E*03 N/A 4856 1248J 2.15E*02 4.58E*03 4.51E*03 Page 49 of 116 NIA 36000 36000 5.00E*OI l.00E*03 l.00E*03 N/A 2866 17338 5.00E*03 5.00E*05 4,08E*05 N/A 66197 20203 1.30E*03 7,OOE*05

. 2.42E*05

United States Nuclear Regulatory Commission SBK-L-ll 00 11 Enclosure

~~~.';'.. '

.:.~:,'.

Rel~ase Catei<irySE~2 I,

C

• .1' Category General Emergency Declaration, sec from SCRAM Plume Duration, sec Plume Start Time, sec from SCRAM Noble Gas Release Fraction Iodine Release Fraction Cesium Release Fraction

'F.,:

Relc!!se Cate~o;" SE-3 Category General Emergency Declaration, sec from SCRAM Plume Duration, sec Plume Start Time, sec tl'om SCRAM Noble Gas Release Fraction Iodine Release Fraction Cesium Release Fraction

!::.; ~. <1'>

Release Category LL-3 Category General Emergency Declaration, sec from SCRAM Plume Duration, sec Plume Stalt Time, sec from SCRAM Noble Gas Release Fraction Iodine Release Fraction Cesium Release Fraction 7200 25200 30600 9.00E-OI 7.00E-04 5.00E-04 I:{ PI'\\ll)e-l 50400 43200 79200 1.50E-OJ 4.00E-03 4.00E-03

,.~.',

. Plume' l 0

5998 1541 1.07E-04 2.31E-OS 2.24E-05

/<-;.

" PluJ;e;2;_i~*:*

c'....

NlA 28800 122400 2.00E-OI 7.00E-03 7.00E-03 N/A 2804 7538 5.60E-OS 4.70E-06 4.52E-06 Page 50 of116 I,:,,:~"B~'~"

I

• ;::r'~~~~\\ !:.\\

NlA 86400 151200 6.50E-OI 2.00E-03 2.00E-03 N/A N/A 86400 86400 10343 122882 4.22E-03 6.8SE-OI 6.08E-05 2.73E-03 2.59E-05 l.32E-03

United States Nuclear Regulatory Commission SBK-L-IIOO I I Enclosure

, ~.

I '

-~.; ;.;.

Release Caiego'ry r,-J;4 Category General Emergency Declaration, sec from SCRAM Plume Duration, sec Plume Start Tune, sec from SCRAM

• Noble Gas Release Fraction Iodine Release Fraction Cesium Release Fraction

','li;i~~s!, C~'iegoi'Y LL-:5

,'.)

Category General Emergency Declaration, sec from SCRAM Plume Duratio~, sec Plume Start Time, sec from SCRAM Noble Gas Release Fraction Iodine Release Fraclion Cesium Release Fraction T-"",.'.. "'."',)-:"

I,

,~~~~g;~~~~ry,/.. '

Category General Emergency Declaration, sec from SCRAM Plume Duration, sec Plume Start Time, sec from SCRAM Noble Gas Release Fraction Iodine Release Fraction Cesium Release Fraction

' ' Plume-!

0 637 IS47 6,43E-07 L06E-07

9. J3E-08 266400 3600 320400 LOOE+OO LOOE-03 LOOE*03

"'j'"

,' Plume-1

~.

6826 378 90 11 3.26E*07 4,98E*09 4,J6E-09

':<; Plurue-2. "

N/A 3481 2484 2.86E-05 L75E-06 L33E-06

',:'\\

':}

' *Plume-2 ';,'

N/A 2772 9389 1.04E-05 7,0IE-08 5,2IE-08 Page 51 of 116

. "Plume-3

' Plume'4 N/A N/A 86400 86400 5965 129218 6,70E-03 9,93E-01 2.SIE.. 05 3,51E-01 8.45E.. 06 2.2lE-01

~ ~;;....

Plume-3

'Plume-\\!

_, I,..

~.:'

N/A N/A 19858 86400 12161 32018 L95E-04 3.25E-03 1.94E-08 7.50E-09 L09E-08 LJ8E-09

United States Nuclear Regulatory Commission SBK-L-llOOll Enclosure R~le~se ~~tegory INTACT2 (a)

Category General Emergency Declaration, sec from SCRAM Plume Duration, sec Plume Start Time, sec from SCRAM Noble Gas Release Fraction Iodine Release Fraction Cesium Release Fraction 6826 378 9011 3.26E-06 4.98E-08 4.16E-08 NlA N/A N/A 2772 19858 86400 9389 12161 32018 1.04E-04 1.95E-03 3.25E-02 7.01E-07 1.94E-07 7.50E-08 5.2IE-07 1.09E-07 J.l8E-08 Note (a): Sum of INTACT I and INTACTI risks are reported in ER as release category INTACT risk.

ER Table F.3.4.3-2 frequency for category INTACT is sUll! of INTACT 1 and INTACT2 frequencies; for reporting, ER Table F.3.4.3-2 INTACT release fractions are for INTACT!, which contributes 91 %

of the INTACT dose risk.

SAMARA12h

2)

Provide the following information relative to the Level 2 analysis:

h. With respect to F&O 3, discuss how this resolution addresses the Peer Review Finding aspect regarding the training, qualification, and familiarity of plant staff with the long-term operation of the turbine-driven (TD) emergency feedwater (EFW) pump.

NextEra Energy Seabrook Response to SAMA RA12h As stated in the resolution to F&O 3, the battelY lifetime was recalculated and battelY life is up to 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> with load shedding/cross tie actions. These actions "preserve" the 125V DC vital bus power needed to control the steam-driven EFW pump arId provide power to necessary instmmentation (e.g., SG level). Therefore, during an extended station black-out condition with successful DC power, long telm control of SO water level and heat sink uses the nonnal control room instrumentation and procedures for which the operators are trained and familiar with.

Page 52 of 116

. United States Nuclear Regulatory Commission SBK-L-llOOll Enclosure RAI Section 3 SAMARAI3a

3)

Provide the following information with regard to the treatment and inclusion of external events in the SAMA analysis:

a. The ER does not address the status of one plant improvement identified in the IPEEE SER:

modification of several extyrior doors so that they will be able to withstand the design pressure differential resulting from high winds. Discuss the status oftms improvement and, ifnot ah'eady

. implemented, provide a cost-benefit evaluation of a SAMA that addresses this improvement recommendation.

NextEra Energy Seabrooll Response to SAMA RAI 3a The door modifications mentioned in the lPEEE submittal and associated SER were completed in 1993.

Therefore, the door modifications need not be evaluated for cost-benefit in the SAMA. The doors subject to the modification included doors: P-900 (RHR Vault E!. 20' -8"), P-604 (P AB RoofE!. 81' -0"), EF-400 and EF-404 (EFW Pump Room E1. 27'-0") and EM-210 (Main Steam Pipe Chase, E1. 3'-0"). The modification to these doors included replacement of the mortise 10cksets with higher capacity locksets.

SAMARAI3b

3)

Provide the following information with regard to the treatment and inclusion of external events in.

the SAMA analysis:

b. ER Section F.3 explains that internally initiated fire events are included in the current PRA and that the fire risk analysis has been updated since the IPEEE. There is no discussion of the fire PRA method in the ER and no presentation of the impOliant fire areas and their contribution to Levell or fire CDF. In light of this, provide:

a) A description of the fire risk analysis method including to what extent the method was based on NUREG/CR-6850. In the response, specifically discuss how fire-induced ISLOCAs are addressed, how fire-induced containment impact is addressed, and model conservatisms. In the response, specifically address whether the RRWs listed in Table F.3.2.1-2 include fU'e-induced sequences where a component required to maintain containment integrity could be failed by the fire itself rather than randomly and independently from failures that induce core damage.

b) Fire PRA results including revised fue zone contribution to the CDF. Additionally,

. explain the reason for significant differences between the IPEEE and updated Fire PRA results.

Page 53 of 116

United States Nuclear Regulatory Commission SBK-L-Il 00 I / Enclosure NextEra Energy Seabrook Response to SAMA RAI 3ba A. Fire Risk Analysis Methodology The most recent update to the Seabrook Station fire PRA was done in 2004, prior to the publication of NUREG/CR-6850. The methodology used to assess the fire risk at Seabrook Station was the most current available at that time. The fire risk analysis consists of the following general steps:

1. Definition offire areas to be analyzed - The Appendix R fire areas and zones were used as a basis for this study since they include all safety related areas and some additional areas. Some areas were combined for this analysis in order to streamline the screening process.

2. Development of a spatial data base of equipment and cable routing for the defined fire areas - A record for each key component within each fire area was created, which included the component location, power and control cable location, motive and control power supplies amI locations, etc. Appendix R was the principle source for equipment location and cable routing. InfOlmation for equipment not addressed in Appendix R was obtained from plant arrangement drawings, walkdov>'IIs, and the CASP computer code cable routings.

3. Qualitatively screening out any areas which did not contain equipment or cables which could cause or mitigate an initiating event.

4. Development offire ignition frequencies for those areas that had not been qualitatively screened out-The fire ignition frequencies were developed by reviewing the latest industry fire frequency data for components and areas and applying them to Seabrook Station areas based on actual components in the defined fire areas.

5. Quantitatively screening out areas based on the area fire ignition frequencies developed in Step 4 -

For this screen, it is conservatively assumed that all equipment and cables in the area could fail at the calculated frequency. The resulting fire initiating events are compared to the existing internal events. If their contribution to the some internal event is negligible (less than about five percent), they are screened from the further analysis. If all possible fire initiating events in a specific area are screened out, the entire area is considered to be quantitatively screened out.

6. Areas remaining after the quantitative screening receive detailed hazards analysis. The methodology for the hazards analysis is based upon the quantitative fire hazard equations and FIVE Methodology. The analysis includes:

Identifying the location of critical equipment (Le., target sets) and the severity of a fire needed to disable this equipment.

Estimating the frequency of a disabling fire based on ignition sources, estimated sevei'ity factor and detection and suppression system availability.

7. Impacts on the plant from failures of those target sets that the hazards analysis identified as potentially important were modeled in detail. This included evaluating human actions needed to respond to the fire, such as controll'Oom evacuation and operation of equipment from remote locations.

Page 54 of 116

United States Nuclear Regulatory Conunission SBK-L-1I001/Enclosure B. Fire-induced ISLOCA & Containment Impact The potential impacts from fire hazard on the containment function, including ISLOCA bypass sequences, are addressed below. As described below, the only credible impact of fires on containment performance is to fail a single train of isolation. For isolation failure of one or more valves of a single train, either redundant isolation would be available or the ability to remove power from fail-closed valves to provide isolation. The frequency of fires that could cause this level of damage is low enough, compared to hardware failures, to not contribute significantly to containment isolation failure. As a result, no fire impacts on containment isolation components are included in the PRA.

Containment performance was evaluated in three areas, with regard to the fire hazard:

1. Contaimnent structure. Fires could have no impact on the containment structural integrity.

2. Containment response to core damage event. Fire-initiated core damage events would have the same impact, with regard to containment response, as internal initiated events. This impact is handled in the Containment Event Tree.

3. Containment isolationlbypass failure. This is the focus of the evaluation described below.

The potential for containment isolationlbypass failure has been evaluated for each containment penetration. The important isolation valves are discussed below.

Cables for containment isolation valves are routed through two distinct paths. Cables for isolation valves located inside the containment building travel through one of the two train-related electrical tunnels.

Cables for valves in the mechanical penetration area travel through the following areas; the P AB electrical chase, duct bankDB-3B, the RHR vaults and the mechanical penetration area. Therefore, afire in any of these areas could only impact the set of isolation valves inside containment or the set outside containment, but not both. The only area where isolation valves both inside and outside containment could be affected would be in the control room or the cable spreading room, and in this event, important isolation valves could be controlled locally at the valve or from the remote shutdown panels.

Fires can affect valves in several ways. Failure ofMOV power cables will fail the valve in its current position. If the valve can be accessed, the position could be changed manually. A hot shott to an MOV control cable can possibly result in a change in position of the valve. Shotts to ground of control cables for AOVs or SOVs will move the valve to its failed positionjust as loss of control power to the circuit.

Hot shotts to de-energized control circuits can also result in inadvettent valve movement, which could be tetminated by removing power to the affected control circuit. Hot shorts to three phase power cables would require multiple contacts and are not considered credible.

Because during operation a number of isolation valves to containment are open, the most likely adverse affect of a fire would be inadvettent isolation of a system. The letdown system has several fail closed AOV s which could isolate letdown with failure of their control cables. It is not credible for all three valves to hot short. The excess letdown system does have AOVs in the closed position. These valves fail closed; therefore, a hot short in 3 valves in series would be required to open a path out of containment and they could easily be closed by removing power to the control circuits. The PCC flow to the thermal Page 55 of 116

United States Nuclear Regulatory Commission SBK-L-1I00ll Enclosure barrier heat exchangers has fail-open AOVs outside of containment, but the PCC flo\\'{ is not open to the RCS, and additional isolation between the RCS and PCC is provided inside containment.

The large RHR suction line MOVs which are modeled for V-sequence events are normally closed and their breakers are locked open at-power. Therefore a fire could not cause them to spuriously open.

The RCP seal return line isolation valves (penetration X-37) are MOVs, which fail as-is on loss of AC power. Fires that cause loss of AC power are modeled to account for this small containment penetration line failing open (given failure of operator action to manually close the outside valve).

Other large isolation valves, such as the containment on-line purge v<\\lves are normally closed and will fail closed on loss of Inshument Air or AC power.

The conclusion of this review is that the only impact of fires on containment peIformance is to fail a single train of isolation. For isolation failure of one or more valves of a single train, either redundant isolation would be available or the ability to remove power from fail closed valves to provide isolation.

The frequency of fires that could cause this level of damage is low enough, compared to hardware failures, to not conh'ibute significantly to containment isolation failure.

C. Model Conservatism Fire risk analysis contains a number of conservative modeling steps. Perhaps the most significant is the postulation that small fires, typical of the generic fire events database, can actually grow to cause the maximum damage assumed. Because it is difficult to model the specific fire gTOwth scenario for each potential fire source, the method uses conservative assumptions to allow the fire risk to be bounded. In general, these fire sequences have such low frequencies and such large uncertainties that it is difficult to detelmine the impact ofthe conservatism on the overall fire CDF.

NextEra Energy Seabrook Response to SAMA RAI 3hh The following table summarizes the fire CDF contribution by fire area from the 2009 Update & the IPEEE:

Fire L'O~aiion (Fire-Zone I Ar~a Control Room (CB-F-3A-A) 52%

34%

Page 56 of 116

United States Nuclear RegulatOly Commission SBK-L-llOOl/Enclosure

~

Seabrook Fire Area

~,_

Fi~e Locatio~'

(Fire Zone I Area ID)

Essential Switchgear Rooms (CB-F-IA-A, CB-F-lB-A)

Turbine Building (TB-F-1A-Z,... TB-F-3-Z)

Primary Auxiliary Building (PAB-F-IA-Z,... PAB-F-3C-Z)

Ocean Service Water Purnphouse (SW-F-IA-Z,... SW-F-lE-Z, SW-F-2-0)

Electrical Tunnels (ET-F-IA-A,... ET-F-lD-A)

FIRE CDF TOTAL

~" '

~

~~

"~ 2009 Update,

~ Contribution.to Fire CDF 41 %

5%

2%

1%

<1 %

1.7e-6/yr WEEE (iJ90),

Contribution to Fire CDF',

18%

13%

26%

9%

<1 %

1.2e-5/yr The 2009 fire CDF is approximately 1.7e-6/yr, a 14% contribution to total CDF. This is in comparison to the IPEEE fire CDF of 1.2e-5, an 11 % contribution to the total CDF in 1990. Thus, while the fire CDF has decreased significantly from 1990 to 2009, the relative contribution to CDF total has remained relatively constant.

The 2009 Update differs from the IPEEE model in the following areas: (1) the inclusion ofthe cutTent plant data and procedures, (2) the performance of detailed walkdowns verifying locations of the major fire sources and impOltant targets, (3) the availability of the new EPR! Fire Database which includes fire records through December 2000, (4) new severity factors were evaluated for cabinets, pumps, control room panels and transients, (5) the quantitative screening results were revisited, (6) the new data on the cabinet release rates were included in the fire hazard evaluation, and (7) total area heat-up was considered in the analysis and evaluated by using quantitative fire hazard equations or CFAST software. The reduction in fire CDF is due to more detailed modeling of the fire areas shown to be important in the IPEEE, using the factors described above. No single change explains the decrease in fire CDF.

SAMARAI3c

3)

Provide the following infonnation with regard to the treatment and inclusion of extemal events in

, the SAMA aualysis:

c. In the description ofPRA model changes made since 1993, on pages F-20 through F-28 of the ER, at least one instance of a major update to the seismic PRA was indicated (Le., on page F-27).

Page 57 of 116

United States Nuclear Regulatory Commission SBK-L-IIOOll Enclosure However, there is no discussion of the updated seismic PRA methodology in the ER. In light of this, provide:

a) A description of the seismic risk analysis method including the seismic hazard curves being modeled. Additionally, provide a discussion of model conservatisms.

b) Seismic PRA results including revised seismic initiator contribution to the CDF.

Additionally, explain the reason for significant differences between tlie IPEEE and updated Seismic PRA results.

NextEr3 Energy Seabrook Response to SAMA RA13c3 The current seismic PRA analysis was updated fi'om the IPEEE model in 2005. The seismic PRA was performed using the following methods:

Hazard Analysis - determination of the frequency of gronnd motions of various magnitudes at the site.

Seabrook Station site specific probabilistic hazard curves were developed for the SSPSA (1983).

However, the present PRA model has been updated to the more recent EPRl hazards. This was done because, while the methodology and experts used in developing the EPRl hazard are essentially the same as the SSPSA, the EPRl hazard is more recent and the EPRl uniform hazard spectrum (UHS) developed for the Seabrook Station site is more realistic than that used in the SSPSA. The EPRT UHS was the basis for the revised equipment fragilities. The sensitivity of different hazard inputs, including the SSPSA and LLNL hazard curves is addressed.

Fragility Analysis - detelmination of the seismically initiated ground acceleration at which plant structures and components are predicted to fail. The probabilistic estimates of seismic capacity of structures and components have been updated to reflect component-specific fi'agility information and a site-specific UHS. Structures and components with a median capacity of2.5g or greater were screened out of the analysis. The sensitivity of this screening level is addressed.

Seismic Model (Systems Analysis) - this involves several tasks, from identifying the equipment list that needs fragility analysis to integrating the results of the above elements into the PRA model and quantifying risk. The RlSKMAN software is used to combine hazard and fragility inputs, develop initiating event frequencies, and fragility values at discrete hazard levels to yield conditional system failure probabilities. These are used to provide initiating event and split fraction values for the plant model. The PRA model event trees used for internal events are used to integrate seismic initiators and seismic-initiated component failures with random (non-seismically caused) hardware failures and maintenance unavailability. A separate event tree "HAZARD" is included in the plant model. The HAZARD tree models the seismic failure contribution of components conditional on the seismic initialing event. Then, all the support and plant response event trees are conditional on the initiator and equipment fi'agilities in the HAZARD tree. In this way, seismic failures are treated-as other failures and are combined with non-seismic failures as appropriate based on the model logic.

Page 58 of 116

United States Nuclear Regulatory Commission SBK-L-lI 00 I 1 Enclosme The most recognizable conservatism in the seismic model is the use of complete correlation of the 1iagility between identical components. An example is both EDGs are assumed to fail at the same seismic hazard level.

NextEra Energy Seabrook Response to SAMA RAI 3cb The following table summarizes the seismic CDF contribution by initiator from the 2009 Update & the IPEEE:

~:.

2009 Update

-Illiti:at"i'ng' Q~nl\\p :.,

Contribution to Seismic CDF TRANS Total 78%

ATWS Total 11%

LLOCA Total 10%

Seismic CDF TOTAL 3.le-6/yr lPEEE (1990)

Contribution to Seismic

. 'CDF 65%

24%

11%

1.2e-51 yr The dominate seismic hazard level for the 2009 Update is the 0.7g earthquake, which accounts for 35% of the seismic CDF. The dominate seismic hazard level for IPEEE was also the 0.7g earthquake, which accounted for 36% of the seismic CDF.

A number of changes were made in the 2005 Update that revised the IPEEE seismic model. The most significant changes include the following:.

The fragility screening of equipment was extended from >2.0g to >2.5g for several reasons. The frequency of exceeding 2.5g is less than lE-7/year, which is considered a small change inrisk and an appropriate cutoff for the seismic hazard. Also, with improved more realistic fragilities, there were fewer components to model. Seabrook has added supplemental diesels, which are expected to reduce seismic station blackout risk. Thus, there was a need to consider extending the seismic initiating events (hazard) to higher accelerations to better capture seismic risk. As a result, the PRA model was revised to consider seismic initiating events and equipment £i'agility up to 2.5g.

A plant-specific uniform hazard spectrum was adopted and used to update the equipment fragilities. The original UHS used in the fragility analysis was conservative based on current understanding of the hazard frequency spectrum. Adjustments to the calculation for equipment fragilities resulted in increased equipment capacities.

However, several new component fragilities were added to the seismic PRA model in the 2005 update. There were cases where equipment was originally screened because its fragility was higher tban other equipment with the same impact. With the update, this ttuncation was not always appropriate. Also, extending the screening value to 2.5g resulted in the addition of equipment to the model.

Page 59 of 116

United States Nuclear RegulatOlY Conunission SBK-L-llOOI/Enclosure Supplemental emergency power supply (SEPS) diesel generators had been added to the plant design and are impOliant to the reduction of station blackout risk. Thus, a component-specific seismic fragility was developed for SEPS and added to the PRA model.

The modeling and documentation of operator actions credited in the seismic PRA were improved.

SAMARAJ3d

3)

Provide the following information with regard to the treatment and inclusion of external events in the SAMA analysis:

d. ER section F.3.1.2.2 explains that the NUREG-1407 procedure for screening high wind, flooding, and other external (HFO) events was used to conclude that contribution to the Seabrook Station total CDF from HFO is less than 1.0E-06 per year. However, the IPEEE discusses two external events that have a CDF contribution greater than 1.0E-06 per year. W11i1e the ER addresses one of these events, flooding caused by a stOim surge, it does not address the second: a huck crash into the SF6 transmission lines having an IPEEE CDF contribution of 1.4E-06 per year. Discuss whether this event is addressed by a loss of off-site power initiator in the current PRA model and, if not, assess the impact of this event on the results of the SAMA evaluation.

NextEl'a Energy Seabrook Response to SAMA RAJ 3d The external initiating event involving a truck crash into the SF6 transmission lines (previous initiator TeTL), has been screened from the PRA model based on the installation of jersey barriers associated with security upgrades. These barriers are located before the access road parallels the SF6 bus duct, which was the "target" for initiator TCTL. Now a huck must completely stop and have a security escort before proceeding into the protected area. Additional guard rails, which were added a number of years ago, also limit the possibility of a slow speed h'uck crash impacting the transmission lines. As a result, there is no impact on the SAMA evaluation.

SAMARAI3e

3)

Provide the following infolmation with regard to the treatment and inclusion of external events in the SAMA analysis:

e. While the ER and IPEEE address flooding resulting from a storm surge caused by a hun'icane, neither appears to specifically address the impact of hurricane-force winds. In light of the potential for hU1l'icanes and "Nor'easters" hitting the Seabrook Station, assess the risk of hurricane-force and the impact of that risk on the results of SAMA evaluation.

Page 60 of 116

United States Nuclear Regulatory Commission SBK-L-llOOI/Eoclosure NextEI'a Energy Seabrook Response to SAMA RAI 3e An assessment of high wind hazards was performed and documented in Section 5.1 of the Seabrook Station IPEEE. The assessment consisted of a review of UFSAR site design basis for high winds and tOlTIildoes against SRP criteria and site-specific weather data. High wind hazards were screened from further detailed assessment based on review of the SRP criteria against Seabrook Station's design. All seismic category I structures exposed to high wind forces are designed to withstand wind velocity of 110 mph at 30 ft above nominal ground elevation. In addition, tornado loads are based on a 290 mph tangential wind velocity and a 70 mph translational wind velocity, with a simultaneous atmospheric pressure drop of3 psi at a rate of2 psi per second. Plant impacts from hurricanes and northeasters are.

considered bounded by these stringent design requirements. The confirmatOlY walkdowns performed for high wind hazards identified the need to modify several exterior doors so that they would be able to withstand the design pressure differential resulting from high winds. These door modifications are complete - refer to NextEra Energy Seabrook Response to RAJ 3a. Based on the above, the risk to plant structures from high winds due to hUlTicanes and northeasters is not significant and there is no impact to the SAMA evaluation.

SAMARAI3f

3)

Provide the following infOimation with regard to the treatment and inclusion of external events in the SAMA analysis:

f. ER Section E.5.5.3 does not identify any reviews of the fire or seismic PRAs. Identify any internal and external reviews of the fire and seismic PRAs, discuss how each finding was resolved, and provide an assessment of the impact of any unresolved findings on the SAMA evaluation.

NextEra Energy Seabrook Response to SAMA RAJ 3f The fire and seismic portions ofthe PRA were reviewed extensively when originally performed in 1983, when revised for the IPEEE and, and when revised in the 2005 update. These reviews were typical internal technical reviews performed foJ' any PRA model revision. No significant comments were documented from these reviews. These reviews were not a fonnal Peer Review with findings and observations.

SAMARAI3g

3)

Provide the following infolmation with regard to the treatment and inclusion of extemal events in the SAMA analysis:

g. NRC Information Notice 2010-18, Generic Issue 199, "Implications of Updated Probabilistic Seismic Hazard Estimates in Central and Eastem United States on Existing Plants," infOims licensees that updated seismic data and models show increased seismic hazard estimates for some Page 61 of 116

United States Nuclear Regulatory Commission SBK-L-Il 00 II Enclosure plants. The NRC report cited in the information notice estimates the seismic CDF for Seabrook Station to be between S.9E-06 and 2.2E-OS per year using 2008 U.S. Geologic Survey (USGS) seismic hazard curves. Depending upon the contribution of seismic CDF to the total CDF, the impact of an increased seismic frequency could be significant enough to increase the estimate of the maximum attainable benefit, based on the cun-ent seismic frequencies employed for the Seabrook Station seismic CDF (to be provided via RAI l.b), such that pt'eviously non-cost-beneficial SAMAs could become cost-beneficial. As the seismic CDF is applied to non-seismic-related as well as seismic-related SAMAs, this could affect the detennination of "cost-bencficiality" for non-seismic-related as well as seismic-related SAMAs.Provide an assessment of the impact of the updated seismic CDF for Seabrook Station on the SAMA evaluation, including the basis for the increased seismic frequency that is chosen if other than the maximum ofthe range (2.2E-05 per year).

NextEra Energy Seabrook Response to SAMA RAI 3g It is noted in the NRC Infonnation Notice 2010-18 that NRC's risk assessment results for Generic Issue 199 ("Implications of Updated Probabilistic Seismic Hazard Estimates in Central and eastern United States on Existing Plants") included the following Limitations ofthe Risk Methodology and Data Used:,

"The approach used to estimate SCDF in the Safety/Risk Assessment is highly sensitive to the inputs used.

While work to date supports a decision to continue to the GJP Regulatory Assessment Stage; the methodology, input assumptions, and data are not sufficiently developed to support other reguiato/y decisions or actions".

Given the preliminary status ofthe in-progress01-199 study and results, use ofthesepreliminaty 01-199 results to the SAMA evaluation may not be meaningful. Nevertheless, using NRC's suggested approach discussed in the RAI, candidate SAMAs that were originally detelmined to be non-cost-beneficial based on their cost exceeding the plant's "maximum attainable benefit" (MAB) have been reassessed for "sensitivity" to their cost benefit value when considering the potential increase in baseline seismic CDF as presented in the latest GI-199 studies.

The approach taken in this assessment isas follows:

Seabrook Station baseline CDF from (at power) internal and extemalevents: 1. 17E-05/yr (SSPSS-2009)

Seismic contribution to CDF: 3.1IE-06/yr 026.5% of total CDF)

Estimated maximum seismic contribution from 01-199: 2.2E-OS/yr Increase factor to apply to the MAB: 2.6 (1.17E-05/yr - 3. llE-06/yr + 2.2E-OS/yr) 11.l7E-OS/yr = 2.6 factor increase in CDF The factor of2.6 increase results in the following MAB estimates:

Page 62 ofll6

United States Nuclear Regulatory Commission SBK-L-llOOl/Enclosure Orig.MAB MAB at2.6 nase Case (7%

Discount)

$O.818M

$2.l3M

$1.28M

$3.34M

,. 8.5% Discount 41 Yr VB

$O.729M

$1.22M

$1.56M

$1.89M

$3.17M

$4.05M The following SAMA candidates that were originally detelmined to be non-cost beneficial based on MAB are reassessed for their sensitivity to a new estimate ofMAB at 2.6.

SAMA #56 - Install an independent RCP seal injection system, without dediCated diesel.

Assessment: Adding an independent RCP seal injection pump should improve reliability ofthe RCP seal injection function and this would tend to reduce the risk of RCP seal LOCA events. Installation of an independent seal injection system is anticipated to realistically cost more than $3M. However, this SAlvfA is similar to other Seabrook Station SAMAs, for which PRA Case RCPLOCA was used to estimate the upper bound (UB) risk reduction benefit of$176K ($457K with assumed 2.6 increase factor).

Therefore, SAMA #56 is not cost-beneficial.

SAMA #65 - Install a digital feed water upgrade.

Assessment: Adding a digital feedwater control system should improve the reliability of the feedwater and SG water level control function and would tend to reduce plant trips due to loss of feedwater. It is noted that Seabrook Station is implementing a phased digital upgrade to balance of plant control systems (including feedwater controls). The initial phase of this upgrade included a digital EHCS. Installation of the entire digital feedwater control system is anticipated to realistically cost over $30M. Therefore, SAMA #65 is not cost-beneficial.

SAMA #77 - Provide a passive secondary-side heat rejection loop consisting of a condenser and heat sink.

Assessment: Installation of a passive SG heat rejection loop with condenser would tend to result in a reduction in core damage events from loss of feedwater events. For a "passive" system to function effectively, the scope of the system would need to include some type oflarge isolation-type condenser positioned and sized in such a way so as to allow a passive means of decay heat removal. The installation of such a system is judged not practical at an existing nuclear power plant such as Seabrook Station. The cost ofthls SAMA would be in excess of$IOM. Therefore, SAMA #77 is not cost-beneficial.

SAMA #90 - Create a reactor cavity flooding system.

Assessment: Installation of a cavity injection system would tend to enhance core debris cooling, reduce core concrete interaction and possibly increase fission product scrubbing. The installation of such a reactor cavity flooding system is judged not practical at Seabrook Station: For Seabrook Station, the design and implementation costs are anticipated to realistically exceed $3M to $4M. This cost is based on a range of cost infOlmation identified in Calvert Cliffs and Indian Point SAMA submittals. Therefore, SAMA #90 is judged not cost-beneficial. It is noted also that Seabrook Station currently has several proceduralized methods to ensure that water is present in the reactor cavity for quenching of core debris.

Page 63 ofl16

United States Nuclear Regulatory Commission SBK-L-1100 II Enclosure These methods include containment injection of the RWST contents and containment injection via the fire protection system through an ILRT flange (refer to SAMA #188). Therefore, it is judged that the current Seabrook Station design meets the intent of SAMA #90.

SAMA #91 - Install a passive containment spray system.

Assessment: Installation of a passive containment spray system would improve containment spray reliability and thus improve fission product scrubbing and reduce long telm containment pressure challenges. The installation of a passive containment spray system is judged not practical at Seabrook Station. Design and implementation costs are judged to realistically exceed $3M to $6M. This cost is based on cost infOlmation identified in Sharon Harris SAMA submittal. However, this SAMA is related to other Seabrook Station SAMAs, for which PRA Case CONTOI was used to estimate the upper bound (UB) risk reduction benefit of $3 10K ($810K if assuming a 2.6 factor increase) when assuming the containment spray system is guaranteed success. Based on the above, SAMA #91 is not cost-beneficial.

SAMA #93 - Install an unfiltered, hardened containment vent.

Assessment: Installation of a hardened containment vent would tend to increase decay heat removal capability for non-ATWS events, without scrubbing released fission products. The scope of this SAMA is to provide a decay heat release path via containment venting. At Seabrook Station, decay heat removal function to protect contaimnent is already provided via anyone of four steam generators, one of two trains ofRHR and one of two trains of containment spray. Design and implementation costs associated with adding a high capacity vent (designed for decay heat removal) are judged to realistically exceed

$3M. Based on the CUlTent plant design, adding yet an additional heat removal capability would not significantly reduce plant risk. This SAMA is related to other Seabrook Station SAMAs, for which PRA Case CONTO! was used to estimate the upper bound (UB) risk reduction benefit of$31OK ($8 10K if assuming a 2.6 factor increase) when assuming the containment spray system is guaranteed success, thus eliminating a challenge to containment overpressure. Based on the above, SAMA #93 is not cost-beneficial. Seabrook Station has SAMG for containment venting at ~ 135 psig to protect containment structure.

SAMA #97 - Create a large concrete crucible with heat removal potential to contain molten core debris.

Assessment: This proposed SAMA infers the installation of a core catcher and cooling system in the reactor cavity. Such a design would tend to increase the reliability of cooling and containment of molten core debris, thus preventing melt-through of the base mat. The installation of such a core catcher/cooling system is judged not practical at Seabrook Station. Design and implementation costs are judged to realistically exceed $40M to $50M. This cost is based on a range of cost infOlmation fi'om the Indian Point, Farley and Calvelt Cliff SAMA submittals. Therefore, SAMA #97 is not cost-beneficial. It is noted that Seabrook Station currently has several proceduralized methods to ensure that water is present in the reactor cavity for quenching of core debris. These methods include containment injection of the RWST contents and containment injection via the fire protection system through an ILRT flange (refer to SAMA # 188). Industry severe accident progression studies show that quenching of the core melt debris with water delivered to the cavity will effectively limit the core concrete interactio~. Therefore, ids judged that the CUlTent Seabrook Station design meets the intent of SAMA #97.

Page 64 of 116

United States Nuclear Regulatory Commission SBK-L-II001/Enclosure SAMA #98 - Create a core melt source reduction systeIn.

Assessment: The scope of this SAMA is to increase cooling and containment of molten core debris.

Refractory material would be placed undemeath the reactor vessel such that a molten core faliing on the material would melt and combine with the material. Subsequent spreading and heat removal from the vitrified compound would be facilitated, and concrete attack would not occur. The installation of this SAMA is judged not practical at Seabrook Station. Design and implementation costs are judged to realistically exceed $40M to $50M. This cost is based on a range of cost infOlmation from the Indian Point, Farley and Point Beach SAMA submittals. Therefore, SAMA #98 is not cost-beneficial. It is noted that Seabrook Station cUtTently has several proceduralized methods to ensure that water is present in the reactor cavity for quenching of core debris. These methods include containment injection of the RWST contents and containment injection via the fire protection system through an ILRT flange (refer to SAMA # 188). lndustty severe accident progression studies show that quenching of the core melt debris with water delivered to the cavity will effectively limit the core concrete interaction. Therefore, it is judged that the current Seabrook Station design meets the intent of SAMA #98.

SAMA #99 - Strengthen primary/secondary containment, e.g., add ribbing to containment shell.

Assessment: The scope of this SAMA is to add stiffening to the containment structure so as to reduce its probability of over-pressurization failure. As mentioned in Seabrook Station's response to RAI 1 c, Seabrook Station's cutTent containment is very large and very strong, with median failure pressure of approximately 215 psig. The installation of stlUctural ribs to effectively increase the strength of the containment is judged not practical at Seabrook Station. Design and implementation costs are judged to realistically exceed $1 OM. This cost is based on costinfOlmation from the Point Beach SAMA submittal.

In addition, this SAMA is related to other SAMAs, for which PRA Case CONTO I was used to estimate the upper bound (UB) risk reduction benefit of$310K ($810K if assuming a 2.6 factor increase) when assuming the containment spray system is guaranteed success, thus eliminating a challenge to containment overpressure. Based on the above, SAMA #99 is not cost-beneficial.

SAMA #100 -lncrease depth of the concrete basemat or use an alternate concrete material to ensure melt-through does not occur.

Assessment: The function of this SAMA is similar to SAMAs 97 and 98 and is focused on reducing the probability of base mat melt-through. Seabrook Station's basemat consists of approximately 10 feet of concrete. Design and implementation costs to increase the depth of concrete are judged to realistically exceed $SM. This cost is based on cost infOlmation from the Indian Point and Farley SAMA submittals.

Therefore, SAMA # 1 00 is not cost -beneficial. It is noted that Seabrook Station currently has several proceduralized methods to ensure that water is present in the reactor cavity for quenching of core debris.

These methods include containment injection of the RWST contents and containment injection via the fire protection system through an ILRT flange (refer to SAMA #188). lndustry severe accident progression studies show that quenching of the core melt debris with water delivered to the cavity will effectively limit the core concrete interaction. Therefore, it is judged that the cunent Seabrook Station design meets the intent of SAM A #100.

SAMA # 101-Provide a reactor vessel exterior cooling system.

Assessment: The scope of this SAMA is to increase the potential to cool a molten core before it causes vessel failure, by submerging the lower head in water. Design and implementation costs of a reactor Page 65 of 116

United States Nuclear Regulatory Commission SBK-L-llOOI/Enclosure vessel external cooling system are judged to realistically exceed $3M and installation is not practical for Seabrook Station. Therefore, SAMA #101 is not cost-beneficial. It is noted that Seabrook Station cUlTently has several proceduralized methods to ensure that the reactor cavity can be flooded to provide a means of extemal vessel cooling. These methods include containment injection of the RWST contents and contaillllent injection via the fire protection system through an ILRT flange (refer to SAMA # 188).

Therefore, it is judged that the current Seabrook Station design meets the intent of SAM A #101.

. SAMA #102 - Constmct a building to be cOffilected to primary/secondmy containment andmaintaiued at vacuum.

Assessment: The focus of this SAMA is to reduce the probability of containment over-pressurization.

As mentioned in RAl 1 c, Seabrook Station's current containment is very large and very strong, with median failure pressure of approximately 215 psig. The CUl1'ent containment design includes a "double" contaiument, with a negative pressure ammlus between the primmy and secondmy containment stmctures to contain leakage. The installation of yet an additional stmctural to effectively reduce overpressure challenges to the primary containment is judged not practical at Seabrook Station. Design and implementation costs are judged to realistically exceed $lOM. In addition, this SAMA is related to other SAMAs, for which PRA Case CONTO 1 was used to estimate the upper bound (VB) risk reduction benefit of $310K ($810K if assuming a 2.6 factor increase) when assuming the containment spray system is guaranteed success, thus eliminating a challenge to containment overpressure. *Based on the above, SAMA #102 is not cost-beneficial.

SAMA #107 - Install a redundant containment spray system.

Assessment: The focus of this SAMA is to increase containment heat removal ability. Seabrook Station cUlTently has a redundant (two-train) contaiument spray and heat removal system. The installation of yet an additional containment spray system to effectively remove contaiument heat and reduce overpressure challenges to the primaty containment is judged not practical at Seabrook Station.

Design and implementation costs are judged to realistically exceed $3M to 4M. This cost is based on our original SAMA assessment and inspection of cost infonnation fi'om the Indian Point SAMA submittal ($61M). In addition, this SAMA is related to other Seabrook Station SAMAs, for which PRA Case CONTO! was used to estimate the upper bound (DB) risk reduction benefit of$310K ($810K if assuming a 2.6 factor increase) when assuming the containment spray system is guaranteed success, thus eliminating a challenge to containment overpressure. Based on the above, SAMA #107 is not cost-beneficial.

SAMA #110 - Erect a balTie!' that would provide enhanced protection of the containment walls (shell) from ejected core debris following a core melt scenario at pressure.

Assessment: The focus of this SAMA is to reduce the probability of containment failure. The

. installation ofbal1'iel' walls to protect the containment shell from ejected corium would have minimal (if any) benefit on improving containment perfonnance. It is noted that the reactor is located within the reactor cavity structure. This provides significant protection of the containment shell during a melt ejection. According to Seabrook Station severe accident phenomena assessments, the majority of the core remains within the cavity area and containment energetic failure is Imlikely. Corium attack on the containment shell structure has not been shown to be a severe accident issue at Seabrook Station. Design and implementation costs of protection walls are judged to realistically exceed $:3M to $4M and would not be practical for Seabrook Station. Based on the above, SAMA #110 is not cost-beneficial.

Page 66 of 116

United States Nuclear Regulatory Commission SBK-L-llOOI/Enclosure RAI Section 4 SAMARAI4a

4)

Provide the following information concerning the Level 3 analysis:

a. Provide the breakdown ofthe baseline population dose-risk (person-rem/yr) and off'site economic cost-risk (OECR, in $/yr) by release categOlY and the total.

NextEra Energy Seabrook Response to SAMA RAI 4a The following table presents the breakdown of the baseline population dose risk and off'site economic cost risk by accident category and the total.

Population dose risk (person-remlyr) 0-50 miles Offsite economic cost risk ($/yr) 0-50 miles

,\\.,. :

"Accident Categ~ry:

Population dose risk (person-remlyr) 0-50 miles Oftsite economic cost risk (S/yr) 0-50 miles Population dose risk. (person-rcmlyr) 0-50 miles Offsite economic cost risk ($I)~)

0-50 miles 1.01 1,234

';.:i, SE-2,

0.16 197 0.22 298 0.59 119

• • • c;,

SE-3 :.:'.;

4.31 12,559 0.0022 0.00 0.0033 0.80 9

2.012

.' :~,~; I.

LL-3 LL-4 2.98 0.64 4,106 3,003 10.70 23,534 The table shows that 40% of the total baseline dose risk and 53 % of the cost risk is from the highest risk category, SE-3. 68% of the total baseline dose risk and 71 % of the cost risk is from the two highest risk categories, SE-3 and LL-3.

Page 67 of 116

United States Nuclear Regulatory Commission SBK-L-llOO I / Enclosure SAMARAJ 4b

4)

Provide the following information conceming the Level 3 analysis:

b. ER Section F.3.4.3 explains that the Cobalt inventOlY was based on the MACCS2 sample problem A, multiplied by the ratio ofthe Seabrook Station projected future power to the reference power (3659 MW /3412 MW). The ER also states that the core inventOlY was estimated using ORIGEN2.1. ClarifY why the cobalt inventOlY required correction. The statement is confusing in that: 1) if a Seabrook Station specific calculation was performed, why was conecting the cobalt required and 2) if sample problem A was scaled for cobalt, why not for iodine? If a Seabrook Station specific core inventOlY was not calculated, quantitatively discuss the impact oflong-lived isotopes that are cycle specific (such as Sr-90, Cs-134 and Cs-137) and not just power-related.

NextEra Energy Seabrook Response to SAMA RAJ 4b A Seabrook-specific core inventory was calculated using ORIGEN2.1. The calculation included all of the nuclides which were patt of MACCS Sample Problem A (and many other nuclides as included in ER Table F.3.1), except Cobalt-58 and -60. The ORIGEN fission product calculation results did not provide detailed isotope inventories of Cobalt-58 and -60 because Co-58 and Co-60 are not core fission products, rather they are activation products fi'om the stainless steel and Inconel components of the RCS pressure boundary materials. For analysis completeness, the isotope inventories of Co-58 and Co-60 were added to the calculated Seabrook specific core inventOlY. The basis for the Cobalt isotopes' inventory was the MACCS Smnple Problem A inventory, conected by the ratio of Seabrook's power level to the MACCS Sample Problem A power level. All other nuclides are represented by their Seabrook specific core inventory.

Adding the Cobalt inventOlY (as described above) to the ORIGEN2.I-calculated Seabrook specific core inventOlY, increases both the total baseline population dose risk and off site cost risk by 0.1 %.

SAMARAJ4c

4)

Provide the following infolmation conceming the Level 3 analysis:

c. Sensitivity analyses are presented in ER Section F.8.4. Provide a quantitative discussion of the results of each of the sensitivity analyses (i.e., provide the percent change in population dose-risk and OECR). Also, discuss the sensitivity of the SAMA results to the population projection assumptions.

N extEra Energy Seabrook Response to SAMA RAJ 4c The following table provides the percent change in total population dose risk and offsite economic cost

. risk (OECR) for each sensitivity analysis.

Page 68 of 116

United States Nuclear Regulatory Commission SBK-L-I 100] IEnclosure Annual Met Data Set Release Height, DPLHlTE Release Heat, DPLHEAT Wake Effects, SIGYINIT, SIGZINIT Evacuation Speed, ESPEED Evacuation Preparation Time, DLTSHL

. Seabrook Statio~; SAMA S~~sitivity 'ri:~SUlts *.

'.'.,' :'. *.****L** *.

._ "",; '.:Jt Inpuimicussion

!iatioto 59-Mile Output DiSCU,S,S.iOU

, Iiaseline Population

_, '. " ~

Dose/CosiRlSk.i,

,.. ","...*.*....: '-;( "

Each year 2004-2008 Baseline assumed release from top of containment vessel.

Releases at ground-level, 25% up containment, 50%

up containment and 75% up containment considered.

Dose ~ 87% (2007) to 95% (2004)

Cost ~ 88% (2008) to 97% (2004)

Dose ~ 97% to No change (note a)

Cost ~ 96% to 99%

Baseline assumed Dose~ 88% to 98%

no heat (ambient).

I and 10 MW heat Cost ~ 91% to 99%

released with each release plume segment for each accident category.

Baseline Dose ~ 99% to 101 %

determined from containment Cost ~ No change to building.

]01%

Oncedainty due to proximity of other buildings. Base case wake size halved and doubled.

One* half and Dose = 96% to 103%

double base case evacuation speed.

Cost = No change Base case was 2 Dose = 91% to 98%

hours (MACCS Cost = No change sample problem A)

One-half and Page 69 of 116 2005, maximum dose ~isk and cost risk year, chosen as baseline.

Decrease in release height increases near-release deposition, Larger downwind population affected by relatively depleted plume. Minimum risks for ground-level release, increasing with increasing release height to top of containment.

Buoyancy associated with increasing heat results in less ground.level consequences near the release. The risk from some accident categories (e.g., LE-l, LE-2) is concentrated near the release, resulting in decreasing risk with increasing release heat; the risk imp0l1ant category, SE-3, does not exhibit this behavior. Total risk increases with decreasing heat.

Minor changes very near release.

Smaller wake results in less plume dispersion near the release. HalVing wake size results in maximum risk Increases ill dose risk as evacuation speed decreases. 0-10 mile risk is minor contributor to 50-mile risk.

Dose generally increases (small changes) or remains the same with increasing preparation time. For categories with ma.ior releases early,

United States Nuclear Regulatory Conunission SBK-L-1l001/Enclosure Pa~~Dieter(MACCS Input Para;netiiriu -:,

CAPS) ',.

Warning to Evacuate (General Emergency Declaration) Time, OALARM Fraction of Population Evacuating,

EZWTFRAC, EZLASMOV Meteorology specification in last spatial segment, double base case preparation period considered (with no change in evacuation speed).

Emergency declaration dependent on accident progression; core uncovery from MAAP simulations used as declaration marker. One-half and double base case preparation period considered_

An important contributor to category SE-3 is a seismically induced station blackout Sensitivity considers that the population does uot evacuate during an SE-3 occun-ence_

Rainfall imposed at all times from 40 to 50 miles from release to force Dose; 94% to 97%

Cost; No change Dose ; 104%

Cost; No change Dose ; 86%

Cost; 83%

Page 70 of 116 the dose can decrease with increasing preparation time if evacuation and.

. plume release are simultaneous; in such a case the dose while sheltered (preparing to evacuate) can be less than the dose while evacuating. The larger total dose risk reduction corresponds with doubling of preparation time.

Changing preparation time had a minor effect on most accident category risks, but larger effect on categories with early releases whose risk is concentrated near the release (e,g, LE-I, LE-2), The risk important category, SE-3, exhibits less than 0.5% change with these changes in preparation

times, Similar behavior as changes in Evacuation Preparation Time (see above). The risk important category, SE-3, exhibits less than 0.5% change with decrease in alarm time but a 6%

increase with a doubling of alarm time (from 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br /> to 28 hours3.240741e-4 days <br />0.00778 hours <br />4.62963e-5 weeks <br />1.0654e-5 months <br />) because evacuation would then coincide with a portion of the release, No evacuation for accident categOlY SE-3 results in a dose-risk increase in lhat category of 9%. The increase on total dose-risk rellects the fraction due to SE-3 Entire decrease is due to removing perpetual rainfall (wet deposition) and specifying measured meteorology in ring from 40 to 50 miles from site.

United States Nuclear Regulatory COJrunission SBK-L-1I00l lEnclosure Seab~ookStatio;'SAMA Sensitivity Res~us

'~ '-.

~

Par~rri~te~ OWACCS ",Input Discussion

'InplI.fpal'amctel'in CAP~L -'

~"

LIMSPA conselvative popUlation exposure for base case. Sensitivity allows 40-50 mile meteorology to temporally follow the onsite meteorology.

Ra~~ to SO-Mile '

Base.\\ine'Popuiati6n Dose/Cost Risk Note (a) - "No change" indicates < 0.5% change in risk.

The baseline modeling conservatism of specifying rainfall in the spatia! ring from 40-50 miles is seen, to more than balance any increases that might be due to alternative specification of release parameters.

The population distribution was exponentially projected to 2050 by applying the county (or counties) specific growth rate applicable to each of the 160 population sectors (16 directions

• 10 distance rings).

As population increases both dose and off site cost risks increase. A linear population projection would result in a lower 2050 population projection and thus lower risks. See RAI response 4h for fmiher discussion of population projections.

SAMARAI4d

4)

Provide the following infonnation concerning the Level 3 analysis:

d. Three SECPOP2000 code enors have been publicized, specifically: 1) incolTect colunm fOimatting of the output file, 2) inconect 1997 economic database file end character resulting in the selection of data from wrong counties, and 3) gaps in the 1997 economic database numbering scheme resulting in the selection of data from wrong counties. Address whether these errors were corrected in the Seabrook Station analysis. If they were not corrected, then provide a revised cost-benefit evaluation of each SAMA with the en'ors cOITected.

NextEl'a Energy Seabrook Response to SAMARA! 4d These SECPOP enors were corrected in the Seabrook Station analysis.

(1) MACCS2 requires spatially distributed agriculture and economic data (the fraction of land devoted to farming, annual fann sales, fraction offarm sales attributed to dailY production, and propeliy values of farm and non-farm land). The SECPOP output file (MACCS input file) containing this information was in a different format than the MACCS code expects. The SECPOP output file was modified to confonn to MACCS input requirements.

Page 71 ofJl6

United States Nuclear Regulatory Commission SBK*L*IIOOl / Enclosure (2) and (3) SECPOP2000 accesses data from the 1997 National Census of Agriculture. The version 3.12.01 data file accessed by SECPOP2000 for that information, County97.dat, was modified to correct two enOl'S that caused SECPOP to select data for inconect counties. These enol'S are sometimes refened to as the missing notes parameter error (inconect database file end character) and the missing county numbers error (gaps in the database numbering scheme).

SAMARAI4e

4)

Provide the following information conceming the Level 3 analysis:

e. ER Section F.3.4.3 states that release fractions for accident categories LE-2, LE-3, SE-2, SE-3 and, LL-5 were taken from Seabrook Station original analyses and all others were from Seabrook Station MAAP simulations. Clarify what this means, and specifically address how release fractions were developed for the original analyses.

NextEra Energy Seabrook Response to SAMA RAI 4e The release fi'actions for release categories LE-2, LE-3, SE-2, SE-3, and LL-5 are based on an assessment of release fractions from the original analyses to support the Seabrook Station Probabilistic Safety Assessment (SSPSA). The basis for these original release category source terms is provided below.

The 2009 PRA model is currently in the process of being updated and is scheduled to be issued in 20 II.

The scope ofthe PRA update will include a reassessment of source telms along with update of other PRA elements, e.g., update to pre-initiator (latent) human actions. Based on preliminary update results, including more CUITent source terms, no changes to the overall SAMA conclusions are expected. An update to the SAMA evaluation will be provided in the next ER annual report.

LE-2, Release S7 A-R: This release category contains sequences involving a large contailllllent bypass.

The sequence is identified as the "V -sequence" based on W ASH-1400 terminology. The release path explicitly considered is a RHR pipe break in the RHR vaults as a result of failure of the RHR MOV isolation valves which pressurizes the low pressure RHR system to the RCS pressure. The break elevation is above the water level accumulated from discharge of the RCS and RWS'j' inventories into the vault. The release will not be scrubbed in this case. The release fi'actions and timing are based on WASH-1400, Table 5-1, PWR-2 release category.

LE-3, Release S6-R: This release category contains sequences involving a large containment isolation failure as a result of the purge valves failure to close. It also includes sequences with large pre-existing leaks. The source term release fractions are based on an !DCOR MAAP analysis for the Zion plant.

Based on the design comparison between Seabrook and Zion, the !DCOR analysis was adopted for Seabrook. The !DCOR analysis employed a SBO sequence with a seal LOCA and an isolation failure.

The !DCOR analysis is a single puff release, which was convelied to a three puff release. The stali timing for the first puff is based on !DCOR analysis. The durations and stmt time for the remaining puffs is based on a sequence with a SLOCA and open 8" purge valves.

Page 72 of 116

United States Nuclear Regulatory Commission SBK-L-ll001! Enclosure SE-2, Release S7B-R: This release category contains sequences involving a small containment bypass.

The release path explicitly considered is a RHR pump seal failure due to over-pressurization of the RHR system as a result of failure ofRCSIRHR MOV isolation valves. The failure results in the discharge of the R WST into the RHR vault at a rate greater than the 50 gpm capacity of the RHR vault sllmp pumps.

The release will be scrubbed by a subcooled pool of water in this case. The release fractions are based on a realistic estimate of the decontamination factor of the pool. The start timing is based on the Seabrook specific MAAP analysis for the V -sequence.

SE-3, Release S2-R: TIJis release categOlY contains sequences where the containment fails with ~n early "Type-A" failure which progresses into a large late failure; Because long-telm contailUnent cooling is failed, Type A failures necessarily progress to a larger failure later since the initial opening is not large enough to arrest the pressllre increase. The source telm release fractions are based on sequences involving LLOCA without CBS or EeCS function. The noble gas release fraction for each puff is such that all noble gas is released. The statt timing for the first puff is based on a transient without EFW available.

LL-5, Release S3-R: This release category contains sequences involving a dlY cavity floor (no RWST injection) long term over-pressurization failure of the containment as a result ofloss of containment heat removal. The release fractions are based on analysis of a SBO/seal LOCA sequence. The results of the IDCOR analysis for Zion station were used based on a design comparison between Zion and Seabrook.

These MAAP generated results were modified to account for the additional Te release during core-concrete interactions and the reduction in Cs and I release as a result of the much longer containment failure time. The release start time was defmed as the 70th percentile estimate of contailUUent failure time, based on a weighted average of contributing sequences. The duration of one hour conservatively models a puff release at containment failure.

SAMARAJ4f

4)

Provide the following information concerning the Level 3 analysis:

f.

Section 2.6.1 of the ER indicates a year 2000 50-mile total popUlation of 4,157,215 (Tetra Tech 2009a) while Table F.3.4.1-1 indicates a total population of 4,232,394. Clarify the discrepancy.

NextEra Energy Seabrook Response to SAMA RAJ 4f The choice of distribution centroids and differences in how populations within the 50-mile radius are

. calculated account for the discrepancy.

One reason for the discrepancy is that the population distribution in Section 2.6.1 was centered on the (longitude, latitude) coordinates for the Seabrook Station used in the SECPOP2000 coordinate data base (42°53'53" 70°51'05"). The Table F.3.4.l-1 population distribution was centered at the reactor (42°53'54" 70°51 '03"). The difference in total 50-mile population as a result of this difference in the distribution centroid's locations is 0.09%.

Page 73 ofll6

United States Nuclear Regulatory Commission SBK-L-11 001 / Enclosure More importantly, the 50-mile population in Section 2.6.1 was calculated using U.S. Census block group data and methodology consistent with the Environmental Justice analysis. When calculating the 50-mile population, block groups that are bisected by the 50-mile radius are clipped, and the percent of census block land area within the 50-mile radius is assumed to be equivalent to the percent of the census block population within the 50-mile radius (i.e., if25 percent ofthe block group area was within the radius, the analysis calculated 25 percent of the total block group population as being within the radius).

SECPOP2000 data was used to detennine the 50-mile population for the Level 3 analysis (Table F.3 04.1-1).SECPOP uses block group data, but assigns the entire block group population to a point corresponding to the centroid of the block. If the centroid of the block falls within the 50-mile radius the entire population of the block is included in the 50-mile radius. If the centroid falls outside the 50-mile radius, the entire population is excluded.

Finally, the SECPOP population was adjusted upward to account for transient populations in the MACCS2 analysis. Therefore, the 50-mile population in Table F.3A.l-l includes an estimate of the transient population, and Section 2.6.1 considers only residents.

In this case, the population used in the Level 3 analysis is 1.8% greater than that of Section 2.6.1.

SAMARAI4g

4)

Provide the following infonnation concel'lling the Level 3 analysis:

g. The ER provides no discussion of the effects of sea-breeze circulation on radionuclide deposition and whether this sea-breeze effect was factored into the MACCS2 calculations. ClarifY whether sea-breeze effects were considered in the SAMA evaluation and, if not, provide an assessment of the sea-breeze effect on the results of the SAMA evaluation.

NextEra Energy Seabrook Response to SAMA RAI 4g Sea-breeze effects are considered in the SAMA evaluation to the extent they are included in the onsite meteorological data (used in the MACCS2 calculations). In response to this RAI, fmiher consideration of sea-breeze effects at the site, as discussed below, are based on the quantitative description of sea-breeze and onshore gradient flow from the Seabrook UFSAR, Rev.12, Section 2.3, Meteorology, pages 32-37.

The two major mechanisms by which sea-breezes could affect the Level 3 MACCS2 calculations are the fOlmation of a thennal internal boundmy layer (TlBL) and a mixing front between the sea-breeze and inland winds. During the summer when the land surface is wanner than the air over the sea, air heated from contact with the land rises and is undercut by cooler denser breezes from the sea. The boundary between these two distinct air layers caused by this difference in air temperature (or density) is the thennal internal boundary layer (TIBL). When these onshore breezes meet inland wind fields, which differ in magnitude and direction from the sea breeze, those differences result in something akin to a weather front with increased turbulence (mixing) along the mixing front betWeen the two wind systems.

Page 74 ofl16

United States Nuclear Regulatory Commission SBK-L-!100J fEnclosure Mixing Front - The mixing fi'ont results in increased phune mixing and dispersion which would, in turn, result in lower population dose. The UFSAR simulates increased mixing (decreased atmospheric stability) below the TIBL by resetting site measured stability classes E through G (most stable classes) to D (less stable) during times ofTIBL occurrence. This stability simulation was not implemented for the SAMA evaluation, thus resulting in a conservative evaluation (less dispersion, greater doses).

Thermal Internal Boundary Layer (TIBL) - The minimum TIBL height is noted in the UFSAR reference as 93 meters (above ground level). The Level 3 release height (top of containment) is 54.6 meters. Thus, the severe accident release plumes would be trapped beneath the TIBL, resulting in limited verticai dispersion and increased ground concentrations (doses). A TIBL was present during 527 hours0.0061 days <br />0.146 hours <br />8.713624e-4 weeks <br />2.005235e-4 months <br /> for the period April 1979 - March 1980 (the period analyzed in the UFSAR). 638 hours0.00738 days <br />0.177 hours <br />0.00105 weeks <br />2.42759e-4 months <br /> (7% of the year) of TIBL fOlmation was found during 2005 (the met year used for the SAMA analysis) by applying the UFSAR's criteria (based on time of year, time of day, wind speed, wind direction and solar radiation) for TIBL formation.

Using additional MACCS2 runs, the effect of TIBL fOlmation was bounded by assuming a summer mixing layer (both morning and afternoon) of 100 meters, the minimum allowable by MACCS2.

Applying this restrictive lid to mixing during the entire surumer (2190 hours0.0253 days <br />0.608 hours <br />0.00362 weeks <br />8.33295e-4 months <br />) bounds the 638 hours0.00738 days <br />0.177 hours <br />0.00105 weeks <br />2.42759e-4 months <br /> of TIBL formation found for 2005. The resulting population dose and offsite economic cost risks were found to increase by 4% and 7% respectively compared to the baseline SAMA case. The sensitivity of the lid height was investigated by specifYing a 110 meter height. This shows that a decrease of 1 0 meters (from 110 to 100 meters) increases the dose and offsite cost risks by 0.2% and 0.5%. A decrease from 100 meters to 93 meters would be expected to show a similar change.

Because the release plumes are trapped beneath the TIBL, no special effect (other than the trapping itself, which is reflected in the MACCS2 runs as described above) is expected on radionuclide deposition.

Given the conservatisms in the above analysis (no accounting for increased mixing with sea-breeze, assuming 25% of aunual hours result in TIBL fOlmation vs. 7% of aunual hours demonstrated for 2005) and the conservatisms built into the baseline analysis (see RAT response 4c and especially the assumption of perpetual rainfall in the 40-50 mile ring), the potential effects of sea-breezes do not change the SAMA

. analysis.

SAMARAI4h

4)

Provide the following infOlmation concerning the Level 3 analysis:

h. Table F.3.4.l-1 indicates that several sector populations extrapolate to zero population in year 2050. For example radius 3 mi to 4 mi, ENE population decreases from 788 to zero). This occurs in several other locations. Clarify why this occurs, and address the potential impact to the SAMA analysis if a more conservative approach were used for extrapolating negative population growths to earlier years.

Page 75 of 116

United States Nuclear Regulatory Commission SBK-L-llOOI/Enclosure NextEra Energy Seabrook Response to SAMA RAI 4h The GIS land layers were not detailed enough to account for the existence of some small islands; GIS water sectors were projected as zero populations. Also, the direction distribution used in the 2050 projection was slightly offset from the existing population, resulting in some sectors considered all water (and thus zero' population) when, in fact, a portion of them includes the coastline and therefore has a population. The population projections have been refined to account for the above and to include the most recent county population growth rates.

The centroid of the refined population distribution was taken as latitude 42°53'53" and longitude 70°51 '05". These coordinates conespond with those from the SECPOP2000 coordinate data base for the site and are the same ones used in Section 2.6 of the ER (see RAl response 4f).

The table below (which is analogous to Table F.3.4.1-1) gives the refined year 2000 and projected 2050 population distributions. The total projected 2050 population shown in this Table, 4,991,410, is 96.3% of that in ER Table F.3.4.1-l.

A sensitivity run of the Level 3 model was performed, replacing the 2050 population distribution used in the ER (Table F.3.4.I-I) with that from the table above. The resulting total population dose and offsite economic risks were 5% and 6% less than the ER baseline risks. Accident category population dose risks declined between 4 and 6% and accident categOly oft'site economic risks declined between 4 and 7%

from those in the ER.

From Radius (mil~)

o o

o o

o o

o o

o o

o o

o o

o o

To Radius

, (miles) 1 1

2 2

2 2

2 2

2

~.

','2000

, ~,

': Populatiol).,

N 1

24 NNE 2

0 NE 3

29 ENE 4

0 E

5 0

ESE' 6

0 SE 7

163 SSE 8

68 S

9 139 ssw 10 65 SW II 10 WSW 12 234 W

13 0

WNW 14 144 NW 15 0

NNW 16 12 N

17 54 NNE 18 36 NE 19 143 ENE 20 12888 E

21 4257 ESE 22 5149 SE 23 188 Page 76 of 116 Projected,

, 2050 Population '

34 o

41 o

o o

228 95 195 91

14.

328 o

202 o

17 76 50 200 18045 5960 7209 263

United Slates Nuclear Regulatory Commission SBK-L-ll0011 Enclosure From Radius ' To Radius

' :Direction '

(miles)

(miles)

I 2

' SSE 1

2 S

I 2

SSW I

2 SW 1

2 WSW 1

2 W

I 2

WNW 1

2 NW I

2 NNW 2

3 N

2 3

NNE 2

3 NE 2

3 ENE 2

3 E

2 3

ESE 2

3 SE 2

3 SSE 2

3 S

2 3

SSW 2

3 SW 2

3 WSW 2

3 W

2 3

WNW 2

3 NW 2

3 NNW 3

4 N

3 4

NNE 3

4 NE 3

4 ENE 3

4 E

3 4

ESE 3

4 SE 3

4 SSE 3

4 S

3 4

SSW 3

4 SW 3

4 WSW 3

4 W

3 4

WNW 3

4 NW 3

4 NNW 4

5 N

4 5

NNE 4

5 NE 4

5 ENE 4

5 E

4 5

ESE 4

5 SE 4

5 SSE 4

5 S

Ie:"

C~~e 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 Page 77 of 116

,, ~~ojected ';,

\\

, 2000 )

-C',, 2050 il'opulationi:

' p~pulatiou 584 818 532 745 1074 1500 1371 1904 474 664 546 764 410 574 385 539 226 316 462 647 2007 2810 2129 2981 1531 2144 83 II6 13 18 II04 1480 535 686 880 1140 1149 1426 471 581 835 1085 5180 7253 122 171 283 396 247 346 1452 2033 3328 4660 3822 5351 788 1103 0

0 0

0 475 586 17035 21030 824 1017 728 899 414 511 493 613 390 546 163 228 265 371 584 818 1290 1806 849 1189 1981 2774 0

0 0

0 0

0 907 1120 570 704 1624 2005

United States Nuclear Regulatory Commission SBK*L*II001/Enc\\osure

Fro~i' Rri*d*i~~*. ;,1'9 Rad[qs *...

(miles)

' ". (miles),

4 5

4 5

4 5

4 5

4 5

4 5

4 5

5 10 5

10 5

10 5

10 5

10 5

10 5

10 5

10 5

10 5

10 5

10 5

10 5

10 5

10 5

10 5

10 10 20 10 20 10 20 10 20 10 20 10 20 10 20 10 20 10 20 10 20 10 20 10 20 10 20 10 20 10 20 10 20 20 30 20 30 20 30 20 30 20 30 20 30 20 30 20 30 20 30 20 30 20 30

.'.~

Di~.~~tion \\

Code SSW 74 SW 75 WSW 76 W

77 WNW 78 NW 79 NNW 80 N

81 NNE 82 NE 83 ENE 84 E

85 ESE 86 SE 87 SSE 88 S

89 SSW 90 SW 91 WSW 92 W

93 WNW 94 NW 95 NNW 96 N

,97 NNE 98 NE 99 ENE 100 E

101 ESE 102 SE 103 SSE 104 S

105 SSW 106 SW 107 WSW 108 W

109 WNW 110 NW 111 NNW 112 N

113 NNE 114 NE 115 ENE 116 E

117 ESE 118 SE 119 SSE 120 S

121 SSW 122 SW 123 Page 78 of 116 2000.,....... ' Projected Population "

2050 '

o Populaiion 481 594 4119 5085 2924 3654 383 536 460 644 197

276, 640 896 4863 6809 11105 15548 1837 2572 0

0 0

0 0

0 0

0 8149 10060 8723 10769 13597 16786 8972 11076 10774 13690 3490 4886 3040 4256 12762 17868 6120 8569 18596 25561 37177 47033 1257 1562 0

0 0

0 0

0 2645 3265 6981 8618 24518 30269 25814 31868 84691 104555 57568 75952 23379 32733 17017 23826 9182 12859 26181 36792 48973 64004 13619 16372 414 498 0

0 0

0 0

0 4in 5157 27814 34338*

196181 242194 132772 154381 243547 272808

United States Nuclear Regulatory Connnission SBK-L-l1001/Enclosure

..... To Radius From Radius Direction..

(I11iles)

(miles)

)1, 20 30 WSW 124 20 30 W

125 20 30 WNW 126 20 30 NW 127 20 30 NNW 128 30 40 N

129 30 40 NNE 130 30 40 NE 131 30 40 ENE 132 30 40 E

133 30 40 ESE 134 30 40 SE 135 30 40 SSE ri6 30 40 S

137 30 40 SSW 138 30 40 SW 139 30 40 WSW 140 30 40 W

141 30 40 WNW 142 30 40 NW 143 30 40 NNW 144 40 50 N

145 40 50 NNE 146 40 50 NE 147 40 50 ENE 148 40 50 E

149 40 50 ESE 150 40 50 SE 151 40 50 SSE 152 40 50 S

153 40 50 SSW 154 40 50 SW 155 40 50 WSW 156 40

'50 W

157 40 50 WNW 158 40 50 NW 159 40 50 NNW 160 Total Page 79 of 116

/'" ~600 ' *.,' *.*.. *.Projected,.-

I. -p~~~i~iion

"':. 2050 ',.

P~plllation 67747 88542 67400 94038 21508 30253 8059

. ))298 25886 36455 30662 37589 24595 29567 I

1 0

0 0

0 0

0 0

0 0

0 54682 69289 859576 942784 162839 163980 146085 170127 103784 140850 95495 139943 19164 28890 15379 22046

))555 14999 54548 65576 158 190 0

0 0

0 0

0 0

0 0

0 190470 218152 812351 879674 123527 130780 52054 59636 27269 37002 37194 53875 24562 37213 10562 15928 4236469 4991410

United States Nuclear Regulatory Commission SBK-L-llOOl/Enclosure RAI Section 5 SAMARAI5a

5)

Provide the following with regard to the SAt\\1A identification and screening process:

a. ER section F.5.1 explains that "the cun-ent plant procedures and training meet CUlTent industty standards" and that there "were no additional specific procedures improvements identified that would affect the result of the HEP calculations" and that therefore "no SAMA items were added.

to the plant-specific list of SAMAs as a result of human actions with risk reduction wOlth greater than 1.005." Describe other mitigation options (besides procedure and training improvement) for addressing each of the human error events that appear in importance Tables F.3.1.1.1-2 and F.3.2.1-1 (e.g., installing or improving automatic control, additional alarms) and provide justification for not considering a non-procedure/training SAMA to address these basic events.

NextEra Energy Seabrook Response to SAMA RAJ Sa Levell Human Actions: The highest CDF-related risk reduction wOlth (RRW) basic event in Table F.3.1.1.1-2 is operator action HH. OSEP I.F A - Operator Fails to Close SEPS Breaker from MCR. The context of this operator action is loss of offsite power with failure of both EDGs and no immediate recovery of power. Once the breaker for SEPS is manually closed to align the bus, the load sequencer connects loads nonnally. Success of SEPS means that seal cooling has been re-established and the potential for RCP seal LOCA minimized/avoided. This basic event has a CDF RRW value of 1.0323.

PRA case study OSEPI was perfOlmed to evaluate the maximmn dollar benefit assuming elimination of 100% of the risk of this basic event, i.e., basic event was set to guaranteed success (value of 0.0) from its nominal value of3.1E-02. Based on this conservative PRA case study, the maximum benefit is $12.SK (nominal benefit at 7% discount rate) and $24K (upper bound benefit). A hardware change such as installing an automatic breaker control or installing an additional alann or indication to improve the human action is estimated to cost >$1 OOK. Based on this result and given that all other CDF-related human action basic events in Table 3.1.1.1-2 have a CDF RRW less than1.0323 (HH.OSEP1.FA), any hardware changes proposed to improve these human actions are judged not cost-beneficial.

Level 2 Human Actions: The highest LERF-related risk reduction worth (RRW) basic event in Table F.3.2.1-1 is operator action HH,XOEFW1.FA - Operator Establishes Feed Flow to Faulted Steam Generator Prior to Significant Release. The context of this operator action is a SGTR accident sequence with EFW available at the onset of core damage: Success ofEFW flow to the ruptured SG will provide water inventOlY for release scrubbing. This basic event has a LERF RRW value of 1.1873. PRA case study XOEFW was performed to evaluate the maximum dollar benefit assuming elimination of 1 00% of the risk of this basic event, i.e., basic event was set to guaranteed success (value of 0.0) from its nominal value of 3.9E-02. Based on this conservative PRA case study, the maximum benefit is $4.1 K (nominal benefit at 7% discount rate) and $7.8K (upper bound benefit). Any hardware changes such as installing an automatic means to maintain/re-establish faulted SG inventOty or installing an additional alarm or indication to improve the human action is estimated to cost >$1 OOK. Based on this result and given that all remaining LERF-related human action basic events in Table 3.2.1-1 have a LERF RRW less than 1.1873 (HH'xOEFWl.FA), any hardware changes proposed to improve these human actions are judged not cost-beneficial. Refer to RAI 2f response for additional evaluation of Level 2 basic events.

. Page 80 of 116

United States Nuclear Regulatory Commission SBK-L-lIOOI/Enclosure SAMARAI5b

5)

Provide the following with regard to the SAMA identification and screening process:

b. Importance Tables FJ.l.1.1-2 and F.3.2.l-1 are not linked to SAMA options except by associated SAMA categOlY (e.g., AC Power SAMAs, Contaimnent SAMAs). It is not always clear, however, how the identified SAMAs address the specific basic events listed (for example, basic events CCE17A.GL and CCE17B.GL. For each basic event identified in the impOliance lists, identify the specific SAMA( s) that address each event and describe how the SAMA(s) address the basic event. For any basic event for which no SAMA is identified, provide justification for not identifying a SAMA(s).

NextEra Energy Seabrook Response to SAMA RAI 5b Tables FJ.1.1.1-2 and FJ.2.1-1 of the SAMA Report provided a link between each basic event and an associated SAMA item "functional" categOlY. For example, basic event FWP37 A.FR, "Turbine Driven Pump FW-P-37A Fails to Run" is identified with SAMAs related to Feedwater & Condensate. In the SAMA repOli, a specific SAMA candidate was not identified to address each basic event due to the basic event's relatively low RR W importance, corresponding low cost-benefit, and the basic event's relationship to the functional SAMA category. In response to this RAI, each basic event was reviewed from a SAMAIcost-benefit perspective. The top 15 basic events related to CDF are provided in the table below. These basic events have an RRW ranging from a high of 1.1713 to a low of 1.0223. Based on a Phase n type assessment of these basic events, no new SAMAs are identified as cost-beneficial. Also, based on the cost-benefit of basic event #15, having an RRW of 1.0223 and a nominal cost benefit of

~$31K and upper bound of $60.6K, it is judged that SAMA candidates associated with basic events having an RRW less than 1.0223 are not cost-beneficial given the minimum estimated cost of$1 OOK to implement plant hardware changes.

The top ranked basic events related to LERF are the same basic events addressed in NextEra response to RAI2f.

Page 81 of 116

United States Nuclear Regulatory Commission SBK*L*llOOI/Enclosure

1. 1 Turbine Driven PUMP FW*P*

FWP37A.FR 37 A fails to run

1. 2 DG* IA fails to run for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> DGDGI A.FR3

1. 3 DG* 1B fails to run for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> DGDGIB.FR3

1. 4 4KV BUS E6 fault EDESWG6.FX

1. 5 Loss of Offiite Power ZZ.SYI.FX subsequent to plant trip 1.1713 1.0774 1.0694 1.0442 1.0391 Feedwater

1. 163 Condensa te SAMAs ACPower

1. 9, #10, #14, SAMAs

1. 155 ACPower

1. 9, #10, #14, SAMAs

1. 155 ACPower Improve Bus SAMAs E6 reliability, eliminate I reduce potential for bus fault ACPower

1. 13, #156, SAMAs

1. 160 Page 82 of 116 TDAFW

$100K (nominal benefit)

Not cost beneficial

$190K CUB benefit) based on assmned OECR Red: -7%

elimination of all Min. Cost: >$250K TDEFWpump failures NOSBO

$155K (nominal benefit)

Not cost beneficial

$295K CUB benefit) based on assumed OECRRed: -14%

elimination of all Min. Cost: >$500K EDG failures NOSBO

$155K (nominal benefit)

Not cost beneficial

$295K CUB benefit) based on assumed.

OECRRed: -14%

elimination of all Min. Cost: >$500K EDG failures E6S

$39K (nominal benefit)

Not cost beneficial

$74.IK (UB benefit) based on* assumed OECRRed: -2%

elimination of all Min. Cost: > lOOK risk associated with Bus fault NOLOSP

$335K (nominal benefit)

Not cost beneflcial

$638K CUB benefit) based on assumed OECRRed: -43%

elimination of all Min. Cost: > 1M risk associated with loss of off sit';

power

United States Nuclear Regulatory Commission SBK-L-11001 / Enclosure

1. 6 ZZ.sY2.FX

1. 7 FWP37A.FSI

1. 8 SEPSDG2A.FR3

1. 9 SEPSDG2B.FR3

1. 10 HH.OSEP1.FA

1. 11 HH.OHPR3.FA Loss of Off'site Power subsequent to LOCA initiator Turbine Driven Pump TURBINE FW-P-37A fails to start on demand I-SEPS-DG-2-A fails to run within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> I-SEPS-DG-2-B fails to run within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> OPERATOR fui!s to close SEPS breaker from MCB OPERA TOR fails to close SEPS breaker from MCB, given SI

. signal 1.0387 1.0376 1.0324 1.0324 1.0323 1.0307 ACPower SAMAs Feedwater Condens.

te SAMAs ACPower SAMAs ACPower SAMAs Seetexf Section F.5,)

See text Section F.5,1

1. 13, #156,

1. 160

1. 163

1. 9, #14, Improve SEPS DO reliability, eliminate potential for SEPS failure

1. 9, #14, Improve SEPS DG reliability, eliminate potential for SEPS failure

1. 9, #14, Provide auto-start and load for SEPS DG

1. 9, #14, Provide auto-start and load forSEPS DG Page 83 of 116 NOLOSP TDAFW SEPES SEPES OSEPI OSEPI

$335K (nominal benefit)

$638K (VB benefit)

OECR Red: -43%

Min. Cost: >lM

$1 OOK (nominal benefit)

$190K CUB benefit)

OECRRed: -7%

Min. Cost: >$250K S39.8K (nominal benefit) $75.8K (VB benefit)

OECR Red: -2%

Min. Cost: >SlOOK.

$39.8K (nominal benefit) $75.8K (VB benefit)

OECR Red: -2%

Min. Cost: >$100K SI2.5K (nominal benefit) $23.9K (VB benefit)

OECRRed: -1%

Min. Cost: >$IOOK

$12.5K (nominal benefit) $23.9K (VB benefit}

OECR Red: -1%

Min. Cost: >$lOOK

,~.,C',

'~Disposit'ion,.

",.::'~'"

Not cost beneficial based on assumed elimination of all risk associated with loss of offsite power Not cost beneficial based on assumed elimination of all TDEFWpump failures Not cost beneficial based *on assumed elimination of all SEPS failures Not cost beneficial based on assumed elimination of all SEPS failures Not cost beneficial based on assumed elimination of all SEPS failures Not cost beneficial based on assumed elimination of all SEPS failures

United States Nuclear Regulatory Commission SBK-L-IlOOI I Enclosure

",;~~::: ~

.. -~ :.. ;,

Basi~ Event 11"

)\\',

1. 12 PORV RC-PCV-456B fails to 1.0300 ECCS Improve PORVRS

$22.8K (nominal RCPCV456B.RS

1. 13 EDESWGS.FX

1. 14 RCPCV456A.RS

1. 15 HH.ORWMZI.FA reseat SAMAs reliability of benefit) $43.4K (UB 4KV BUS E5 fault PORV RC-PCV-456A fails to reseat OPERA TOR minimizes ECCS flow wi recirc. failure 1.0279 1.0265 1.0223 ACPower SAMAs ECCS SAMAs See text Section F.5.1 PORV reseat benefit)

function, OECR Red: -1%

eliminate Min. Cost: >$IOOK PORV re-seat failure Improve Bus ES reliability, eliminate I reduce potential for bus fault Improve reliability of PORVreseat

function, eliminate PORV re-seat failure Provide hardware change for automatic ECCSflow

control, E6S PORVRS ORWS

$39K (nominal benefit)

$74. IK (UB benefit)

OECR Red: -2%

Min. Cost: > lOOK

$22.8K (nominal benefit) $43.4K (UB benefit)

OECRRed: -1%

Min. Cost: >$IOOK

$31.8K (nominal benefit) $60.6K (UB benefit)

OECR Red: -5%

Min. Cost: >$lOOK Page 84 of 116 Not cost beneficial based on assumed elimination of all PORV reclose failure potential Not cost beneficial based on assumed elimination of all risk associated with Bus fault Not cost beneficial based on assumed elimination of all PORV reclose failure potential Not cost beneficial based on assumed elimination oftbe human failure risk

United States Nuclear Regulatory Commission SBK-L-11 001 I Enclosure SAMARAISc

5)

Provide the following with regard to the SAMA identification and screening process:

c. Importance Tables FJ.1.1.1-2 and FJ.2.1-1 combine the impoltance of internal, fire, and seismic events, so that it is not possible to determine the relative impottance of each basic event for each hazard category. As a result, the SAMA identified to address each event may not address the more important initiator (e.g., fire). Provide a Levell and 2 importance list for each hazard categoty (internal, fire, and seismic) and identify which SAMA(s) address each event. For any basic event for which no SAMA is identified, identify and evaluate a new SAM-A.

NextEra Energy Seabrook Response to SAMA RAISe The Risk Reduction Worth (RRW) impottance values provided in Tables FJ.1.1.1-2 and FJ.2.1-l represent the basic event's total importance contribution to CDFILERF from all hazards. The basic event RR W contribution to individual hazards will be a subset of the total RR W. Based on the response to RAl 5b, it is judged that hardware changes proposed to address individual hazard contributors would not be cost-beneficial based on a conservative minimum cost for a hardware change of$lOOK.

SAMARAISd

5)

Provide the following with regard to the SAMA identification and screening process:

d. Importance Tables FJ.1.1.1-2 and FJ.2.1-l identify only one event (Le., COTK25.RT -

Condensate Storage Tank CO-TK-25 ruptures/excessive leakage) to be related to seismic fragility (based on the basic event descriptions presented). SAMA 162, "Increase the capacity margin of the condensate storage tank (CST)," appears to have been identified to address this event.

However, it is not clear that this SAMA addresses the seismic fragility of the CST since the SAMA is described as increasing the capacity margin of the CST. Provide an assessment of a SAMA to increase the seismic fragility of the CST.

NextEra Energy Seabrook Response to SAMA RAI Sd The CST is seismically qualified and is protected fi'om external events by a lugged, seismic structure.

The CST has a median seismic fragility of 1.65g and a high confidence low probability of failure (HCLPF) of 0.65, without giving credit to the concrete shield structure surrounding the CST, which is judged to have a median seismic capacity of greater than 2.5g. A Phase 2 PRA case study was perfolmed to assess the potential cost-benefit of possible seismic upgrades to the CST. The PRA case (QCSTS) sets the seismic-induced failure of the CST to guaranteed success (assumes CST does not structurally fail during all seismic initiating events. The following results were obtained from PRA case QCSTS:

Page 85 of 116

United States Nuclear Regulatory Commission SBK-L-1I001 fEnclosure Nominal Cost-Benefit:

-$IK Upper Bound Cost-Benefit: -$2K

% Reduction in OECR: <1%

Cost of SAMA: The engineering analysis costs and cost of potential upgrades to increase the seismic capacity of the CST are estimated to exceed $1 OOK.

Based on this case study, any modifications to further increase the seismic capacity of the CST would exceed the upper bound cost-benefit of$2K and thus are judged lIot cost-beneficial.

SAMARAI.Se

5)

Provide the following with regard to the SAMA identification and screening process:

e. SAMA 92, "Use the fire water system as a backup source for the contaimnent spray system," was screened in Table F.6-1 because the contaimnent spray function is not important early. Yet, RCPCV456A.FC and RCPCV456B.FC (Spray Valves fail to open on demand) appear on the LERF impOliance Table F.3.2.1-1 and may also provide benefit in late releases. In light of this, provide an assessment of this SAMA.

NextEra Energy Seabrool. Response to SAMA RAISe LERF lmpOliance Table FJ.2.1-1 basis events RCPCV456A.FC and RCPCV456B.FC refer to modeling ofthe pressurizer Power Operated Relief Valves (PORVs), and not the contaimnent spray valves. The descriptions of basic events RCPCV456A.FC and RCPCV456B.FC inadvertentlyrefen-ed to these valves as PORV spray valves. The PORV function is unrelated to the contaimnent spray function. Therefore, no SAMA assessment ofPORV contribution to contaimnent spray is necessary.

SAMARAISf

5)

Provide the following with regard to the SAMA identification and screening process:

f.

SAMA 143 (Upgrade fire compartment barriers) was screened in Table F.6-1 because the Seabrook Station plant design includes 3-hour rated fire bal1'iers. ClarifY how additional barriers for fire areas were considered and assess the impact that adding additional barriers would have on the SAMA results.

. Page 86 of 116

United States Nuclear Regulato!), Commission SBK-L-llOOI/Enclosure NextEra Energy Seabrook Response to SAMA RAI 5f Additional ban-iers for fire areas were screened out based on the Seabrook Station plant design with 3-hour rated fire barriers. In addition, a review of the fire risk by location in the most recent update supports this screeningthat additional fire baniers would not significantly impact fire risk. Assessment of specific fire barriers is provided in the table below.

,"to.:" <',"

Fire Location '

". ~~".

Control Room Essential Switchgear Rooms Turbine Building Primary Auxiliary Building Ocean Service Water Pumphouse Electrical Tunnels (ET) 52%

41 %

5%

2%

1%

<1 %

Additional fire barriers are not physically possible in the CR.

These rooms are train separated. These scenarios involve fire in a single Essential Switchgear room, with subsequent fire-independent hardware failures.

Additional baniers would have no impact on these scenarios.

The key contribution to fire risk from the Turbine Bldg is the potential for loss of offsite power. In addition, fire in the TB is a minor contribution to overall fire CDF. Thus, additional balTiers would no significant impact on [u'e I*isk.

While PCC pumps are in the same area in the P AB, additional fire barriers would do little to reduce the overall fire CDF. This is due to the detailed fire modeling perfonned for this area.

While SW ocean pumps are in the same area in the SW pumphouse; additional fire balriers would do little to reduce the overall fire CDF. This is due to the redundant SW cooling tower pumps, physically separated from the ocean pumps.

These tunnels are train separate.

Page 87 of116

United States Nuclear Regulatory Commission SBK-L-IIOOI/Enclosure SAMARAISg

5)

Provide the following with regard to the SAMA identification and screening process:

g. SAMA 79, "Install bigger pilot operated relief valve so only one-isrequired," was screened in Table F.6-1 because the intent of the SAMA has already been implemented. However, the Phase 1 Disposition colunm explains that 2-of-2 PORVs is needed for intermediate head Safety Injection. In light of this success criteria, provide an assessment of this SAMA.

NextEm Energy Seabrook Response to SAMA RAI Sg The.context ofSAMA #79 is to increase ihe capacity of the pressurizer PORVs such that opening of only one PORV would satisfy the feed and bleed success criteria for all loss offeedwater-type sequences. At Seabrook, only one PORV is needed for feed and bleed if feed is provided by one of two high head charging pumps. However, opening oftwo PORVs is needed if feed is provided by one of two safety i1*ction (S1) pumps. A Phase II PRA case study was performed to assess the potential cost-benefit of possible upgrades/replacement of the PORVs to increase their capacity such that opening of only one PORV would satisfy the feed and bleed success criteria under all combinations of feed. The PRA case (FWOI) conservatively eliminates loss offeedwater events and this reduces the risk contribution from failure to feed and bleed. The following results were obtained from PRA case FWOl:

Nominal Cost-Benefit:

-$73K Upper Bound Cost-Benefit: -$140K Reduction in OECR: -6%

SAMA Cost: Modifications to the PORV s and associated safety analysis to increase their flow capacity is estimated to exceed $250K.

Based on this case study, any modifications and safety analysis needed to increase the flow capacity of the PORVs is judged to exceed the upper bound cost-benefit of $140K and thus are judged not cost-beneficial.

SAMARAISh

5)

Provide the follo.wing with regard to the SAMA identification and screening process:

h. SAMA 64, "Implement procedure and hardware modification for a component cooling water

. header cross-tie," was screened in Table F.6-1 because a cross-tie already exists to support a maintenance activity. Clarify whether exiting plant procedures provide foi* the cross-tie between divisions A and B of the PCCW system in the event of a loss of cooling water and, if not, provide an assessment of a SAMA to develop and implement a procedure to perform the cross-tie.

Page 88 of 116

United States Nuclear RegulatOly Commission SBK-L-11001/Enclosure NextEra Energy Seabrook Response to SAMA RAI 5h Seabrook Plant Operations Procedures provide explicit instructions for alignment ofthe PCCW A and B cross-tie. The cross-tie alignment is primarily used during maintenance activities; however it could be used during an extreme off-normal event involving a failure of heat sink in one division with failure of frontline components in the opposite division, provided that adequate time is available. The cross-tie is not cUlTently modeled in the PRA and is judged to nQt provide a significant benefit if it were to be considered in the PRA.

SAMARAI5i

5)

Provide the following with regard to the SAMA identification and screening process:

i.

SAMA 127, "Revise emergency operating procedures to direct isolation of a faulted steam generator," is screened in Table F.6-1 using Criterion B. However, the explanation provided in the Phase I Disposition colunm, "Faulted SG refers to Steam line break and Ruptured SG refers to SG rupture," does not explain why this is not a viable SAMA candidate. Clarify whether the existing emergency operating procedures (EOPs) implement this SAMA and if the EOPs distingnish between a faulted steam generator and a ruptnred steam generator.

NextEra Energy Seabrook Response to SAMA RAI 5i The context of SAMA # 127 is to have specific emergency operating procedures for isolation of a steam generator for the purpose of reducing the consequences of a steam generator tube rupture. Seabrook Emergency Operating Procedures currently direct specific operator actions to diagnose a steam generator tube rupture and to perfOlID its isolation. The Reactor Trip or Safety Injection EOP requires operators to check ifSG U-tubes are intact. SG tube integrity is detelTllined based on monitoring of main steamline radiation, condenser air evacuation radiation, steam generator blowdown radiation, and SG nmrow range level-no uncontrolled increase. This procedure transfers to the Steam Generator Tube Rupture EOP, which provides specific steps, beginning at Step 3, for isolation of a "ruptured" steam generator. It is also noted plant EOPs specifically provide actions for the identification and isolation of a "faulted" stemn generator (e.g., steam line break). These actions are delineated in the Reactor Trip or Safety Injection and Faulted Steam Generator Isolation procedures.

Based on the above, SAMA #127 was screened in Phase 1 - intent met.

Page 89 of 116

United States Nuclear Regulatory Commission SBK-L-llOOI/Enclosure SAMARAISj

5)

Provide the following with tegard to the SAMA identification and screening process:

j. SAMA 82, "Stage backup fans in switchgear rooms," and SAMA 84, "Switch for emergency feedwater room fan power supply to station batteries," are screened in Table F.6-1 as not applicable to the Seabrook Station. However, the explanation in the Phase I Disposition column does not appear to preclude the viability of these SAMAs. Provide fmiher justification for screening out these SAMAsor provide an evaluation of each.

NextEra Energy Seabrook Response to SAM;A RAI Sj SAMA #82 - The context of SAMA #82 is to enhance the availability/reliability of ventilation to the essential switchgear rooms in the event of a loss of switchgear room ventilation. This SAMA was initially judged as "not applicable" to Seabrook Station in Phase I screening. A more accurate Phase I screening criterion for SAMA #82 is "intent met". Switchgear room compensatory ventilation procedures exist for maintaining acceptable room temperatures when ventilation components become unavailable.

Switchgear room high temperature alarm response procedures reconmlend several actions to restore ventilation including monitoring of room temperatures, opening of doors, and the set up portable fans to supply external cooling.

SAMA #84 - The context of SAMA #84 is to enhance the availability/reliability of ventilation to the EFW pump house in the event of a loss of pump house ventilation by switching the pump house ventilation fane s) power supply to station batteries. This SAMA was initially judged as "not applicable" to Seabrook Station in Phase I screening. Further evaluation of this SAMA is provided below.

EFW pump house ventilation is needed only for long term sequences involving EFW SG inventory control due to the relatively slow heat-up rate of the pump house. EFW pump house compensatory ventilation procedures exist for maintaining acceptable room temperatures when ventilation components become unavailable. EFW pump house high temperature alarm response procedures recommend several actions to restore ventilation including the monitoring of pump house temperatm*e and to provide pOliable ventilation as required. These compensatOlY actions are credited in the PRA and as a result, failure of the.

ventilation system, which is reliable, is not a significant contributor to core damage frequency.

The EFW pump house ventilation system consists of two trains of ventilation fans and associated damperss. The ventilation fan trains use AC power for fan motive power, not DC power. To operate the ventilation fans using DC motive power would require a redesign of the system with the possibility of requiring a new, larger capacity station battelY. Conceptually, a conservatively low cost of these modifications is estimated at >$250K. A Phase II PRA case study was perfOlmed to assess the potential cost-benefit of possible upgrades to the EFW ventilation system.

The PRA case (OEFWVS) conservatively assumes guaranteed success of EFW ventilation. The following results were obtained from PRA case OEFWVS:

Nominal Cost-Benefit: <$IK Page 90 of 116

United States Nuclear Regulatory Commission SBK*L*I 1001 IEnclosure Upper Bound Cost-Benefit: <$IK Reduction in OECR: <1 %

SAMA Cost: Hardware or procedural modifications aimed at improving EFW pump house ventilation reliability, particularly redesigning the system for DC motive power, is >$1 OaK.

Based on this case study, the proposed SAMA is judged not *cost -beneficial.

SAMARAJSk

5)

Provide the following with regard to tbe SAMA identification and screening process:

k. The SAMA identification process (ER Section F.5) did not appear to include a review of the cost-beneficial SAMAs identified for other Westinghouse 4-100p plants for which license renewal applications have been submitted. Provide an itemized review of the cost-beneficial SAMAs identified in the following recent license renewal applications for Westinghouse 4-100p plants:

Salem, Diablo Canyon, Vogtle, Indian Point 2/3, and Wolf Creek. In the response, provide a Phase I screening of each and, if not screened, provide a Phase II evaluation.

NextEra Energy Seabrook Response to SAMA RAJ Sk Evaluation of candidate SAMAs found to be potentially cost-beneficial at Salem, Diablo Canyon, Vogtle, Indian Point 2/3, and WolfCreek plants is provided below.

~

" i '

SAMA Salem Station SAMA#I Enhance Procedures and Provide Additional Equipment to Respond to Loss of Control Area Ventilation.

Screened Phase I, Criterion B - Intent Met. Seabrook currently has a specific abnormal procedure for implementing compensatory actions for loss of control room ventilation or air conditioning. A compensatory procedure for loss of Essential Switchgear Room ventilation also exists (as addressed in Seabrook SAMA #82).

Page 91 of 116

United States Nuclear Regulatory Conunission SBK-L-llOOI/Enclosure

........,<.*** *.!'!extIil:~.E!1,~rgy:S~ltbm,~}~ ];~~lu~tiiJ?~ipth;;*~L~!1t~1@~n~ L~,.n..,..'.......

,-.......... '.' '?:\\'!. 'y

. '. ". S~~ilr~gl{ 'E~aiJ~;iori ') ~7 S~

.. ' SAMf Description

~ '"....

SAMA#2 Re-configure Salem 3 to Provide a Screened Phase 2, Not Cost-BeneficiaL More Expedient Backup AC Power Seabrook Station is a single unit site with no other onsite Source for Salem 1 and 2.

power generation capability (except for onsite emergency and supplemental diesel generators). Illstallatioll of additional onsite power generation capability was evalnated in Seabrook SAMA #13 (buried offsite power source),

SAMA #14 (gas turbine), and SAMA #156 (install alternate ofTsite power source). These SAMA candidates were judged not cost-beneficial.in Phase 2.

SAMA#3 Install Limited EDG Cross-tie Screened Phase 1, Criterion A - Not Applicable. Seabrook Capability Between Salem 1 and 2, Station is a single unit site with no other onsite power generation capability (except for onsite emergency and supplemental diesel generators). Pllase 1 screening is consistent with Seabrook SAMA# 12, Create AC Cross-tie Capability with Other Unit (multi-unit site).

SAMA#4 Install Fuel Oil Transfer Pump on Screened Phase 1, Criterion B - Intent Met.

"C" EDG & Provide Procedural This SAMA is similar to Seabrook SAMA #11 and SAMA Guidance for Using "C" EDG to

1. 17. Seabrook currently has 2 EDGs that are dedicated to Power Selected "A>> and "BH Loads.

their respective emergency bus. In addition, Seabrook has a third diesel generator unit - supplemental emergency power supply (SEPS), which can be aligned to either emergency power division. In addition, Seabrook has the capability to cross-tie the divisional BDG fuel oil tanks.

SAMA#5 IrlstaU Portable Diesel Generators to FDP Replacement - Screened Phase 2, Not Cost-Beneficial Charge Station Battery and Replacement of PDP and installing a diesel generator for Circulating Water Batteries &

motive power is similar to Seabrook SAMA #26 (install Replace PDP with Air-Cooled Pump.

additional high pressure injection pwnp with independent diesel). This SAMA IVas shown to be not cost-beneficial.

Fortable Battery Charger - Not screened - Potentially Cost-Beneficial - however, similar to Seabrook SAMA # 157 Installing portable diesel generator for charging batteries is similar to Seabrook SAMA #157 (provide independent AC power source for battery chargers). This SAMA was shown to be potential cost beneficial to extend main station battery life during long term SBO sequences.

Regarding switchyard DC cono'o! power, Seabrook has an 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> switchyard battery system, independent orthe main station baUeries. lf switchyard batteries become depleted, Seabrook switchyard breakers can be operated locally/manually without DC control power. This locaVmanual action is accounted for in the PRA by assWlling an additional hour is 'needed to recover off site power dming long tenn SBO sequences. Removal of the additional hom for local action is judged to have a negligible positive impact on SBO core damage risk.

Page 92 of 116

• United States Nuclear Regulatory Conunission SBK-L-llOOI/Enclosure

.-:,ii,-!,,"

. ~e~\\~I*~ ~ne7gf§el!bl:o~kEvaIUatiOnOf Otbe"PIi~tSAMAs:C*... ~.

'. SA1~"

SANIAnescJi>!i()~

< I: *

.J'eab~6()kEvalu~ti()n... iCC. -

SAMA #5A Install Portable Diesel Generators to Not screened - Potentially Cost-Beneficial - however, SAMA#6 SAMA#7 SAMAY8 Charge Station Battery and similar to Seabrook SAMA #157 Circulating Water Batteries.

Enhance Flood Detection for 84' Aux Building and Enhance Procedural Guidance for Responding to Service Water Flooding.

Install "B" Train AFWST Makeup Including AlIernate Water Source.

Install High Pressure Pump Powered with Portable Diesel Generator and Long-term Suction Source to Supply the AFW Header.

Installing portable diesel generator for charging balteries is similar to Seabrook SAMA #157 (provide independent AC power source for battery chargers). This SAMA was shown to be potential cost beneficial to extend main station battery life during long teon SBO sequences.

Regarding switchyard DC control power, Seabrook has an 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> switchyard battery system, independent of the main station batteries. If switchyard batteries become depleted, Seabrook switchyard breakers can be operated locally/manually without DC control power. This locallmanual action is accowlted for in the PRA by assuming an additional hour is needed to recover offsite power during long term SBO sequeuces. Removal of the additional hour for local action is judged to have a negligible positive impact on SBO care damage risk.

Screened Phase I, Criterion B - Intent Met.

Flood alarms, sump alaons, response procedures and associated opemtor actions were reviewed as part of updated internal nooding assessment. Current alarnl response procedures judged adequate and are reflected in assessment.

Screened Phase 2, Not Cost-Beneficial This SAMA involvingIong teon makeup to the CST is similar to Seabrook SAMA #162 (increase capacity ofthe CST) and SAMA #164 (modity JO" condensate filter flange to have 2-II2-inch female fire hose adapter with isolation valve). These SAMAs were shown 10 be not cost-beneficial. In addition, it is noted that the time to empty the CST is approximately 17 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br />. Seabrook currently has several means of making-up water inventory to the CST.

These include alignment ofthe Demineralized Waler (DW) system pumps or alignment of the gravity drain ofthe Demineralized Water Storage Tanks to the CST, makeup from the condenser hot well spill or by use of the cooling tower portable pump. Therefore, the intent of the Salem SAMA is judged to be met regardless of cost-benefit.

Screened Phase I, Criterion B - Intent Met.

This SAMA is similar to Seabrook SAMA #66 and SAMA

1. 75. Seabrook currently has the capability to connect several water sources to supply the EFW suction header.

These include: the fire system using a diesel-driven fire pump; the portable diesel-drivcn pwnp with suction connected to the fire main; and the cooling tower portable makeup pump using either the cooling tower basin or Browns illver. Iii addition, the water contents in the DWST can be gravity drained to the CST.

Page 93 of 116

United States Nuclear Regulatory Commission SBK-L-llOOI/Enclosure SAMA#9 SAMA#IO SAMA#l1 SAMA#12 SAMA#14 SAMA#17

'. ~ ", NextE,:il EI1(;r~ se~~\\.o~~Eval.;'~ti~d of()tb~i.~~~n~ ~M1As '.*** '.:,.. " *.,.

SAMA Desc'*iptloi'-..;/.... "....

. ;'sJ~i;i,~cil{Eyalu~ti,o~ '. **.

Connect Hope Creek Cooling Tower Basin to Salem Service Water System as Alternate Service Water Supply.

Provide Procedural Guidance for Faster Cooldowll on Loss ofRCP Seal Cooling.

Modify Plant Procedures to Make Use of Other Unit's PDP for RCP Seal Cooling.

Improve Flood Barriers Outside of 220/440V AC Switchgear Rooms.

Expand AMSAC Function to Include Backup Breaker Trip on RPS Failure.

Enhance Procedures and Provide A<;lditional Equipment to Respond to Loss ofEDG Control Room Ventilation.

Screened Phase I, Criterioll B - Intent Met.

Seabrook is a single unit site. Seabrook is currently equipped with a four-train (two-division) Oce~n Service Water System. The alternate service waier supply is provided by the service water cooling tower, which consists of two pump divisions and a cooling tower water basin with a capacity of greater than 3M gallons. The cooling tower service water supply is independent of the ocean supply.

Screened Phase I, Criterion B - Intent Met.

RCS cooldown and depressurization procedures are addressed in Seabrook SAMA #50. Seabrook's emergency procedures include actions to cooldown and depressurize the RCS"t a rate of30 to 50 FIhr while maintaining adequate subcooling margin and to establish conditions for operation ofRHR shutdown cooling. These actions are currently modeled in the PRA and any changes to further enhance the procedures are judged to have negligible benefit on the accident sequences.

Screened Phase I, Criterion B - Intent Met.

Seabrook's charging system design includes two centrifugal charging pumps A and B and a single positive displacement pump (PDP). The PDP is powered Ii'om a non-safety electrical power source and requires component cooling from PCCW train B. The PDP is credited for seal irUection in the loss of RCP seal injection initiating event in the PRA.

Replacement of this pump with a centrifugal pump was evaluated as Seabrook SAMA # 170.

This SAMA is applicable to Seabrook SAMA #192, installation of flow limiting device in Control Building fIre protection piping. Refer to Seabrook response to RAI I a.

Screened Phase 2, Not Cost-Beneficial.

Seabrook SAMA # 174 is judged similarto this Salem Station SAMA for ATWS mitigation. SAMA #174 evaluated the cost-benefit of providing an alternate scram button to remove power from the MG-sets to the control rod drives. This change was judged not cost -beneficial ill the Seabrook }>hase 2 assessment.

Screened Phase I, Criterion A - Not Applicable.

Seabrook Station does not have EDG control rooms or EDG control room ventilation systems.

Page 94 of 116

United States Nuclear Regulato!), Commission SBK-L-1l00 1/ Enclosure

. ',..' Ne~~ra Ene~~ Seabl'ook Evaluatio? of Ot~e~; Plan~~~s

'-'.~,'

c_

....~

', ',.'0/

~ "

I

' S~

'.' : *.S~

Descl:iption Seabl'ook Evaluation.'

SAMA#24 Provide Procedural Guidance to Screened Phase 1, Criterion B - Intent *Met.

Cross-tie Salem 1 and 2 Service Seabrook is a single lInit site. Seabrook is currently Water Systems.

equipped with a four-train (two-division) Ocean Service Water System. The alternate service water supply is provided by the service w.ter cooling tower, which consists of two pump divisions and a cooling tower water basin with a capacity of greater than 3M gallons. The cooling tower service water supply is independent of the ocean supply.

SAMAII27 In Addition to the Equipment Screened Phase 1, Criterion B - Intent Met.

Installed for SAMA 5, Install This SAMA, involving seismic connections to alternate Permanently Piped Seismically Qualified Connections to Alternate AFW water sources, is similar to the cuneot Seabrook AFW Water Sources.

capability. At Seabrook, the CST is seismically qualified and is protected from external events by a rugged, seismic structure. A pennanent, seismically qualified flanged connection exists to allow connection of an alternate water supply to the EFW turbine driven pump suction andlor refill of the CST.

Diablo Canyon SAMA#12 Improve Fire Barriers for ASW and Screened Phase I, Criterion B - Intent Met.

CCW Equipment in the Cable Additional barriers for fire areas were screened out based Spreading Room.

on the Seabrook Station plant design with 3-hour rated fife barriers. Also, a review of the ftre risk by location in the most recent update supports this screening that additional fife barriers would not signiftcantly impacffire risk (refer to RAI5.£)

SAMA#13 Improve Cable Wrap for the PORVs Screened Phase I, Criterion B - Intent Met.

in the Cable Spreading Room.

The Cable Spreading Room (CSR) at Seabrook was screened from detailed evaluation based on the absence of ignition sources and the fact that only small amounts of transient combustibles would likely be found in the area.

Also, the area is equipped with fIre detection and suppression. Cables for the two safety trains enter the room at different locations and are separated. Cables Iravelli'om the separate trays to the conlrol room from through metal wire ways to the penetrations in the ceiling.

Page 95 of 116

United States Nuclear Regulatory Commission SBK-L-11001 / Enclosure o

0 Co'_'

SAMA,

SAMA #24 Prevent Clearing ofRCS Cold Leg Water Seals.

SAMA#25 VogUe SAMAII2 Fill or Maintain Filled The Steam Generators to Scrub Fission Products.

Maintain full-time black start capability oftbe Plant Wilson combustion tllrbines.

Screened Phase 2, Not Cost-Beneficial.

The context of this SAMA is that clearing of the water seals in the cold legs after core damage could result in a greater challenge (0 a SG thennally induced tube rupture (SGTI) due to the resulting unobstructed flow path of hot gas between the reactor vessel and SG( s). Thus, a procedilre change to ensure that tile cold legs are not cleared would reduce the SGTI challenge and reduce the associated release. Tq assess this SAMA for Seabrook, PRA case XSGTIS Was run, which eliminated 100% of the potential for SOT!. The results of this case show that SGTI has minimal contribution to release. The reason SGT! does not contribute significantly to release is because successfully maintaining SG water inventory or depressurizing the RCS also effectively mitigate the SGT! tube rupture challenge. It is noted that the [u'st two steps in Seabrook's Severe Accident Management Guideline (SAMG) diagnostic flow,

chart is to ensure SG water level greater than 10% nalTOW range and RCS pressure <285 psi. The associated nominal aDd upper bound cost-benefits of this SAMA are both less than $1K. This SAMA is judged not cost-beneficial based on the estimated cost for a procedure modification, which is ill the range of$I510 $20K.

Screened Phase I, Criterion B - Intent Met.

The context of this SAMA is for plant procedures to ensure that there is sufficient water level in the SGs (preferably a faulted SG) so that fission product release through a postulated tube rupture will be reduced by scrubbing effects. At Seabrook, procedures specifically address SG water level and isolation of a faulted SG to minimize the possibility of release. In addition, severe accident guidelines, which are entered before the onset of core damage (at 1100 'F), further address the imp0l1ance to inject into the SGs and to maintain SG inventory for the purpose of: (1) protecting the SG tubes from creep lUpture,

. (2) sClUbbing fission products that could enter the SG from tube leakage, and (3) making the SGs available as a heat sink for the RCS.

Screened Phase 2, Not Cost-Beneficial.

Seabrook Station is a single unit site with no other onsite power generation capability (except for o!lsite emergency and supplemental diesel generators). Installation of additional onsile power generation capability was evaluated in Seabrook SAMA #13 (buried offsile power source),

SAMA #14 (gas turbine), and SAMA #156 (install alternate offsite power source). These SAMA candidates were judged not cost-beneficial in Phase 2.

Page 96 of 116

United States Nuclear Regulatory Commission SBK-L-llOOI/Enclosure

..,. 'i' NextEra Energy Seabrook Evalnation of Other Plant SAMAs

" i' i

SAMA SAMA Description SAMA # 4 Prepare procedures and operator training for cross-tying an opposite unitDG.

. SAMA #6 Implementation of a bypass line for the cooling tower return isolation valves.

SAMA #16 Enhance procedures for ISLOCA response.

Screened Phase I, Criterion A - Not Applicable. Seabrook Station is a single unit site with no other onsite power generatioll capability' (except for onsite emergency and supplemental diesel generators). Phase I screening is consistent with Seabrook SAMA# 12, Create AC Cross-tie Capability with Other Unit (multi-unit site).

Screened Phase 2, Not Cost-Beneficial.

The purpose of this Vogtle SAMA is to reduce the likelihood ofa Vogtle-specific SW failure mode that could lead to loss ofEDG cooling following a J.oSP event. The proposed moc)ification is to install redundant, quick opening valves in parallel with existing cooling tower isolation valves, which are signaled to close and fe-open during a LOSP event. Failure to re-open would result in loss of cooling to the respective EDG. Thus, redundant parallel valves would tend to increase the reliability ofEDG cooling. The EDGs at Seabrook are also cooled by Service Water (SW). SW to each EDG is nOimally isolated by a closed AOV designed to fail open on loss of power or instrument air. The AOVs open automatically upon EDG demand and are not cycled closed and opened similar to the Vogtle design. The Seabrook AOVs have a history of reliable operation. As a result, installation of redundant, parallel SW valves would not significantly improve EDG reliability and would not be cost-beneficial based on inspection ofPRA Case NOSBO (which is very conservative) and cost of implementation judged to be on the order of $500K. Detailed (less optimistic) evaluation would further show this option as being not cost-beneficial.

Screened Phase 2, Not Cost-Beneficial.

TillS Vogtle SAMA did not identity any specific procedural changes to improve the risk of ISLOCA events. Vogtle simply calculated a decrease in risk assuming elimination of all ISLOCA risk. Seabrook PRA Case LOCA06 also quantified the risk benefit associated with elimination of ISLOcA risk. Based on this optimistic assessment, the UB benefit was determined to be on lhe order of$53K. Based on a review of procedures, any changes to procedures are judged not to have a significant impact on ISLOCA risk.

The cost of procedure changes and training are judged to be in the range of$30 to $40K. Detailed (less optimistic) evaluation would further show this option as being not cost-beneficial.

Page 97 of 116

United States Nuclear Regulatory Commission SBK-L-llOO I / Enclosure

.i:

SAMA Indian Point SAMA #30 Provide a portable diesel-driven battery charger.

SAMA#52 SAMA#55 SAMA#61 SAMA#62 Open city water supply valve for altemati ve AFW pump suction.

Provide hard-wired connection to one SI or RHR pump I1'OIn the Appendix R bus (MCC 312A).

Upgrade the Alternate Safe Shutdown System (ASSS) to allow timely restoration of seal injection and cooling.

Install flood alann in the 480VAC switchgear room.

Not screened - Potentially Cost-Beneficial-however, sinlilar to Seabrook SAMA # I 57 Installing portable diesel generator for charging batteries is sinlilar to Seabrook SAMA #157 (provide independent AC power source for battery chargers). This SAMA was shown to be potential cost beneficial to extend main station battery life during long tenn SBO sequences.

Screened Phase I, Criterion B -, ntent Met.

This SAMA is similar to Seabrook SAMA #66 and SAMA

1. 75. Seabrook currently has the capability to connect several altemate water sources to supply the EFW suction header. These include: the fire system using a diesel-driven flfe pump; the pOilable diesel-driven pump with suction connected to the flfe main; and the cooling tower portable makeup pump using either the cooling tower basin or Browns River. In addition, the water contents in the DWST can be gravity drained to the CST.

Screened Phase I, Criterion B - Intent Met.

Seabrook currently has two, separate Remote Safe Shutdown (RSS) panels, one panel for each division. The panels are located in the respective divisional emergency switchgear room. Each RSS division has the capability to use the associated EOG and emergency buses to controVstabilize RCS inventory using boric acid transfer pump and/or high head charging pump; perform plant cooldown using ASDVs and EFW, and initiate shutdown cooling using an RHR pump.

Screened Phase I, Criterion B - Intent Met.

Seabrook's Remote Safe Shutdown (RSS) panels currently have the capability to provide both seal injection and cooling. Seal injection is provided using a high head charging pump. Seal cooling is provided nsing a thennal barrier cooling (TBC) pump with PCCW and SW pumps for heat sink.

This SAMA is applicable to Seabrook SAMA # 192, installation of flow limiting device in Conb'ol Building flfe protection piping. Refer to Seabrook response to RAIla.

Page 98 of 116

United States Nuclear Regulatory Commission SBK*L*lIO01/Enclosure

-c o' :"_

NextEra En~rgySeabro~k Eval,u~,tionofp.thel' Plant S~s

'c I.'

, **.. SAMA WolfCreek SAMA#I SAMA#2 SAMA#3 Pennanent, Dedicated Generator for the NCP with Local Operation ofTD AFW After 125V Battery Depletion, ModifY the Controls and Operating Procedures for Sharpe Station to Allow for Rapid Response, AC Cross*tie Capability_

Screened Phase I, Criterion B - Intent Met.

The context of this WolfCreek SAMA is to reduce SBO RCP seal LOCA scenarios by installing an additional, dedicated diesel generator power source for the normal charging pump (NCP), Seabrook currently has 2 EDGs that are dedicated to their respective emergency bus, In addition, Seabrook has a third diesel generator"unit

• supplemental emergency power supply (SEPS), which can be aligned from the control room to either emergency power division, Once the breaker for SEPS is manually closed onto the aligned bus, the load sequencer connects the required loads, Because SEPS can be quickly aligned from the control room to re-energize an emergency bus,.

restoration of the thennal barrier cooling and seal injection functions is accomplished in time to preserve RCP seal integrity and avoid seal failure, Screened Phase 2, Not Cost*BeneficiaL Seabrook Station is a sUlgle unit site with no other onsite power generation capability (except for onsite emergency and supplemental diesel generators), lnstallation of additional onsite power generation capability lVas evaluated in Seabrook SAMA #13 (buried offsite power source),

SAMA #14 (gas turbine), and SAMA #156 (install alternate offsite power source), These SAMA candidates were judged not cost*beneficial in Phase 2, Screened Phase I, Criterion B - Intent Met.

The context of this WolfCreek SAMA is to provide enhanced capability to cross*tie the emergency 4kV buses.

This SAMA is similar to Seabrook SAMA # I L Seabrook currently has 2 EDGs that are dedicated to their respective emergency bus, In addition, Seabrook has a third diesel generator unit* supplemental emergency power supply (SEPS), which can be aligned to either emergency power division.

Page 99 of 116

United States Nuclear Regula\\OlY Commission SBK-L-l 1001 / Enclosure SAMA #4 Upgrade emergency procedures to direct local. manual closure of the RHR EJRVS809A and EJHV8809B valves if they fail to close remotely.

SAMA #5 Enhance procedures to direct operators to open EDG Room doors for altemate room cooling.

Screened Phase 2, Not Cost Beneficial.

The context ofthis WolfCreek SAMA is to provide a procedural enhancement for operators to locally, manual close RHR valves to isolate an ISLOC should the valves fail to close remotely. At Seabrook, ISLOCA events are postulated to occur in the RHR system due to: (I) multiple MOV failures in the RHR suction supply from the Res hot legs, and (2) multiple check valve failures in the RHR injection lines to the RCS coldlhot legs. Remote manual isolation of an ISLOCA event in RHR is detailed in plant EOPs. From a PRA perspective, isolation of a postulated ISLOCA originating from the RHR suction valve failures is assumed /lot possible and thus is not credited in the PRA.

Isolation of a postulated ISLOCA originating from RHR Dljection valve failures is possible via manual closure of valve RH-MOV-,l4 or RH-MOV-26. However, only limited isolation credit is given in the PRA for events when RHR reliervalves are successful and RHR piping and heat exchanger pressure boundaries remain intact. Thus, ISLOCA leakage is a result of relief valve opening and RHR pump seal degradation. Failure of the operator action to isolate RHR valve 14 (26) has an RR W of only 1.0001.

Given this extremely low risk reduction, the cost-benefit of making improvements to associated proc.edures or valve hardware will be well below the range of $28K to $53K, which assumed elimination of all ISLOCA risk in PRA Case LOCA06. Therefore, this SAMA is judged not cost-beneficial.

Screened Phase I, Criterion B - Intent Met.

The context of this WolfCreek SAMA is to provide for altemate EDG room cooling during favorable outdoor ambient temperature conditions by OpenDlg room doors. At Seabrook, the diesel air handling (DAR) systems consists of a supply and exhaust fan and associated dampers for velltilatinglcoolulg the respective diesel room. TIle DAH is included in the PRA model due to the relatively sholt heatup times. Note, however, the:tl during a substantial time of the year, the outside temperature is low enough that ventilation may not be required. DAR compensatory ventilation actions exist for maintaining acceptable DG room temperatures depending on outdoor ambient temperatures when DAR components arc unavailable. In addition, DG room high temperature alarm response procedures recommend several actions including monitoring ofDG temperatures and providing poltable fans to cool the affected room.

Page 100 of 116

United States Nuclear Regulatory Commission SBK-L-1l001/Enclosure

.c:.

~. _..'

NextEra Ene~'gySeabrork Evaluationvf ()tbe~PJant SAMAs

_" *i.*",,,

i***

. ~ :.

SAMA Description '.:;*.Seabl*ook}i;valuation.'.

SAMA #13 Altemative Fuel Oil Tank with Gravity Feed Capability.

SAMA #14 Permanent, Dedicated Generator for the NCP, one Motor Driven AFW Pump, and a Battery Charger.

SAMARAISI Screened Phase I, Criterion B - Intent Met.

Seabrook currently has 2 EDGs that are dedicated to their respective emergency bus. Tile capability to cross-tie the divisional EDG fuel oil tanks exists. Fuel oil makeup to the respective EDG day tank is accomplished via the divisional fuel oil transfer pump. Gravity drain from the filel oil storage tanks to the associated day tank is not practical.

However, the discharge /i'om each fuel oil transfer pump can be aligned to supply the opposite division EDG day tank. This provides flexibility to maintain day tank inventory should a problem develop with one transfer pump.

In addition, as noted above, Seabrook has a third diesel generator unit - supplemental emergency power supply (SEPS), which can be aligned to either emergency power division. The SEPS has a dedicated fuel oil, independent of the EDG I\\lel oil.

Screened Phase I, Criterion B - Intent Met The context ofthis WolfCreek SAMA is similar to Wolk Creek SAMA # I but with increased diesel generator capacity to also operate a motor-driven AFW pump. As noted above, Seabrook has already ulStalled a third diesel generator unit - supplemental emergency power supply (SEPS), which can be aligned from the control room to either emergency power division. Once the breaker for SEPS is manually closed onto the aligned bus, the load sequencer connects the required loads. Because SEPS can be quickly aligned from the control room to re-energize an emergency bus, restoration of the thermal barrier cooling and seal injection functions is accomplished in time to preserve Rep seal integrity and avoid seal failure. The SEPS diesel generator load is maintained at or below 5280kW. This is sufficient capacity to operate many important systems including: motor-driven EfW pump or the startup feedwater pump, SW pump, PCCW pump, battery charger, etc.

5)

Provide the following with regard to the SAMA identification and screening process:

L SAMAs were identified for.all basic events having a RRW greater than or equal to Loos.

Provide the maximum dollar benefit of a SAMA that eliminates 100% of the risk of a basic event having an RRW value of 1.005.

Page 101 of1l6

United States Nuclear Regulatory Commission

. SBK-L-llOOI / Enclosure NextEra Energy Seabrook Response to SAMA RAI 51 The maximum dollar benefit of a SAMA that eliminates 100% of the risk of a basic event having an RRW value of 1.00S is $2.SK (nominal benefit at 7% discount rate) and $4.8K (upper bound benefit).

This benefit result is based on setting human action basic event HH.RDGL2Q.FL from its point estimate valve of 1.8E-Ol to guaranteed success, a value of 0.0. Human action basic event HH.RDGL2Q.FL models operator failure to manually load needed pumps onto the EDG or SEPS given a seismic event with failure of the emergency power sequencer. This human action basic event has an RR W of 1.00S7 as provided in Table F.3.1.1.l-2 of the SAMA report.

SAMARAISru S)

Provide the following with regard to the SAMA identification and screening process:

m. Table F.S.6-l identifies the source of38 SAMAs as being plant-specific SAMAs based on review of the IPE, IPEEE, plant personnel, and expelt panel. Identify the specific source for each of these SAMAs.

NextEra Energy Seabrook Response to SAMA RAJ Sm The primary source for each plant-specific SAMAs is provided in the table below.

= 7:-

-c

".2" SAMA Source

• • • • . *.*..t

. SAl\\:lA,JDs lPE

1. 155, #156, #157, #158, #159, #167, #168, #169, #174, #184, #185, #186, #187 IPEEE

1. 160,#171,#173,#175,#176,#177,#178,#179,#180,#181,#182,#183 Plant Personnel

1. 154,#162,#164,#165,#166,#188,#189,#190,#191 Expert Panel

1. 161,#163,#172,#170 SAMARAISn S)

Provide the following with regard to the SAMA identification and screening process:

n. SAMAs 105 and 191 were screened as not applicable to Seabrook Station because they would violate the current licensing basis. This is not a valid basis for screening the SAMAs as not applicable. Provide nllther justification for why these SAMAs should not be considered in the Phase II evaluation.

Page 102 of 116

United States Nuclear Regulatory Commission SBK-L-JlOOJ IEnclosure NextEra Energy Seabrook Response to SAMA RAI5n SAMA # 1 05 - Delay containment spray actuation after a large LOCA. The context of this SAMA is to extend Reactor Water Storage Tank (RWST) availability during a large LOCA event. This is interpreted as extending the RWST inventOlY during a large break LOCA so as to deplete the inventOlY at a reduced rate. This would provide operators with more time to complete the semi-automatic transfer to cold leg recirculation for large breaks. A Phase II evaluation was performed using PRA Case OLPRS. This case conservatively assumed guaranteed success of the operator action to complete/ensure the RHRlLHSI transfer to long term recirculation during large LOCA events. The results of this case study show that the operator action does not contribute significantly to core damage frequency. Thc following l'CSUItS were obtained from PRA case OLPRS:

Nominal Cost-Benefit:

~$7.2K Upper Bound Cost-Benefit: -$13.7K

% Reduction in OECR: <1 %

Cost of SAMA: The engineering analysis costs to justify a delay in containment spray for large break LOCA events and corresponding changes to emergency and operating procedures is expected to exceed

$100K.

Based on this case study, engineering analysis and modifications to justify a delay in containment spray for large LOCA events and increase operator action time is judged not cost-beneficial.

SAMA #191 - Remove the 135°F temperature trip of the PCCW pumps. The context of this plant-specific SAMA is to improve PCCW reliability by eliminating the potential for a spurious trip due to inadvelient actuation of the temperature switches. Each PCCW division is protected from high temperature with a 2-out-of-2 temperature element logic. It is noted that significant engineering re-analysis would be needed to ensure that elimination of this protective trip would not violate the PCCW system design basis. Spurious operation of this logic could fail the associated division of PCC by tripping off the pumps. A Phase II evaluation was perfOlmed using PRA Case PCTES. This case assumed elimination of 1 00% of the inadvertent failure of the redundant temperature element/logic as a failure mode of the associated PCC division for both loss ofPCCW (AlB) initiating events (during the year) and loss ofPCCW (AlB) mitigative function (mission time). The results of this case study show that inadvertent actuation ofthe temperature/element logic does not contribute significantly to the system

. unreliability. The following results were obtained from PRA case PCTES:

Nominal Cost-Benefit:

<$lK Upper Bound Cost-Benefit: <$lK

% Reduction in OECR: <1%

Cost of SAMA: The engineering analysis costs to justify a change in the design basis temperature and impacts to pipe stress and stlUctural support is estimated to exceed $IOOK.

Based on this case study, engineering analysis and modifications to eliminate the high temperature trip and preserve the design basis are judged to exceed the upper bound cost-benefit and thus are judged not cost-beneficial.

Page 103 of 116

United States Nuclear Regulatory Commission SBK-L-llOOl / Enclosure SAMARAISo

5)

Provide the following with regard to the SAMA identification and screening process:

o. SAMAs 173 and 185 both are described as "improve procedural guidance for directing depressurization of ReS" and both are' dispositioned as already implemented. Clarify the difference between these two SAMAs.

NextEra Energy Seabrook Response to SAMA RAJ 50 SAMA #173 and #185 are both related to improved procedure guidance for directing operators to perform RCS depressmization. Table F.6-1 provides the description ofthese SAMAs.

SAMA #173 -The context ofSAMA #173 is to have adequate procedure guidance directing operators to depressurize the ReS before core damage, during SBO-type sequences with success oflhe tmbine-driven EFW pump, to minimize loss of primary inventolY due to RCP sealleakage. This action is modeled in the Levell PRA to prevent/delay core damage. The procedure guidance is explicitly provided in the EOP.

SAMA #185 - The context of SAMA # 185 is to have adequate procedure guidance directing operators to depressurize the RCS after core damage to reduce the potential for a high pressure core melt ejection (HPME) and challenge to contaiwnent due to direct containment heating (DCH). Tllis action is modeled in the Level 2 PRA prior to hot leg creep rupture and prior to RPV failure to prevent an energetic HPME event from occurring. The procedure guidance is explicitly provided in the EOP.

These SAMAs were identified from review of the Seabrook IPE and IPEEE and have been completed.

The operator guidance provided in these procedures is judged adequate. Thus, both SAMAs were screened in Phase 1, Criterion B - intent met.

SAMARAI5p

5)

Provide the following with regard to the SAMA identification and screening process:

p. ER Table F.S.6-1 Footnote A states that "Plant-specific SAMA candidates based on review of IPE, IPEEE, presentation and solicitation of plant personnel and expert panel" were the source for several non-industry or NEI SAMAs. Clar'ify that the RR W listing was used to identifY SAMAs consistent with Tables F.3.1.1.1-2 and F.3.2.l-2.

Page 104 ofl16

United States Nuclear Regulatmy Commission SBK-L-11 00 1/ Enclosure NextEra Energy Seabrook Response to SAMA RAI Sp The SAMA identification process specifically included a review ofthe top RR W basic events provided in Tables F.3.1.1.1-2 and FJ.2.1-2. As provided in Section F.5.1 of the Seabrook SAMA Report, the risk reduction wotth (RR W) of the components in the baseline model was used to identify the basic events that could have a significant potential for reducing risk. Components with RR W > 1.005 were identified as the most impottant components. A similar review was performed on a system basis. The components and systems were reviewed to ensure that each component and system was covered by an existing generic or plant -specific SAMA item based on a "functional" category. For example, basic event FWP3 7 A.FR, "Turbine Driven Pump FW-P-37A Fails to Run" is identified with SAMAs related to Feedwater &

Condensate. A specific SAMA candidate was not identified to address risk reduction of each basic event due to their relatively low RRW importance, corresponding low cost-benefit, and the basic event's relationship to the assigned functional SAMA category. Refer to Seabrook response to RAI 5b.

SAMARAISq

5)

Provide the following with regard to the SAMA identification and screening process:

q. ER Table F.5.6-1 presents SAMA 188 to modify "a containment ILRT lO-inch test flange to include a 5-inch adapter" that Table F.6-1 screen outs with an explanation in the Phase I Disposition colunm that "flange and procedure exists." The applicability of this disposition is unclear. Is there already a 5-inch adapter on the ILRT 10-inch flange to connect fire water?

NextEra Energy SeabrooI{ Response to SAMA RAI 5q The 10-inch flange with fire hose adapter has been pre-fabricated; is stored in a designated, controlled area and is available for attaching to the I O-inch ILRT flange for the purpose of containment flooding via Severe Accident Guideline instructions. The flange/adapter would be used to connect the fire system as one alternate means of performing containmentllooding. The entire containment flooding evolution via the fire water hose connection is expected to take several days. Therefore, there is no significant time savings to be realized by pre-installation of the flange adapter.

SAMARAISr

5)

Provide the following with regard to the SAMA identification and screening process:

r. Section F.6 presents screening criteria used in the Phase 1 analysis. Neither screening criterion D (Excessive Implementation Cost) nor E (Very Low Benefit) is used in Table F.6.1. Phase IT Table F.7.1 seems to use screening criterion D via Footnote 1: "Riskreductionnot specifically evaluated because estimated cost exceeds the possible maximum avelted cost-risk." Clarify that criterion D was used in Phase II and not Phase I and why. Also clarify why criterion E was not used at all?

Page 105 of 116

United States Nuclear Regulatory Commission SBK-L-llOOI (Enclosure NextEra Energy Seabrook Response to SAMA RAI 5r Screening criteria A (not applicable), B (already implemented or intent met), or C (combined) were used as the primary Phase I criteria conceptually to force evaluation of more SAMA candidates under Phase II.

Although criterion "D" or "E" could have been used under Phase I, an attempt was made to judge SAMA costs and benefits as part of Phase II instead of Phase I. In doing so, criterion D and E were not nsed.

SAMARAI5s

5)

Provide the following with regard to the SAMA identification and screening process:

s. The IPE identifies an improvement to install an "alternate, independent emergency feedwater pump (e.g., dieseJ firewater pump hard piped to discharge of startup feed pump." SAMA 29, "provide capability for alternate injection via diesel-driven fire pump," was screened in Phase 1 as "implemented through altemate mitigation strategy." In addition, SAMA 163, "install third EFW pump (steam-driven)," was determined to not be cost-beneficial based on the estimated benefit of

$IOOK in the baseline analysis and $190K in the uncertainty analysis, and a cost of>$250K.

Describe the altemate mitigation strategy that was the basis for screening SAMA 29 and, since SAMA 29 appears to be a lower cost alternative than SAMA 163 and would achieve much ofthe estimated benefit of SAMA 163, clarify why SAMA 29 should not be further considered in the Phase 2 evaluation.

NextEra Energy Seabrook Response to SAMA RAI 5s SAMA #29, "Provide Capability for Alternate Injection via Diesel-driven Fire Pump", was screened in Phase I using Criterion B - mtent Met, implemented through alternate mitigation strategy. The capability to use the fire system (with diesel-driven fire pump) for injection to the steam generators exists and implementation instructions are provided in Seabrook's Severe Accident Management Guidelines. In addition, two portable diesel-driven pumps are also available to perfolm this function using (1) the suction from the fue protection systemlhydrant, (2) the cooling tower basin, or (3) the Browns River. Use of the portable diesel-driven pump is also included in the Severe Accident Management Guidelines.

Therefore, SAMA #29 was screened as intent met in Phase I and further consideration under Phase II is not necessary.

Page 106 of 116

United States Nuclear Regulatory Commission SBK-L-Il 00 1 / Enclosure RAl Section 6 SAMARAI6a

6)

Provide the following with regard to the Phase II cost-benefit evaluations:

a. Provide the % reduction in OEeR for each SAMA evaluated in Table F.7-l and any other SAMAs evaluated in response to RAls.

NextEra Energy Seabrook Response to SAMA RAI 6a The percent reduction in off site economic cost risk (OEeR) for each SAMA evaluated in Table F.7-1 is pIOvided in the table below. The reduction in OEeR is assumed to be the ratio ofthe SAMA's property damage savings to the maximum attainable property damage savings and is based on the base case (best estimate 7% discount rate).

SAMAPRA*-.,

Approx,i1iulte.

,** SAMA#

. Case.",;":,. 'OECR ReduCtion

1. 2, #20, # 154, NOSBO 14%

1. 161,#190

1. 13,1114,1116,1124, NOLOSP 43%

1. 156

1. 21 BREAKER 1%

1. 41 LOCA01 2%

1. 25,#26,#39 LOCA02 49%

1. 28 LOCA03 22%

1/35,11106 LOCA01 13%

1. 147 LOCA05 16%

1. 113, #115, 1/187 LOCA06 1%

1. 43 SWOl 1%

1. 44, #59, CCWOl 19%

1. 55, #167, #168,.

RCPLOCA 10%

1/169,11170,1/172 1/80 HVAC2 1%

1194, 11186 CONTOI 32%

11112,11114 CONT02 48%

1. 96,#108,11109 H2BURN 9%

11119,11121,11125, NOSGTR 14%

11126,11129 11130, #131, #133, NOATWS 14%

11174 11153 NOSLB 0%

11157,1/159 lNDEPAC 2%

11162,11164 CSTOI 0%

11163 TDAFW 7%

11165 NORMW 8%

Page 107 ofl16

United States Nuclear Regulatory Commission SBK-L-llOOl/Enciosure I*

\\ SAMA#

1. 175

1. 179

1. 181

1. 182

1. 184

1. 189 SAMARAI6b SAMAPRA,

App~*Oldril.ate Case OECRRedil'ction.

FIRE2 0%

FiREl 0%

SElSl:v11CO I 17%

SEISl:v11C02 0%

PURGE 0%

lOF2SEPS 2%

6)

Provide the following with regard to the Phase II cost-benefit evaluations:

b. ER Section E.7.2 and Table F.7-1 state thai an expelt panel developed the implementation cost estimates for each of the SAMAs. Describe the level of detail used to develop the cost estimates (i.e., the general cost categories considered). Also, clarity whether the cost estimates accounted for inflation, contingency costs associated with unforeseen implementation obstacles, replacement power during extended outages required to implement the modifications, and maintenance and surveillance costs during plant operation.

NextE.-a Energy Seabrook Response to SAMA RAI 6b The costs associated with implementation of SAMA candidates were determined based on the experience and judgment of plant staff serving as an independent, expelt panel. In most cases, a detailed cost estimate was not performed because of the large margin between the individual SAMA cost-benefits and the judged cost associated with implementation. For procedure modifications, a minimum cost range of

$15K to $40K was used for implementation of a procedure change depending on the type of procedure (AOPfEOP, etc.), extent of procedure change and with no engineering analysis required. For hardware modifications, a minimum cost of $1 OOK was used for development and implementation of a non-complex hardware modification. These costs represent expeli panel judgment for procedure and facility changes that affect risk significant systems or initiating events. No specific allowance is given for inflation, contingencies, implementation obstacles, replacement power, etc.

These lower bound cost estimates are comparable with industry experience as shown in other recent License Renewal fER submittals for Indian Point, Duane Arnold, Pilgrim, Cooper, and Velmont Yankee.

SAMARAI6c

6)

Provide the following with regard to the Phase II cost-benefit evaluations:

c. For celtain Phase II SAMAs listed in Table F. 7 -1, the information provided does not sufficiently describe the associated modifications and what is included in the cost estimate. Provide a more detailed description of both the modification and cost estimate for SAMAs 44, 59, 94, 112, 114, 163,186, and 187.

Page 108 of 116

United States Nuclear Regulatory Commission SBK-L-I 1001 I Enclosure NextEl'a Energy Seabrook Response to SAMA RAI 6c SAMA #44 - Eliminate ECC Pwnp Cooling Water Dependency - The context of this SAMA is to enhance Bces pump reliability by eJimimiting a direct component cooling water dependency. CUlTently, the EeeS pumps require component cooling as follows: Safety Injection and Charging pumps need CCW for lube oil cooling; RHR pumps need CCW for cooling of the mechanical seal during RHR shutdown cooling and long telm sump recirculation. The modifications needed to eliminate the cew direct dependency would involve either a redesign/replacement of the BeeS pumps (6 pumps) or a redesignlreplacement of ecw to incorporate independent cooling. It is noted that room cooling for these pumps also depends on CCW, thus complete elimination ofthe CCW dependency is judged not practical.

The cost estimate of>$500K shown in Table F.8-l was determined ii-om expert panel judgment, based on the SAMA description and consideration of potential costs of design engineering, material procurement, installation, test, operations procedures, etc. for the development and installation of an independent cooling mechanism or new pumps.

SAlVIA #59 - Install Additional CCW Pump - The context of this SAMA is to enhance CCW reliability by installing an additional ecw pump. Given the high reliability of the existing CCW system (each division of CCW has two pumps (four CCW pumps total), an additional pump would need to be completely independent of the existing pumps to maximize the reliability benefit. The cost estimate of

>$500K shown in Table F.8-! was determined from expeli panel judgment, based on the SAMA description and consideration of potential costs of design engineering, material procurement, installation, test, operations procedures, etc. for the development and installation of an additional pump. It is noted that a benchmarked plant estimated the cost of this type of modification to be in the range of$1.5M.

SAMA #94 - Install Containment Filtered Vent - The context of this SAMA is to eliminate containment overpressure failure events by removing decay heat from containment via a filtered vent which would retain fission products. The design concept for this SAMA improvement was provided in the industry SAMA tables as consisting of gravel bed filters or venture scrubber. The cost estimate of>$500K shown in Table F.8-! was detelmined from expeli panel judgment; based on the SAMA description and consideration of potential costs of design engineering, material procurement, installation, test, operations procedures, etc. for the development and installation of a filtering system. It is noted that a benchmarked plant estimated the cost of this type of modification to be in the range of $5M to $6M.

SAMA #112 - Install Containment Isolation Valve Limit Switches - The context of this SAMA is to improve the containment isolation function by increasing contaiwnent isolation valve reliability. The design concept for this SAMA was provided in the industry SAMA tables as consisting of adding redundant and diverse limit switches to each containment isolation valve. At Seabrook, containment isolation valves are alI'eady equipped with limit switches. The limit switch function is primarily for valve position indication/verification and judged not to contribute significantly to the overall reliability ofthe containment isolation valves themselves. Adding yet additional limit switches is judged not practical and would not significantly improve reliability..The cost estimate of>$500K shown in Table F.8-! was determined from expeli panel judgment, based on the SAMA description and consideration of potential costs of design engineering, material procurement, installation, test, operations procedures, etc. for concept development and installation of improved isolation valve limit switches.

Page 109 of 116

United States Nuclear Regulatory Commission SBK-L-11 0011 Enclosure SAMA #114 - Install Self-Actuated Containment Isolation Valves - The context of this SAMA is to improve the containment isolation function by increasing containment isolation valve reliability. The design concept for this SAMA was provided in the industry SAMA tables as consisting of installation of self-actuated containment isolation valves. At Seabrook, isolation of containment penetrations is typically performed using motor operated valves (MOV), air operated valves (AOV) and check valves (CV), and combinations of these valves, depending on the operational function and isolationrequirements of the specific penetration. Check valves are considered to be self-actuated valves. MOVs and AOVs automatically close upon receipt of Engineered Safety Actuation Signals. Based on the. conceptual SAMA description and Seabrook's current design features, a more accurate disposition of this SAMA is to screen it in Phase I, Criterion B - intent met. In addition, given that the conservative, upper-bound cost-benefit ofthis SAMA was limited to -$379K, a detailed estimate of this proposed modification was not performed. The cost estimate of>$500K shown in Table F.8-1 was determined from expert panel judgment, based on the SAMA description and consideration of potential costs of design engineering, material procurement, installation, test, operations procedures, etc. for development and installation of possible improved isolation design of containment penetrations.

SAMA #163 - Install Steam-Driven EFW Pump - The context ofthis SAMA is to improve the reliability of the EFW system by installing a second steam-driven EFW pmnp. Given that the conservative, upper-bound cost-benefit of this SAMA was limited to ~$l90K, a detailed cost estimate of this proposed modification was not performed. The cost estimate of>$Z50K shoWll in Table F.8-1 was determined from expert panel judgment, based on the SAMA description and consideration of possible costs of design engineering, material procurement, installation, test, operations procedures, etc. Based on the Seabrook plant layout, the concept for adding a steam-driven EFW pump, while maintaining the design basis and reliability of other systems, e.g., impact of high-energy piping on existing SSCs, would include construction of a new buiiding/structure to house the pump. The cost of design/engineering, construction, material procurement, procedures, training, testing and maintenance would clearly exceed the cost-benefit of this modification.

SAMA #186 - Install Containment Leakage Monitoring System - The context of this SAMA is to improve the reliability of containment performance by installing improved leakage monitoring systems that would reduce to potential for pre-existing leakage. Given that the conservative, upper-bound cost-benefit of this SAMA was limited to - $31 OK, a detailed estimate of this proposed modification was not performed. The cost estimate of >$500K shown in Table F.8-1 was determined from expert panel judgment, based on the SAMA description and consideration of possible costs of design engineering, material procurement, installation, test, operations procedures, etc.

SAMA #187 - Install RHR Isolation Valve Leakage Monitoring System -The context of this SAMA is to improve the reliability ofRHRisoJation integrity and thus reduce the risk ofISLOCA events. Given that the conservative, upper-bound cost-benefit of this SAMA was limited to -$53K, a detailed estimate of this proposed modification was not performed. The cost estimate of>$1 OOK shown in Table F.8-1 was determined from expert pand judgment, based on the SAMA description and consideration of possible costs of design engineering, material procurement, installation, test, operations procedures, etc. needed to enhance the ability to monitor the integrity of individual RHR valves located both inside and outside of containment.

Page 110 ofl16

United States Nuclear Regulatory Commission SBK-L-ll00l/Enclosure SAMARAI6d

6)

Provide the following with regard to the Phase II cost-benefit evaluations:

d. The benefit and cost evaluation of SAMA 80, "Provide a redundant train or means of ventilation,"

assumes removal ofHV AC dependency for cooling system (CS), safety injection (SI), residual heat removal (RR), and containment building (CB) spray pumps. It is possible that just one of these systems provides most of the benefit. Provide an assesSment of a SAMA to remove HV AC dependency for just the highest risk system.

NextEra Energy Seabrook Response to SAMA RAI 6d The baseline PRA assumes that all of the ECCS pumps require HV AC (via the emergency air handling-EAH) during sequences involving long tenn containment sump recirculation.

The cost-benefit detennination for SAMA #80 is based on PRA case study HVAC2. This PRA case conservatively assumed 100% elimination of the HV AC dependency of all ECCS systems during the long term recirculation sequences. This conservative assumption resulted in a maximum cost-benefit of $32K (nominal at 7% discount rate) and $61K (upper bound). An assessment of this SAMA to remove the HV AC dependency for just the highest risk ECCS system would result in the same or lower cost-benefit.

Installation of a redundant HVAC train to either a single ECCS pump/system or to multiple BCCS pumps/systems is judged to cost >$500K, significantly more than [he upper bound cost-benefit.

SAMARAI6e

6)

Provide the following with regard to the Phase II cost-benefit evaluations:

e. The estimated cost of SAMA 65, "Install a digital feed water upgrade," is $30M while the estimate cost of SAMA 147, "Install digital large break LOCA protection system," is >$500K.

Provide justification for the cost estimates for these two systems. In the response, address the reason for the large cost difference between what appear to be two similar modifications.

NextEra Energy Seabrook Response to SAMA RAI 6e The $30M cost estimate for SAMA #65 is based on a detailed assessment of costs associated with Seabrook's long range plan for digital upgrade of BOP control systems including the feedwater control system. The >$500K cost estimate for SAMA # 147 was determined from expelt panel judgment, based on the SAMA description and consideration of potential costs of design engineering, material procurement, installation, test, operations procedures, etc. for concept development and installation of that system. Given that the conservative, upper bound cost-benefit is ~$196K, a detailed cost estimate of the digital LOCA protection system was not pel'fonned. It is noted that a benchmarked plant estimated the cost of this type of modification to be ~$2M.

Page 111 of 116

United States Nuclear Regulatory Commission SBK -L-J J 00 J / Enclosure SAMARAJ6f

6)

Provide the following with regard to the Phase II cost -benefit evaluations:

f.

The estimated benefits for SAMAs 96, 108, and 109, which assume elimination of all hydrogen ignition/bums, are negative for the reduction in dose-risk (i.e., the dose-risk increases). Describe the reason for this anomalous result.

NextEra Energy Seabrook Response to SAMA RAJ 6f Hydrogen burns do not provide a significant challenge to the containment because of the robust design strength of Seabrook's containment building. Thus, from a risk perspective, hydrogen burns do not contribute to containment failure and there is no change (zero) in containment release when assuming that all hydrogen burns are eliminated. The % reduction in offsite dose in Table F.7-1 was inadvelient1y shown as a -0.05% due to round off and subtraction from the base case result. The conect % reduction in offsite dose for SAMA #96, 108 and 109 is 0.0%.

SAMARAI6g

6)

Provide the following with regard to the Phase II cost-benefit evaluations:

g. The estimate cost for SAMA 113, "Increase leak testing of valves in ISLOCA paths," of $1 OOK seems high for what does not appear to be a hardware modification. Provide justification for the cost estimate.

NextEra Energy Seabrook Response to SAMA RAI 6g The >$1 OOK cost estimate for SAMA # 113 was detennined from expert panel judgment, based on the SAMA description and consideration of potential costs of procedure changes and performance of more frequent inspections. Most of the candidate ISLOCA valves are located inside contaiument. Leak testing of the ISLOCA valves is typically done during plant refueling/cold shutdown conditions, when the valves are assessable and the systems can be aligned/configured to allow installation of test equipment and performance of the testing. Leak testing, as is currently done, but on a more frequent basis, would require costly plant shutdown. It is noted that a benchmarked plant estimated the cost of this lYpe of modification to be -$ 190K.

Page 112 of 116

United States Nuclear Regulatory Commission SBK-L-llOO I / Enclosure.

SAMARAI6h

6)

Provide the following with regard to the Phase II cost-benefit evaluations:

h. It is unclear from the description of SAMAs 157 and 159 (SAMA Case INDEP AC) what changes were made to the PRA model to generate the estimated benefits. Provide a more detailed description of the PRA model changes made to evaluate these SAMAs.

NextEra Energy Seabrook Response to SAMA RAI 6h SAMA #'s 157 and 159 are concerned with improving battelY life of station batteries during SBO conditions when nOlmal and emergency AC power is not available to charge the batteries. Alternate AC power to charge the batteries (#157) or a pre-charged, independent standby battery that could be aligned

(#159) could improve long term SBO sequence by allowing more time to recover offsite power. Two modifications were made to the PRA model to estimate the potential cost-benefit of these SAMAs:

(1) Top Event ODC12, Operator Action to Shed DC Loads to Extend Batteries to 12 Hours: Operator action ODC12 was changed from its nominal value of5.4E-03 to guaranteed success (value of 0.0). This ensures that DC battelY 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> life is guaranteed successful provided that the DC buses, batteries and related equipment are available (have not previously failed randomly).

(2) Top' Event ROSP, Recover Offsite Power: The following offsite power recovery split fractions were set from their nominal failure probability value (shown in parenthesis) to guaranteed success (value of 0.0):

Plant-related LOSP: ROSPPB (1.60E-02), ROSPPC (6.90E-03), ROSPP7 (8.30E-02)

Grid-related LOSP: ROSPGB (1.70E-02), ROSPGC (4.30E-03), ROSPG7 (1.92E-Ol)

Weather-related LOSP: ROSPWB (1.43E-01), ROSPWC (8.50E-02), ROSPW7 (3.63E-01)

These split fractions were chosen because they apply to SBO sequences that include 12 hom battery life and core damage does not occur lUltil approximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> or greater. The 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> sequences are the relevant SBO sequences that would benefit from extending battelY life beyond 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> by providing an independent ac power source (connection of the proposed portable generator) to charge the station batteries, thus allowing continued operation of the steam-driven EFW pump and ASDV s and extending the time available for recovelY of offsite power befOl'e core damage occurs. Other sequences that result in core damage before approximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> would not significantly benefit from the proposed SAMA and the related PRA modeling spiit fraction values are maintained at there baseline value.

Page 113 of 116

United States Nuclear Regulatory Commission SBK-L-11 00 I I Enclosure SAMARAI6i

6)

Provide the following with regard to the Phase II cost-benefit evaluations:

i. The estimated cost of SAMA 157, "Provide independent AC power source for battery chargers,"

of $30K seems low for what is described as a hardware change. Provide justification for the cost estimate.

NextEra Energy Seabrook Response to SAMA RAI 6i The estimated cost of SAMA #157 is based on expeli panel judgment and includes: procurement of a small portable, non-safety related 480V generator, associated connection cables, and operation guideline.

Cost considers that the portable generator would be stored in a convenient location on site (warehouse) and moved into position/coooected if ever needed during an extended SBO event.

SAMARAI6j

6)

Provide the following with regard to the Phase II cost -benefit evaluations:

j. The evaluation of SAMA 179, which assumed eliminating initiator FCRPL, resulted in a Table F.7-1 reduction in CDF of 0.69%, while Table F.3.1.1.1-1 reports the contribution to CDF from initiator FCRPL to be 1.00%. The evaluation of SAMAs ll9, 121, 125, 126, and 129, which assume SGTR events do not occur, resulted in a Table F.7-1 reduction in CDF of 3.47%, while Table FJ.1.l.1-1 repOlis the contribution to CDF from initiator SGTR to be 4.00%. The evaluation of SAMAs 113, ll5, and 187, which assmne ISLOCA events are all eliminated, resulted in a Table F.7-1 reduction in CDF of 2.08%, while Table F.3.l.l.l-1 repOlis the contribution to CDF from initiator LOCI VS to be 2.30%. Clarify the reason for these, and any other, discrepancies and their impact on the SAMA analysis.

NextEra Energy Seabrook Response to SAMA RAI 6j Each of the above differences in initiating event CDF contribution between Table F.3.1.1.1-1 and Table F.7-1 were reviewed. The difference in contribution result is due to rounding. Table FJ.1.1.1-1 was developed from internal PRA documentation and was intended to provide generallmowledge of CDF contributions. Table F. 7 -1 results are judged more precise and these fonn the. basis for SAMA.

Therefore, there is no impact on the SAMA results.

Page 114 of 116

United States Nuclear Regulatory Commission SBK-L-IIOOII Enclosure SAMARA16k

6)

Provide the following with regard to the Phase II cost-benefit evaluations:

k. The ratio of the 95th percentile'CDF to the mean value CDF was reported to be 1.9 in Section F.8.2 of the ER. While this is a "typical" result for internal event CDF, it seems quite low for the fire and seismic CDFs which generally"have wider uncertainty bands than internal events.

Describe how the unceltainty distribution was developed and discuss how and why the CDF distribution is different for internal, fire, and seismic CDF.

NextEra Energy Seabrook Response to SAMA RAI 6k The Seabrook Station PRA model is an integrated model of both internal and external events. As stated in Section F.3.l of the SAMA Report, the Seabrook Station baseline, at power CDF, including both internal and external initiators is 1.44E-5/yr (mean value). The CDF uncertainty for the integrated model is represented by the following, distribution:

5th Percentile = 7.37E-6 50th Percentile = 1.26E-S 95th Percentile = 2.7SE-5 This uncertainty distribution was developed based on a saved-sequence model with 1450 sequences at lE-9 bin cutoff, and Monte-Carlo sample size of 10,000. This distribution includes the integrated contribution of all events, both internal and external. Use of the integrated 9Sth percentile to determine the upper bound ratio between mean and 95th provides a reasonable approximation of upper bound cost-benefit for PRA case results, which are based on the integrated model. This approach is consistent with industry guidance, NEI 05-0 I. Individual distributions for contributions ofinternal, fire and seismic CDF were not developed.

Page 115 of 116

United States Nuclear Regulatory Conuuission SBK-L-llOO 1 / Enclosure RAJ Section 7 SAMARAI7

7)

For certain SAMAs considered in the ER, there may be lower-cost alternatives that could achieve much of the risk reduction at a lower cost.. In this regard, discuss whether any lower-cost altematives to those Phase II SAMAs considered in the ER would be viable and potentially cost-beneficial. Evaluate the following SAMAs (previously found to be potentially cost-beneficial at other plants), or indicate if the particular SAMA has already been considered. If the latter, indicate whether the SAMA has been implemented or has been determined to not be cost-beneficial at Seabrook Station.

a. Use a pottable generator to extend the coping time in loss of AC power events (to power selected instrumentation and DC power to the tnrbine-driven auxiliary feedwater pump). This is an expanded version ofSAMA 74.

b. Provide altemate DC feeds (using a portable generator) to panels supplied only by DC bus.

c. Purchase or manufactnre of a "gagging device" that could be used to close a stuck-open steam generator safety valve for a SGTR event prior to core damage.

NextEra Energy Seabrook Response to SAMA RAI 7 Seabrook has a robust design and many plant safety enhancements have already been implemented at the station. The CDF is relatively low and the conti'ibution to CDF is not significantly dominated by a single initiator. As a result, many SAMAs were screened in Phase I as intent met. Also, individual component contributions to core damage frequency are relatively low, making associated cost-benefits relatively low, even when assuming 100% of the component's contribution is eliminated. Based on inspection of the SAMA list, no new potentially cost-benefit SAMA candidates were identified.

Regarding Items 7a and 7b: Use of a portable generator to extend the SBO coping time by charging station batteries, which supply DC buses, is identified as a potentially cost-beneficial modification. Refer to SAMA #157.

Regarding Item 7c: Use of a gagging device for closing a stuck-open stearn generator safety valve dnring a SGTR event prior to core damage - a Phase II evaluation was performed using PRA Case MSSVRS.

This case assumed guaranteed success of main steam safety valve re-closnre during a SGTR event, provided that operators were successful at controlling EFW flow, Safety Injection (SI) and RCS depressurization. If one of these actions is not successful, MSSV re-closure is assumed failed. The results of this case study show that MSSV failure to reclose does not contribute significantly to plant riskresults.

The following results were obtained from PRA case MSSVRS:.

Nominal Cost-Benefit:

<$IK Upper Bound Cost-Benefit: <$lK

% Reduction in OECR: <1%

Based on this case study, implementation hardware and procednre changes are judged not cost-beneficial.

Pa~e 116 nf 116