Revision as of 05:33, 9 October 2018

Text

Surry Power Station, Units 1 and 2 - Response to Request for Additional Information Regarding Supporting Probabilistic Risk Assessment for License Amendment Request for Emergency Service Water Pump Allowed Outage Extension
	ML17066A187
Person / Time
Site:	Surry
Issue date:	03/01/2017
From:	Sartain M D; Virginia Electric & Power Co (VEPCO)
To:	; Document Control Desk, Office of Nuclear Reactor Regulation
References
	Download: ML17066A187 (79)
	v • d • e

{{#Wiki_filter:VIRGINIA ELECTRIC AND POWER COMPANY RICHMOND, VIRGINIA 2326I U. S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, DC 20555-0001 March 1, 2017 VIRGINIA ELECTRIC AND POWER COMPANY SURRY POWER STATION UNITS 1AND2 PROPOSED LICENSE AMENDMENT REQUEST Serial No.: SPS/LIC-CGL: Docket Nos.: License Nos.: 10CFR50.90 17-016A R1 50-280/281 DPR-32/37 EMERGENCY SERVICE WATER PUMP ALLOWED OUTAGE TIME EXTENSION RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION REGARDING SUPPORTING PROBABILISTIC RISK ASSESSMENT By letter dated July 14, 2016 (Serial No. 16-263), Virginia Electric and Power Company (Dominion) submitted a license amendment request (LAR) to extend the allowed outage time (AOT) in Surry Technical Specification 3.14.B for one inoperable emergency service water (ESW) pump from 7 days to 14 days. In a February 1, 2017 e-mail from Ms. Karen Cotton Gross (NRC Project Manager) to Mr. Gary Miller (Dominion Corporate Licensing), the NRC technical staff requested additional information regarding the probabilistic risk assessment supporting the proposed LAR. Dominion's response to the NRC request for additional information is provided in the attachment to this letter. The information provided in this letter does not affect the conclusions of the significant hazards consideration or the environmental assessment included in the July 14, 2016 LAR. Should you have any questions or require additional Information, please contact Mr. Gary D. Miller at (804) 273-2771. Respectfully, Mark D. Sartain Vice President Nuclear Engineering and Fleet Support COMMONWEAL TH OF VIRGINIA COUNTY OF HENRICO The foregoing document was acknowledged before me, in and for the County and Commonwealth aforesaid, today by Mr. Mark 0. Sartain, who is Vice President -Nuclear Engineering and Fleet Support, of Virginia Electric and Power Company. He has affirmed before me that he is duly authorized to execute and file the foregoing document in behalf of that company, and that the statements in the document are true to the best of his knowledge and belief. Acknowledged before me this -1..!!_ day of , 2017. My Commission Expires: \ 'l. \ i1 I 7..D . NOt ;;plJbliC

'\. Commitments contained in this letter: None Serial No. 17-016A Docket Nos. 50-280/281 RAI Response Page 2 of 2

Attachment:

Response to NRC Request for Additional Information Regarding the Supporting Probabilistic Risk Assessment -Proposed License Amendment Request -Emergency Service Water Pump Allowed Outage Time Extension cc: U.S. Nuclear Regulatory Commission -Region II Marquis One Tower 245 Peachtree Center Avenue, NE Suite 1200 Atlanta, GA 30303-1257 State Health Commissioner Virginia Department of Health James Madison Building -ih floor 109 Governor Street Suite 730 Richmond, VA 23219 Ms. K. R. Cotton Gross NRC Project Manager -Surry U.S. Nuclear Regulatory Commission One White Flint North Mail Stop 08 G-9A 11555 Rockville Pike Rockville, MD 20852-2738 Ms. B. Mozafari NRC Project Manager -North Anna U.S. Nuclear Regulatory Commission One White Flint North Mail Stop 08 H-12 11555 Rockville Pike Rockville, MD 20852-2738 NRC Senior Resident Inspector Surry Power Station Attachment Serial No. 17-016A Docket Nos. 50-280/281 Response to NRC Request for Additional Information Regarding the Supporting Probabilistic Risk Assessment Proposed License Amendment Request Emergency Service Water Pump Allowed Outage Time Extension Virginia Electric and Power Company (Dominion) Surry Station Units 1 and 2 Serial No. 17-016A Docket Nos. 50-280/281 Page 1 of 76 RESPONSE TO NRC REQUEST FOR ADDITIONAL INFORMATION SURRY POWER STATION UNITS 1 AND 2 By letter dated July 14, 2016 (Serial No. 16-263), Virginia Electric and Power Company (Dominion) submitted a license amendment request (LAR) to extend the allowed outage time (AOT) in Surry Technical Specification (TS) 3.14.B for one inoperable emergency service water (ESW) pump from 7 days to 14 days. In a February 1, 2017 e-mail from Ms. Karen Cotton Gross (NRC Project Manager) to Mr. Gary Miller (Dominion Corporate Licensing), the NRC technical staff requested additional information regarding the probabilistic risk assessment supporting the proposed LAR. The request for additional information and Dominion's response are provided below. APLA, RA/ #1: Section 2. 5. 3 of RG 1. 17 4, Revision 2, states that "the impact of using alternative assumptions or models may be addressed by performing appropriate sensitivity studies or by using qualitative arguments, based on an understanding of the contributors to the results and how they are impacted by the change in assumptions or models." Furthermore, the regulatory guidance states that in general, the results of the sensitivity studies should confirm the guidelines are still met even under the alternative assumptions. The license amendment request refers to Sensitivity

1, Sensitivity
2, and Sensitivity
3. a. Discuss and summarize each sensitivity study. Dominion Response:

Sensitivity

1 was conducted to measure how sensitive the application results are to Human Error Probabilities used in the model. It was conducted by multiplying all initiator, initiator, post-initiator, and dependent Human Error Probabilities by a factor of 10. This sensitivity bounds the expected aggregate impact of the outstanding documentation and modeling issues related to HRA in the S007Aa model. Sensitivity
2 was conducted to measure how sensitive the application results are to common cause failure probabilities used in the model. It was conducted by multiplying component common cause failure basic event probabilities by a of 10. This sensitivity bounds the expected aggregate impact of the outstanding documentation and modeling issues related to common cause failure modeling in the S007 Aa model. Sensitivity
3 was conducted to measure how sensitive the application results are to Loss of Offsite Power (LOOP) frequencies used in the model. It was conducted by multiplying all LOOP and initiating event frequencies by a factor of 10. This sensitivity t'.

Serial No. 17-016A Docket Nos. 50-280/281 Page 2 of 76 bounds the expected aggregate impact of all outstanding documentation and modeling issues related to LOOP frequencies in the S007 Aa model. b. Provide the risk results for RG 1.177 and RG 1.174 (i.e., ICCDP, ICLERP, delta CDF, and delta LERF) after simultaneously applying all the sensitivity study changes. If the simultaneous risk results cannot meet the risk acceptance guidelines, then provide justification for the request to extend the allowed outage time. Dominion Response: The sensitivity study changes from Sensitivity #1, #2, and #3 were applied simultaneously. The results are displayed in Table RAl1 b below and meet the acceptance guidelines for RG 1.177 and RG 1.17 4. Table RAI 1 b: Analysis #1 Results -Simultaneous Application of All Sensitivity Study Changes Ul i\CDF U2 i\CDF 2x ESWP Unavailability 8.03E-09 l.03E-08 Ul ICCDP U2 ICCDP Single 14 day TS entry 2.60E-09 3.33E-09 APLA. RA/ #2-1998 Peer Review: a. Attachment 4, Page 2 of 41 states that, "[s]ubsequent to the peer review, the model has been updated to address all Category A, B, and D F&Os. Category B items from the 1998 NE/ PRA Peer Review (all closed) are listed below." Are there no Category A F&O's? If there are, provide their summary and disposition in the response to RAJ #7. Dominion Response:

No Category A F&Os were identified in the 1998 WOG Surry peer review.

APLA, RA/ #3: Serial No. 17-016A Docket Nos. 50-280/281 Page 3 of 76 Attachment 4, in the table entitled "HRA SCIENTECH PEER REVIEW COMMENTS" the resolution for PRA changes implemented is not provided. The staff requests the licensee provided the resolution (e.g., PRA change) for each peer review comment and explain its impact to the requested extension of the allowed outage time in the response to RA/#7. Dominion Response: Table RAl7, which is provided in the response to RAl#7, includes the resolution for the 2004 HRA Scientech Peer Review F&Os and includes a discussion of the impact to the requested extension where applicable. APLA, RA/ #14 -2010 Peer Review: The NRG staff expects that the PRA used to support the LAR has been reviewed according to RG 1.200, Revision 2 and associated NE/ guidance. The NRG staff also expects the LAR includes all findings (F&Os, gaps, etc.) that have been identified, excluding those removed by a subsequent peer review on the associated SRs. ' Attachment 4 of the submittal explains that a focused peer review was performed in . February 2010 for the SPS internal events PRA. It states that a review was conducted to determine if a focused scope or full scope peer review was required per the ASME PRA standard (RA-Sb-2005). Further discussion explains that an evaluation was supported by Addendum to ASMEIANS PRA Standard RA-S-2008 and NRC clarification and qualifications as provided in Appendix A of RG 1. 200, Rev. 2. The staff requests that the licensee provided the following additional information:

a. Clarify which version of . the standard the February 2010 peer review was performed against? Dominion Response:

Table RAl4a that follows is provided for clarification as to which versions of the PRA standard and RG 1.200 the Surry PRA model has been reviewed against. As specified in this table, the February 2010 peer .review was reviewed against PRA Standard RA-Sa-2009 and RG 1.200 Rev. 2. Serial No. 17-016A Docket Nos. 50-280/281 Page 4 of 76 Table RAl4a: Surry Peer Review History Review Reg. Guide Review Type Year Scope PRA Standard 1.200 Revision WOG Peer Review Peer Review 1998 Full Process N/A Peer Review 2004 Partial RA-S-2002 Trial Use Self Assessment I Gap Assessment 2009 Full RA-Sb-2005 1 Peer Review 2010 Partial RA-Sa-2009 2 Peer Review 2012 Partial RA-Sa-2009 2 b. Provide summary of each F&Os identified by the peer review team and a disposition of each F&O (i.e., assess the impact to the risk analysis supporting the requested extension for the service water flow path allowed outage time change) in the response to RAJ #7. Dominion Response: Table RAl7 provided in the response to RAl#7 includes the resolution for the 2010 Peer Review F&Os and includes a discussion of the impact to the requested extension where applicable.

c. Attachment 4, Page 20 of 41 states that, " ... changes in the IE, HR and QU elements only required specific HLR verification." Confirm that the focused scope peer review addressed all SR's in each HLR that required verification or provide further clarification.

Dominion Response: The 201 O focused scope peer review addressed all the SRs in the QU Element. For the HR element, the peer review addressed all SRs in the following HLRs: HLR-HR-E, HR-F, HLR-HR-G, HLR-HR-H, and HLR-HR-1. For the IE element, the peer review addressed the following SRs in HLR-IE-C: IE-C6, IE-C7, IE-C8, IE-C9, IE-C10, IE-C11, and IE-C12. Serial No. 17-016A Docket Nos. 50-280/281 Page 5of76 d. Attachment 4, Page 20 of 41 states that, "[t]he AS and IF elements required a full review against the high level requirements (HLRs)." Confirm that the focused scope peer review addressed all the SRs in the AS and IF technical elements or provide further clarification. Dominion Response: The 2010 focused scope peer review addressed all the SRs in the AS and IF technical elements. APLA. RA/ #5 -Self-Assessment: Jn accordance with RG 1.200, Revision 2, it is expected that the differences between the current version of the PRA standard (i.e., ASMEIANS RA-Sa-2009) and the earlier version of the standard used in the internal events PRA peer review (i.e., ASMEIANS RA-Sb-2005) be identified and addressed (i.e., perform a gap assessment).

a. Explain how the self-assessment was performed to identify the eleven gaps? What version of the standard did the self-assessment use to evaluate the SPS internal events model? Dominion Response:

The self assessment was performed by reviewing the Surry internal events PRA model files and documentation against the requirements in PRA standard RA-Sb-2005 and RG 1.200 Rev. 1. Dominion PRA engineers reviewed the model against each SR and identified gaps wherever it was judged the PRA model did not meet Capability Category 11. b. Provide a gap assessment of the internal events PRA against ASMEIANS RA-Sa-2009, as qualified by RG 1.200, Revision. 2 [Section 3.3, "Gap Assessment for PRAs Reviewed against RG 1.200, Revision 1," of NE/ 05-04, Revision 3, provides guidance on performing a gap assessment. Dominion Response: Table RAl5b below lists all SRs in PRA Standard RA-Sa-2009 and RG 1.200 Rev. 2 and documents the most recent peer review that assessed each SR. A gap assessment is provided for SRs that were not reviewed against RG 1.200 Rev. 2. --------------=-----::---------- ---*---**--------::---- Serial No. 17-016A Docket Nos. 50-280/281 Page 6 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment RA-Sa-Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment 2009 SR Peer Review Reviewed IE-A1 1998 Full None *Note 1 Scope 2012 IE-A2 Focused 2 None Required Scope IE-A3 1998 Full None *Note 1 Scope IE-A4 1998 Full None *Note 1 Scope 2012 IE-AS Fopused 2 None Required Scope IE-A6 1998 Full None *Note 1 Scope IE-A? 1998 Full None *Note 1 Scope IE-A8 1998 Full None *Note 1 Scope IE-A9 1998 Full None *Note 1 Scope IE-A10 1998 Full None *Note 1 Scope IE-81 1998 Full None *Note 1 Scope IE-82 1998 Full None *Note 1 Scope IE-83 1998 Full None *Note 1 Scope IE-84 1998 Full None *Note 1 Scope IE-85 1998 Full None *Note 1 Scope IE-C1 1998 Full None *Note 1 Scope IE-C2 1998 Full None *Note 1 Scope ..

*------

--*---- RA-Sa-2009 SR IE-C3 IE-C4 IE-CS IE-C6 IE-C7 IE-CB IE-C9 IE-C10 IE-C11 IE-C12 IE-C13 IE-C14 IE-C15 IE-01 IE-02 IE-03 Serial No. 17-016A Docket Nos. 50-280/281 Page 7 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2012 Focused 2 None Required Scope 1998 Full None *Note 1 Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 1998 Full None *Note 1 Scope 2012 Focused 2 None Required Scope 2012 ,., Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required ' Scope 2012 Focused 2 None Required Scope _*-

-*--_-r_

-**-::----

*---------.--

RA-Sa-2009 SR AS-A1 AS-A2 AS-A3 AS-A4 AS-AS AS-A6 AS-A7 AS-A8 AS-A9 AS-A10 AS-A11 AS-81 AS-82 AS-83 AS-84 AS-85 Serial No. 17-016A Docket Nos. 50-280/281 Page 8 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 -. Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope . ---* -. ------**--:.-::--- -".:'--:-"":' __ -.---_-_ ----::_---_: -*--*- RA-Sa-2009 SR AS-86 AS-87 AS-C1 AS-C2 AS-C3 SC-A-1 SC-A2 SC-A3 SC-A4 SC-A5 SC-A6 SC-81 SC-82 SC-83 SC-84 SC-85 SC-C1 Serial No. 17-016A Docket Nos. 50-280/281 Page 9 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope* 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope .. 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope .* .,.; RA-Sa-2009 SR SC-C2 SC-C3 SY-A1 SY-A2 SY-A3 SY-A4 SY-AS SY-A6 SY-A? SY-AS SY-A9 SY-A10 SY-A11 SY-A12 SY-A13 SY-A14 Serial No. 17-016A Docket Nos. 50-280/281 Page 10 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 1998 Full None *Note 1 Scope The 2009 Self Assessment was performed to assess potential gaps to RA-Sb-2005 I RG 1.200 Rev. 1 requirements. This assessment assessed 1998 Full None the SR as not meeting capability category II. This Scope issue was resolved in the S007 Aa model that was used to support the proposed AOT extension risk assessment. Per NEI 05-04 Rev. 3 no re-evaluation is required for RA-Sa-2009 I RG 1.200 Rev. 2. 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full. None *Note.1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope .. RA-Sa-2009 SR SY-A15 SY-A16 SY-A17 SY-A18 SY-A19 SY-A20 SY-A21 SY-A22 SY-A23 SY-A24 SY-81 SY-82 SY-83 SY-84 SY-85 SY-86 SY-87 Serial No. 17-016A Docket Nos. 50-280/281 Page 11 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope -* RA-Sa-2009 SR SY-88 SY-89 SY-810 SY-811 SY-812 SY-813 SY-814 SY-815 SY-C1 SY-C2 SY-C3 HR-A1 HR-A2 Serial No. 17-016A Docket Nos. 50-280/281 Page 12 of 76 . Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1. Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope The 2009 Self Assessment was performed to assess potential gaps to RA-Sb-2005 I RG 1.200 Rev. 1 requirements. This assessment assessed the SR as not meeting capability category II 1998 Full None because system walkdowns were not fully Scope documented in the S007 Aa model. This documentation issue has no impact on the proposed AOT extension. Per NEI 05-04 Rev. 3 no re-evaluation is required for RA-Sa-2009 / RG 1.200 Rev. 2. The 2009 Self Assessment was performed to assess potential gaps to RA-Sb-2005 I RG 1.200 Rev. 1 requirements. This assessment assessed 1998 Full None the SR as not meeting capability category II. This Scope issue was resolved in the S007 Aa model that was used to support the proposed AOT extension risk assessment. Per NEI 05-04 Rev. 3 no re-evaluation is required for RA-Sa-2009 I RG 1.200 Rev. 2. 2004 Focused Trial Use *Note 1 Scope 2004 Focused Trial Use *Note 1 Scope , .. RA-Sa-2009 SR HR-A3 HR-81 HR-82 HR-C1 HR-C2 HR-C3 HR-01 HR-02 HR-03 HR-04 HR-05 HR-06 HR-07 HR-E1 Serial No. 17-016A Docket Nos. 50-280/281 Page 13 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2004 Focused Trial Use *Note 1 Scope 2004 Focused Trial Use *Note 1 Scope 2004 Focused Trial Use *Note 1 Scope 2004 Focused Trial Use *Note 1 Scope 2004 ( Focused Trial Use *Note 1 Scope 2004 Focused Trial Use *Note 1 Scope 2004 Focused Trial Use *Note 1 Scope 2004 Focused Trial Use *Note 1 Scope 2004 Focused Trial Use *Note 1 Scope 2004 Focused Trial Use *Note 1 Scope 2004 Focused Trial Use *Note 1 Scope The 2009 Self Assessment was performed to assess potential gaps to RA-Sb-2005 I RG 1.200 2004 Rev. 1 requirements. The clarification for HR-06 in Focused Trial Use RG 1.200 Rev. 2 was reviewed. There are no gaps Scope in the characterization of uncertainty for pre-initiator human failure events, so there is no impact on the proposed ESW AOT extension from the RG 1.200 Rev. 2 clarification. 2004 Focused Trial Use *Note 1 Scope 2010 Focused 2 None Required Scope - Serial No. 17-016A Docket Nos. 50-280/281 Page 14 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment RA-Sa-Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment 2009 SR Peer Review Reviewed 2010 HR-E2 Focused 2 None Required Scope 2010 HR-E3 Focused 2 None Required Scope 2010 HR-E4 Focused 2 None Required Scope 2010 HR-F1 Focused 2 None Required Scope 2010 HR-F2 Focused 2 ' None Required Scope 2010 HR-G1 Focused 2 None Required Scope 2010 HR-G2 Focused 2 None Required Scope* 2010 HR-G3 Focused 2 None Required Scope 2010 HR-G4 Focused 2 None Required Scope 2010 HR-G5 Focused 2 None Required Scope 2010 HR-G6 Focused 2 None Required Scope 2010 HR-G7 Focused 2 None Required Scope 2010 HR-G8 Focused 2 None Required Scope 2010 HR-H1 Focused 2 None Required Scope 2010 HR-H2 Focused 2 None Required Scope 2010 HR-H3 Focused 2 None Required Scope , ---::-------'.'.'.- -- --* . --------_:-- --*-

RA-Sa-2009 SR HR-11 HR-12 HR-13 DA-A1 DA-A2 DA-A3 DA-A4 DA-81 DA-82 DA-C1 DA-C2 DA-C3 DA-C4 DA-C5 DA-C6 DA-C7 DA-C8 Serial No. 17-016A Docket Nos. 50-280/281 Page 15 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 1998 Full None *Note 1 . Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope -------._ -. .. RA-Sa-2009 SR OA-C9 OA-C10 OA-C11 OA-C12 OA-C13 OA-C14 OA-C15 OA-C16 OA-01 OA-02 DA-03 OA-04 OA-05 OA-06 OA-07 OA-08 OA-09 Serial No. 17-016A Docket Nos. 50-280/281 Page 16 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 1998 Full None *Note 1 Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 1998 Full None *Note 1 Scope This SR was added in RG 1.200 Rev. 2 and has not None NIA been peer reviewed. The S007 Aa Surry PRA model does not model equipment repair, so there is no impact on the proposed ESW AOT extension. RA-Sa-2009 SR DA-E1 DA-E2 DA-E3 IFPP-A1 IFPP-A2 IFPP-A3 IFPP-A4 IFPP-AS IFPP-81 IFPP-82 IFPP-83 IFSO-A1 IFSO-A2 IFSO-A3 IFSO-A4 IFSO-AS Serial No. 17-016A Docket Nos. 50-280/281 Page 17 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused

2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 ' Focused 2 None Required Scope RA-Sa-2009 SR IFSO-A6 IFS0-81 IFS0-82 IFS0-83 IFSN-A1 IFSN-A2 IFSN-A3 IFSN-A4 IFSN-A5 IFSN-A6 IFSN-A7 IFSN-A8 IFSN-A9 IFSN-A10 IFSN-A11 IFSN-A12 Serial No. 17-016A Docket Nos. 50-280/281 Page 18 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope ' 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope RA-Sa-2009 SR IFSN-A13 IFSN-A14 IFSN-A15 IFSN-A16 IFSN-A17 IFSN-81 IFSN-82 IFSN-83 IFEV-A1 IFEV-A2 IFEV-A3 IFEV-A4 IFEV-A5 IFEV-A6 IFEV-A7 IFEV-AB Serial No. 17-016A Docket Nos. 50-280/281 Page 19 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2010 Focused 2 None Required Scooe 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scooe 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scooe 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required . Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scooe 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scooe 2010 Focused 2 None Required Scooe 2010 Focused 2 None Required Scope RA-Sa-2009 SR IFEV-81 IFEV-82 IFEV-83 IFQU-A1 IFQU-A2 IFQU-A3 IFQU-A4 IFQU-AS IFQU-A6 IFQU-A7 IFQU-A8 IFQU-A9 IFQU-A10 IFQU-A11 IFQU-81 IFQU-82 Serial No. 17-016A Docket Nos. 50-280/281 Page 20 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required -Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope RA-Sa-2009 SR IFQU-83 QU-A1 QU-A2 QU-A3 QU-A4 QU-A5* QU-81 QU-82 QU-83 QU-84 QU-85 QU-86 QU-87 QU-88 QU-89 QU-810 Serial No. 17-016A Docket Nos. 50-280/281 Page 21 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 *, Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 1998 Full None *Note 1 Scope 1998 Full None *Note 1 Scope RA-Sa-2009 SR QU-C1 QU-C2 QU-C3 QU-01 QU-02 QU-03 QU-04 QU-05 QU-06 QU-07 QU-E1 QU-E2 QU-E3 QU-E4 QU-F1 QU-F2 Serial No. 17-016A Docket Nos. 50-280/281 Page 22 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2010 Focused 2 None Required Scope 2010 . Focused 2 None Required Scope 1998 Full None *Note 1 Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope RA-Sa-2009 SR QU-F3 QU-F4 QU-FS QU-F6 LE-A1 LE-A2 LE-A3 LE-A4 LE-AS LE-81 LE-82 LE-83 LE-C1 LE-C2 LE-C3 LE-C4 Serial No. 17-016A Docket Nos. 50-280/281 Page 23 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required Scope 2010 Focused 2 None Required
Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused . 2 None Required Scope 2012 Focused 2 None Required Scooe 2012 Focused 2 None Required Scope 2012 Focused 2 Non.e Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope RA-Sa-2009 SR LE-C5 LE-C6 LE-C7 LE-CS LE-C9 LE-C10 LE-C11 LE-C12 LE-C13 LE-01 LE-02 LE-03 LE-04 LE-05 LE-06 LE-07 Serial No. 17-016A Docket Nos. 50-280/281 Page 24 of 76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope RA-Sa-2009 SR LE-E1 LE-E2 LE-E3 LE-E4 LE-F1 LE-F2 LE-F3 LE-G1 LE-G2 LE-G3 LE-G4 LE-GS LE-G6 Serial No. 17-016A Docket Nos. 50-280/281 Page 25of76 Table RAl5b: Peer Review SRs Reviewed and Gap Assessment Last Surry RG 1.200 Version RA-Sa-2009 I RG 1.200 Gap Assessment Peer Review Reviewed 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope , 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope 2012 Focused 2 None Required Scope *Note 1 -The 2009 Self Assessment was performed to assess potential gaps to RA-Sb-2005 I RG 1.200 Rev. 1 requirements.

Per NEI 05-04 Rev. 3, no re-evaluation is required for RA-Sa-2009 I RG 1.200 Rev. 2. Serial No. 17-016A Docket Nos. 50-280/281 Page 26 of 76 APLA. RA/ #6 -2012 Focused Scope Peer Review: Attachment 4 of the submittal explains that a focused peer review was performed in June 2012 for the SPS internal events PRA. It states that a review was performed against the requirements of the ASMEIANS PRA standard RA-Sb-2005. Previous discussion from the 2010 Peer Review states that an evaluation was supported by Addendum to ASMEIANS PRA Standard RA-S-2008.

a. Clarify which version of the standard the June 2012 peer review was performed against and provide a list identifying all the technical elements that were reviewed for this focused scope peer review. Dominion Response:

The June 2012 Peer Review was performed against PRA standard RA-Sa-2009 and RG 1.200 Rev. 2. The SRs assessed in this peer review are documented in Table RAl6a below. Table RAl6a: Scope of the 2012 Surry Peer Review Technical High Level Supporting Requirements Element Requirement (SRs) Covered Focused Scope s (HLR) IE-A IE-A2, IE-A5 Focused on the ISLOCA analysis. This IE-C3, IE-C5, IE-C6, IE-CB, included consideration of applicable AS IE IE-C IE-C9, IE-C10, IE-C11, IE-SRs. However, since not all event trees C12, IE-C14, IE-C15 were reviewed, any issues identified with the ISLOCA event tree were documented IE-D IE-D1 I IE-D2, IE-D3 under the IE SRs. SY SY-8 SY-81, SY-82, SY-83, SY-Focused on modeling of common cause 84 failures. DA-A DA-A 1, DA-A2, DA-A3, DA DA-D DA-D2, DA-D3, DA-D5, DA-Focused on data for common cause D6, DA-D7 failures. DA-E DA-E1 I DA-E2, DA-E3 Focused on inclusion of state-of-QU-A QU-A3 knowledge correlation in the ISLOCA au frequency analysis. QU-8 QU-87 Focused on logic for removal of mutually exclusive CCF basic events. APLA, RA/ #7: Serial No. 17-016A Docket Nos. 50-280/281 Page 27of76 Provide a single table with all current peer review findings, a summary of each finding, the licensee's resolution (e.g., PRA change) of each resolved finding, and an explanation (i.e., disposition) of the impact of each unresolved finding to the requested extension of the allowed outage time. Dominion Response: Table RAl7, shown below, was constructed to document a consolidated list of all current peer review findings and the additional information requested. In some cases the changes documented in the 'Dominion Resolution' column for older peer review findings describe legacy modeling changes that have been superseded by more recent modeling changes. The original resolutions are provided because they most directly address the peer review findings. For example, the table entry F&O IE-8 in Table RAl7 below describes how this 1998 F&O was originally addressed without a Loss of Circulating Water fault tree, even though this fault tree has since been added to the Surry model and is included in the S007 Aa model used to support the proposed AOT extension. Current Peer Peer Review Review Findings WOG 1998 F&O AS-2 Level B (Element AS-5 and AS-9) Serial No. 17-016A Docket Nos. 50-280/281 Page 28 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension The models and analyses are consistent, as best PRA update guidance was developed, which F&O is resolved. as the reviewers could determine, with the as includes a review of plant procedures built plant, and were consistent with plant (EOPs), assumptions for components, and operating procedures at the time the IPE was system recovery models based upon human completed. However, there is no process in actions. This guidance is described in place to identify and incorporate changes in plant procedure NF-AA-PRA-410 PRA operation into the PRA model. This process Configuration Control Program. The current should also include periodic review of industry industry guidance suggests a voluntary standards that may impact the PRA. Some periodic review of industry standards, which examples of where such a process could impact will be considered as resources allow. the model include, the timing for switchover to hot leg recirculation after event initiation (9 hours in the current EOP), and a review. of potential impacts on the PSA due to the power uprate program. The focus of this comment is on the lack of process more than any current discrepancies found in the model, and is related to the IPE Maintenance and Update Process elements. Current Peer Peer Review Review Findings WOG 1998 F&O AS-8 Level B (Element AS-12) Serial No. 17-016A Docket Nos. 50-280/281 Page 29 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension The RCP seal LOCA model appears to include An RC pump seal failure model (the so-called F&O is resolved. an optimistic interpretation of the WOG and NRC Rhodes model) that is acceptable to the NRC models, and does not include a contribution from for use in PRA was developed and early seal failure. implemented for the T4, T1A and T6 event trees. This model addresses the probability of early seal failure, and does not allow restoration of seal cooling after a relatively brief cooling loss. For Surry, the model is discussed fully in SM-1296, implemented in the SOA-D PRA models. For the S03A model, the T1A (SBO) accident sequence model was revised to be fully consistent with the WOG2000 RCP seal LOCA model, and the T 4 and T6 accident sequences models were revised to incorporate simplified logic consistent with the WOG2000 model. Future model updates are planned to include additional model detail (and remove conservatisms) for the seal failure contribution for events other than T1A. Current Peer Peer Review Review Findings WOG 1998 F&O DA-8 Level B (Element DA-12/DE-9) WOG 1998 F&O DA-9 Level B (Element DA-9) Serial No. 17-016A Docket Nos. 50-280/281 Page 30 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension (Implementation of NUREG/CR-4780 The common cause fault (CCF) approach is F&O is resolved. methodology) Reviewers question the validity of revised to incorporate the following: Alpha the approach used for defining CCF terms, by factor model, INEEL data base of CCF events adding fail to start and fail to run data variables. from NUREG/CR-6268, different failure Method added value of QD and Lambda, but the modes (run and demand), and different CCF events are not consistent (i.e. per-demand and events based upon population size (e.g., 2 of per-hour). Assuming a mission time of one hour 3 as well as 3 of 3 CCF events.) Guidance and a demand for the device, the terms can be for the CCF models was taken from added. But what if: NUREG/CR-5485, which extends the 1. Common cause failure is dominated by technology developed for NUREG/CR-4780. running failures, there is no mission time associated with the use of the common cause term -non-conservative result 2. Running failure rate is comparable to start term, but common cause dominated by start terms -overly conservative result. The common cause failure probability of valves The common cause fault (CCF) approach is F&O is resolved. failing due to plugging is (0.1)(1.25-7 f/hr)(2160 revised to incorporate the following: Alpha hrs), or about 1 E-4. The 0.1 beta factor used for factor model, INEEL data base of CCF events this calculation may be overly conservative. The from NUREG/CR-6268, different failure net result is that many of the top sequences (for modes (run and demand), and different CCF the 3-year maintenance case) involve common events based upon population size (e.g., 2 of cause valve plugging terms. It is unusual to 3 as well as 3 of 3 CCF events.) Guidance have passive equipment failures be so prominent for the CCF models was taken from in the dominant cutsets (more prominent than NUREG/CR-5485, which extends the active equipment failures). technology developed for NUREG/CR-4780. Current Peer Peer Review Review Findings WOG 1998 F&O DA-6 Level B (Element DA-11) WOG 1998 F&O DE-3 Level B (Element DE-8) Serial No. 17-016A Docket Nos. 50-280/281 Page 31 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O ** Summary of Finding Dominion Resolution on Proposed AOT extension The models for the EDGs do include common Miscalibration of instrumentation channels is F&O is resolved. cause failures of fuel oil system. In general the resolved as a human reliability rather than an models do consider common/shared equipment common cause fault. The HEP components and support systems explicitly. The fault behaves the same as an equipment models do not appear to include the effects of CCF, but is quantified on the basis of human common maintenance crews or l&C technicians. error rather than equipment reliability. HEP Specifically, there is no consideration of common events are created for the following cause miscalibration of instrumentation instrument channels: EOG, 1-LMPT-channels. 100A/B/C/D, MS flow, MS differential pressure, steam generator level, pressurizer pressure RWST level, intake canal level and RC delta T and TAVE. The methods used to determine CCF groups are The common cause fault (CCF) approach F&O is resolved. simplistic. Determination of the set of active was revised to incorporate the following: components based on 1 % contribution to CDF Alpha-factor model, INEEL data base of CCF severely limits the number and type of common events from NUREG/CR-6268, different cause terms used in the model. As an failure modes (run and demand), and evaluation tool for plant vulnerabilities (i.e., the different CCF events based upon population IPE), it is more than sufficient, but as an size (e.g., 2 of 3 as well as 3 of 3 CCF evaluation tool for Risk-informed Applications, it events.) Guidance for the CCF models was is not enough. taken from NUREG/CR-5485, which extends the technology developed for NUREG/CR-Events that should be considered include: 4780. Breaker fail to operate (Open/Close) Auxiliary Feedwater Pumps (back-leakage) Ventilation fans Current Peer Peer Review Review Findings WOG 1998 F&O HR-4 Level B (Element HR-15) WOG 1998 F&O HR-5 Level B (Element HR-26) Serial No. 17-016A Docket Nos. 50-280/281 Page 32 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension HEP development for the IPE model was The HEP events developed since the IPE F&O is resolved. extensively documented; however, HEPs have received detailed analysis. developed for subsequent updates of the IPE model were not as well documented (and by implication, were not developed in as much detail). For many of the HEPs in subsequent updates, a value of 0.1 was used. It is not clear whether this is a screening value or some other value. In a sensitivity analysis (SM-1174, Addendum A) The HRA dependencies were reanalyzed for F&O is resolved. to evaluate dependency among His contained in the 8007 A model. Potential HEP cutsets, time between actions was listed as the dependencies include: (a) the time required major factor in establishing independence of the to complete all actions in relation to the time operator actions. In most cases, time (itself) is available to perform the actions, (b) factors not an adequate factor, but is a parameter which that could lead to dependence (e.g., common can be associated with more defensible factors. instrumentation, common procedures, For example, one cutset contained two HEPs -increased stress, etc.), and (c) availability of one for early SG isolation following a SGTR and resources (e.g., personnel accounting for one for late SG isolation. The time difference of time of day). several hours between the actions was cited as the basis for the actions' independence. Better factors for independence might have been different clues caliing for the need to isolate the SG or actuation of the TSC, or additional/new crew for the late isolation. All of these are related to time, but time (itself) is not the factor. Current Peer Peer Review Review Findings WOG 1998 F&O HR-2 Level B (Element HR-4,7/DE-7/SY-8) WOG 1998 F&O IE-4 Level B (Element IE-7) Serial No. 17-016A Docket Nos. 50-280/281 Page 33 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension Table D.1-1 of Section D.1 of the Surry IPE lists Miscalibration of instrumentation channels is F&O is resolved. the pre-initiator errors considered in the analysis. resolved as a human reliability rather than an The list contains only mispositioning events equipment common cause fault. The HEP (valves, blank flanges, etc.). fault behaves the same as an equipment CCF, but is quantified on the basis of human No instrument miscalibration events are error rather than equipment reliability. For contained in the list. The procedure for system Surry, HEP events are created for the analysis (page 19 of 58) indicates that common following instrument channels: EDG, 1-LMPT-cause His should be modeled for miscalibration 100NB/C/D, MS flow, MS differential of instruments used to initiate systems following

pressure, steam generator level, pressurizer an action or in any standby equipment items pressure RWST level, intake canal level and such as the level instrumentation in storage RC delta T and TAVE. tanks. A recent industry event (Oconee) involved a The Oconee failure mechanism is not F&O is resolved.

small break LOCA (>10 gpm) at the charging line considered valid for the Surry design, and connection to the RCS. The mechanism for the should not require LOCA frequency crack in the thermal sleeve at the connection adjustment. The current Surry LOCA point was thermal fatigue. Is the Surry piping_ frequencies are developed from NUREG/CR-subject to this type of event? If so, has it been 5750. This document observed that no small considered in the initiating event frequency? LOCA events had occurred in U. S. nuclear power plants up to 1995. However, the 1997 Oconee 2 event could possibly be categorized as a very small LOCA / leak, and four such events from 1987 -1995 are included within the NUREG/CR-5750 initiating event frequencies. (ref. additional discussion in NE-1436) Current Peer Peer Review Review Findings WOG 1998 F&O IE-5 Level B (Element IE-14) WOG 1998 F&O IE-3 Levers (Element IE-13) Serial No. 17-016A Docket Nos. 50-280/281 Page 34 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension An industry issue of about 5-6 years ago was the The suggested scenario is isolatable and F&O is resolved. creation of an ISLOCA caused by a leak in an therefore does not represent a credible RCP thermal barrier heat exchanger and a ISLOCA. The event would be a small LOCA failure to isolate the CCW lines that provided for which the frequency is captured in the cooling water to the heat exchanger. How was overall random small LOCA frequency. this potential ISLOCA pathway treated by the initiating events analysis? Does it apply to the Surry model? Initiating event frequencies have not been The initiating events are updated by several F&O is resolved. updated since the IPE submittal in 1991. As a sources. The rare initiator frequencies from result, recent industry information and operating NUREG/CR-5750 are used as priors for experience have not been incorporated into the Bayesian updating with plant specific initiating events analysis. This information could histories. The moderate frequency transient alter the initiating event frequencies currently initiating event frequencies are created from contained in the model. plant specific data (1990-2000 LERs) and a non-informative gamma prior distribution. Finally, some plant unique initiating events are quantified with new fault tree models directly linked to the integrated PRA model. These models are discussed fully in SM-1307, implemented in SOA-D, with closure on 5/16/2001. Current Peer Peer Review Review Findings WOG 1998 F&O IE-8 Level B (Element IE-13) Serial No. 17-016A Docket Nos. 50-280/281 Page 35of76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension The FMEA portion of the Initiating Events The Surry LOSW IE is actually an evolution of F&O is resolved. notebook (page 12 of 28) states that screen a loss of circ water initiator. We do not wash pumps do not have to operate during an currently model this with a fault tree, so there accident. The implication is that because of this is no specific screen clogging contribution there is no need to consider the screen wash identified for this initiator. Currently, the Surry system further. However, clogged screens can PRA uses a plant specific Bayesian update of cause plant trips, and this failure mechanism generic industry experience for loss of should be considered in the development of Circulating Water to evaluate the IE-T6 (Loss initiating event frequencies. Recent industry of Circulating Water) frequency. The plant events at Salem and Wolf Creek illustrate a

specific clogged screen failure event is plant's susceptibility to clogged intake screens. therefore considered, since it is part of the overall industry and plant-specific experience leading to loss of CW. However, an evaluation has been performed for the S03A model update to establish that the IE frequency used for this event would encompass contributions from events such as clogged intake screens I screen wash faults. Should the model be changed in the future to model this IE with a fault tree, this failure mechanism would be addressed explicitly.

Current Peer Peer Review Review Findings WOG 1998 F&O IE-9 Level B (Element IE-16) WOG 1998 F&O L2-2 Level B (Element L2-8 & 10) Serial No. 17-016A Docket Nos. 50-280/281 Page 36 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension The reactor core has been upgraded to 2586 The effect of the 4.5% core power uprate on F&O is resolved. MWt. Has the effect of this change been the timing of HEPs used in the SPS PRA considered on the moderator temperature Model and on the success criteria of coefficienUreactivity feedback, particularly for hardware credited in the SPS PRA Model has early in a core's life? Also, has the increased been evaluated using MMPA.0.5. The. decay heat load been considered in the success results of the analysis show that no changes criteria for decay heat removal? are required to the current success criteria or HEP calculations. The details of the analysis are documented in SPS Notebook SPS-RA.MD.SC.001 Rev 0. The consequences of operator actions after core The Level I model takes into account all F&O is resolved. damage are not considered in the PSA or the human error probabilities due to EOP actions. LERF assessment. After core damage has The human error probabilities based on occurred, the control room staff will continue to SAMG actions are not incorporated into the attempt to implement EOP actions (and now model. According to WOG procedures, once SAMG actions). the core temperature reaches 1200 °F, the operator leaves the EOPs and enters the Considering the EOP actions, only those that SAMGs (control room guideline SACRG-1). prevent core damage (have an impact on the Our Level I model was developed CDF) are modeled in the Level 1 PSA. Several independent of the SAMG actions. It is EOP actions that can impact the LERF are: recognized that inappropriate SAMG actions

FR-C.1 actions to depressurize the RCS at the may cause negative consequences which onset of core overheating greatly decreases the may result in greater source term releases to probability of a high pressure reactor vessel the atmosphere.

For this reason, a Technical failure, while significantly increasing: a) the Support Center (TSC) is formed that reviews potential for core concrete interactions, and b) real time plant parameter data and provides the fission product release from RCS to expert guidance to the operation staff during containment (which, in tum, increases the source a severe accident condition. Additionally term for containment failures). training on the SAMGs is provided every 3

FR-H.1 actions to establish some type of years which includes a discussion of these feedwater flow to the SGs increases the chances cautions and recommendations to the Current Peer Peer Review Review Findings Serial No. 17-016A Docket Nos. 50-280/281 Page 37 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension of SG tube failure due to thermal stresses of cold operators.

The operators are taught to be water being injected onto hot SG tubes, but can aware of their plants most dominant accident also increase the potential for arresting the core sequences and the consequences of damage in-vessel. These two aspects can inadequate actions. impact the LERF.

ECA-0.0 actions to start sprays when offsite power is restored.

This can prevent overpressure failure of containment, but can also de-inert containment and lead to a hydrogen burn. When combined with the added hydrogen from in-vessel recovery, the hydrogen burn may ch,allenge containment. Also, these operator actions should be substantiated by an HRA analysis to determine the HEP. The plant has also completed implementation of the SAMG. The SAMG contains a set of accident management strategies that would be implemented for each of the core damage accidents. The implementation of some of the strategies has negative consequences that should be addressed. Current Peer Peer Review Review Findings WOG 1998 F&O MU-4 Level B (Element MU-4) WOG 1998 F&O MU-3 Level B (Element MU-4) Serial No. 17-016A Docket Nos. 50-280/281 Page 38 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension Activities to evaluate the effects on the PSA of The Surry PRA model has been updated to F&O is resolved. changes to equipment failure rates, initiator include changes in data. The initiating events frequencies, and human error probabilities are were updated as documented in calculation minimal. SM-1370. Component unavailabilities were updated for Surry as documented in calculation SM-1308. The component reliabilities for risk. significant pumps and the EDGs were Bayesian updated with plant specific data for Surry as documented in SM-1311. Further, a full data update was performed in 2005 for the MSPI update, as documented in the DA series of PRA notebooks. The HEPs were reviewed and updated as necessary following the PRA self assessment per RG1 .200 and in response to the subsequent HRA re-peer review comments. Input from Operations personnel at Surry was obtained to provide better estimation of the times associated with the performance of emergency procedures. A PRA update process is described in procedure NF-AA-PRA-410 PRA Configuration Control Program. Requirements for review of operating PRA update guidance was developed, which F&O is resolved. experience, plant procedures, and plant includes a review of: (1) Technical controlled documents in support of a PSA update Specification revisions, (2) station are not detailed in the PSA guidance documents. engineering Design Change lists, (3) station procedures, and (4) operating experience. Current Peer Peer Review Review Findings WOG 1998 F&O MU-2 Level B (Element MU-4) WOG 1998 F&O SY-11 Level B (Element SY-5) WOG 1998 F&O SY-2 Level B (Element SY-5) Serial No. 17-016A Docket Nos. 50-280/281 Page 39 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension The core power has been upgraded. Effects of The effect of the 4.5% core power uprate on F&O is resolved. this change have not been incorporated into the the timing of HEPs used in the SPS PRA PSA model. Factors that could be affected by the Model, and on the success criteria of core power upgrade include the moderator hardware credited in the SPS PRA Model has temperature coefficient (for ATWS) and the been evaluated using MMP 4.0.5. The decay heat load (for several accident results of the analysis show that no changes sequences). are required.to the current success criteria or HEP calculations. Dependency table from IPE model wasn't Updated System Success criteria and F&O is resolved. updated in SM-1162 or SM-1165 to account for Dependency table and documented in SY CCW dependency. Also, success criteria section notebooks. of system notebook was not updated. The program does not appear to have a formal PRA update guidance was developed, which F&O is resolved. requirement for incorporating changes based on includes a review of: plant design changes. For example, a later EOP (1) Technical Specification revisions, (2) change the time to hot leg recirculation station engineering Design Change lists, (3) switchover as 9 hours. The model says 16 station procedures, and (4) operating hours. experience. This guidance is described in procedure NF-M-PRA-410 PRA There is an advantage to identifying operator Configuration Control Program. actions to specific procedure steps. The downside is, procedures change. Thus, the models and documentation need to be updated periodically. Current Peer Peer Review Review Findings WOG 1998 F&O SY-5 Level B (Element SY-5) WOG 1998 F&O SY-4 Level B (Element SY-12) WOG 1998 F&O TH-2 Level B (Element TH-8) Serial No. 17-016A Docket Nos. 50-280/281 Page 40 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact ofF&O Summary of Finding Dominion Resolution on Proposed AOT extension The RPS logic model is incorrect. The fault tree Fault tree RP1 was revised to include F&O is resolved. indicates that success of either logic train allows separate logic for RTA and RTB including the challenge to both reactor trip breakers. Actual input logic signal with recovery. The models design is logic train A sends signal to RTA and also include failure of the trip breaker logic train B sends signal to RTB. (RTA/RTB), and RTA/RTB. recovery through the shunt trip relay (including failure of 125 VDC, human reliability model, and failure of the shunt trip relay). The RPS model does not properly identify the Fault trees RP1 for Surry were revised to F&O is resolved. required support systems. RPS logic receives include separate logic for RT A and RTB power from Class 1E125V DC buses 1A and 1 B. including the input logic signal with recovery. Failure of the DC buses removes power to the The models also include failure of the trip RTB shunt trip coils which limits operator action breaker (RTA/RTB), and RTA/RTB recovery in the control room if reactor trip fails. through the shunt trip relay (including failure of 125 VDC, human reliability model, and failure of the shunt trip relay). Several HVAC systems are modeled in detail As part of the 2007 model update, a room F&O is resolved. and are well documented. These include ESGR cooling dependency matrix was developed room cooling and the Auxiliary Building and added to notebook SY.1 rev 2. This Ventilation System, but these are the only matrix was used to determine which areas ventilation dependencies modeled in the PSA. had room heat up calculations developed to Some of the systems models provide a one line justify not modeling room ventilation. For assumption stating that room cooling is not areas that did not have a calculation or where required, but little if any basis is provided for the calculations indicated the need for these assumptions. Based.on discussions with ventilation, the ventilation systems were the PSA group engineers during this review, it added to the model. appears that the HVAC requirements were adequately addressed in the modeling process, but the assumptions were not clearly documented, and no process is defined for the determination of the need for room cooling. Current Peer Peer Review Review Findings HRA SJ-1 Level A/B Scientech (SR HR-G7) Peer Review 2005 HRA SJ-2 Level A/B Scientech (SR HR-G9) Peer Review 2005 Serial No. 17-016A Docket Nos. 50-280/281 Page 41 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension In the Surry HRA Post-Initiator Notebook, the Item 1 was resolved in S007 Aa by inclusion F&O is resolved. following dependency issues need to be of dependent combination replacement addressed. events for the two HEPs. Item 2 was no longer applicable as this HEP was removed 1. Section 2.4.4 Detailed Analysis of HEP-1 E0-8, from the modeling. HEP-2E0-8 Dependency. I would imagine this HFE occurs in cutsets along with SIS manual actuation (HEP-1 E0-4, HEP-2E0-4) and there may be a cognitive dependence.

2. Section 2.4.8 Detailed Analysis of HEP-1 EO-13-S1, HEP-2E0-13-S1 Dependence.

Good bullet identifying the relationship with other basic events. The next step is to describe how this dependence is evaluated (also applies to HEP-1E0-16). In the Surry HRA Post-Initiator Notebook, the Unresolved but tracked per PRA The impact of this following dependency issues need to be Configuration Control Program. gap on this addressed. . application was The Surry decision tree for evaluating evaluated by performing a dependencies is unique and key assumptions sensitivity study should be evaluated via sensitivity studies. For (Sensitivity

1) example, in the Surry approach the time window which multiplies (total time available) has more impact than on HEPS in the the time that the actions would occur. SOD? Aa model by Specifically, if two actions occurred within 5 minutes of each other but there was 2 hours 10. This available, then the dependence is assessed as. sensitivity low. demonstrates that this issue does not impact the results of this analysis.

Current Peer Peer Review Review Findings HRA F&O 2J-1 Level Scientech A/B (SR HR-Peer Review G5/G9) 2005 ** Serial No. 17-016A Docket Nos. 50-280/281 Page 42 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension In the Surry HRA Post-Initiator report the both Unresolved but tracked per PRA The impact of this the HCR/ORE and CBDTM HRA methods use Configuration Control Program. Ultimate gap on this timing data that was derived from operator resolution of this item will involve detailed application was surveys. talk-throughs with plant operations and evaluated by training personnel in accordance with performing a Section 2.3 Results of Operator Surveys. Supporting requirement HR-E3. a. How many crews? If these are all from 1 crew sensitivity study then this is not a representative, random set. (Sensitivity

1) b. Use of surveys is a good idea and has been which multiplies used at other utilities, but is somewhat non-HEPS in the standard.

Would be good to submit this idea to 8007 Aa model by the EPRI HRA Users Group. (Related to 10. This Guideline comment #5) sensitivity

c. Recommend adding the use of survey to the demonstrates HRA Guideline (since the guideline says, "The that this issue only source of data for such estimates is does not impact simulator experiments on the appropriate scenario.")

the results of this d. The weighting based on experience is non-analysis. standard and a would be beneficial. If there is not a referenceable basis, then this should be listed as a key assumption.

e. At some plants the STA shift length, training, and organization they are part of are different than the panel operators and thus might affect the weighting.

Recommend adding a statement as to whether these are the same or different at Surry. f. The total experience appears to be 99 years instead of 100. g. Tables 1-3 Precision. The data shown in Tables 1-3 are displayed as seconds. In some cases the data from the surveys was in minutes Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution (e.g. 10 minutes for a median response time is shown as 600 seconds). The number of significant figures in the Average should correspond to the input (e.g., 2 decimal places in the average may be too many). h. The weighting employed gives operator #7 more than 50% of the total weight. My observation of operators in a training environment is that often times the newest operators have a better cognizance and feel for -"-rare evenV accident scenarios (based on their recent training) than operators with 5-6 years experience (who are more cognizant of normal plant operations such as surveillance tests and routine maintenance). A sensitivity analysis with equal weighting, or higher weighting for newer operators, should be conducted. Serial No. 17-016A Docket Nos. 50-280/281 Page 43 of 76 Impact of F&O on Proposed AOT extension Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding . Dominion Resolution HRA F&O 4J-2 Level The following comments are made relative to the Unresolved but tracked per PRA Scientech NB detailed analysis documentation associated with Configuration Control Program. Peer Review each basic event in the Surry HRA Post-Initiator 2005 Notebook. These comments apply to the detailed analysis of each human failure event. 1. Section 2.4.1 Detailed Analysis of HEP-1 E0-1, HEP-2E0-1 Usage of the Basic Event. Need to explain the mismatch between the "IPE usage" bullet with that at the end of the section (current fault trees). 2. Section 2.4.1 Detailed Analysis of HEP-1 E0-1, HEP-2E0-1 HFE Usage. (Related to comment #1 in this section) -Add a description of the differences between scenarios (e.g. which case is the bounding case, what are the differences between scenarios and how/why the HFE applies to all scenarios modeled). For example, in Section 2.4.2 the Detailed Analysis of HEP-1 E0-4, HEP-2E0-4 shows the current model usage expands from the SINOHH1 to CHNOCH1. The IPE usage was for only S2 LOCAs but this BE could be modeled in steam/feedwater line breaks, consequential LOCAs such as RCP seal LOCA, and other initiators.

3. Section 2.4.14 Detailed Analysis of HEP-1 EO-10, HEP-2E0-10 CBDT. The cues should be stated. Serial No. 17-016A Docket Nos. 50-280/281 Page 44 of 76 Impact of F&O on Proposed AOT extension This is judged to be primarily a documentation consideration only. This issue does not affect sequences relevant to this application.

Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution HRA F&O 4J-3 Level The following comments are made relative to the Unresolved but tracked per PRA Scientech NB detailed analysis documentation associated with Configuration Control Program. Peer Review each basic event in the Surry HRA Post-Initiator 2005 Notebook. These comments apply to the detailed analysis of each human failure event. Section 2.4.1 Detailed Analysis of HEP-1 E0-1, HEP-2E0-1 Time Window. Add more specific references for the time window (e.g. the MAAP run document). HRA F&O 4J-4 Level The following comments are made relative to the Unresolved but tracked per PRA Scientech NB (SR HE-G6) detailed analysis documentation associated with Configuration Control Program. Peer Review each basic event in the Surry HRA Post-Initiator 2005 Notebook. These comments apply to the detailed analysis of each human failure event. Section 2.4.11 Detailed Analysis of HEP-1 E0-16, HEP-2E0-16 Cognitive and Execution. Many, if not all, CBDT cognitive modeling comes out to be 3E-3. Execution inconsistencies such as stating 15 valves to open but modeled as a simple action. Many, if not all, execution modeling comes out to be 1 E-4. For example, in Section 2.4.114 Detailed Analysis of HEP-1 E2-4, HEP-2E2-4 Execution, one of the required actions stated in the "Notes" is to locally close valves but this modeled as a simple action inside of the control room (see comment #22). This comment aoolies to HEP-1 E3-3 as well. Serial No. 17-016A Docket Nos. 50-280/281 Page 45 of 76 Impact of F&O on Proposed AOT extension This is judged to be primarily a documentation consideration only. This issue does not affect sequences relevant to this aoolication. The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

1) which multiplies HEPS in the S007 Aa model by 10. This sensitivity demonstrates that this issue does not impact the results of this analysis.

Current Peer Peer Review Review Findings HRA 78-1 Level A/B Scientech (SR HR-G9, H3) Peer Review 2005 Serial No. 17-016A Docket Nos. 50-280/281 Page 46 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summ.ary of Finding Dominion Resolution on Proposed AOT extension There are numerous places in the Surry HRA Unresolved for bullets 5 and 6 but tracked per This is judged to where exceptions are taken to standard PRA Configuration Control Program. be primarily a methodologies, or where specific modeling Resolved Bullets: documentation assumptions have been made. Specific 1)The factors considered in ASEP have been consideration comments have been provided on some of these considered, for example the time to take the only. in other Observations. Where such exceptions actions. Based on ASEP, no any two pre-and assumptions are used, however, it is initiators are dependent, because no any two Exceptions to important that they be obvious to users of the HFEs are taken within 2 minutes. Surry HRA standard PRA, and that their potential impact on PRA improved the ASEP model. methodologies results be examined, understood, and 2)SPS notebook HR.2 R1 added attachment have been documented. E sensitivity on weighting factors applied to documented. All Examples of methodology exceptions or surveys and concluded current weighting HR-G SRs were modeling assumptions that should be identified factors may get most conservative results. reviewed as assumptions, and evaluated (e.g., via 3)The internal flooding HRA is addressed in subsequently in sensitivity studies) for consideration as "key internal flooding notebook. 2010 focused assumptions" (i.e., those for which use of 4)0perator actions for cooldown and injection peer review and alternate assumptions might result in significantly are discussed in LE notebook. different results) include: 7) Documented per GARD NF-AA-PRA-2052. no findings were

Use of a dependency model different than For HR-G8 (equivalent to HR-G9 2005 stnd) identified defined in ASEP, for the pre-initiator actions assessed as MET, 2010 peer review stated associated with
Approach and assumptions related to the 'Error factors were developed using a this particular interpretation and quantification averaging of the standard, accepted approaches and were F&O finding. Operator Survey results provided for individual HEPs in HR.2 and for
Lack of treatment of internal flooding and dependent HEPs in HR.4.' external events human actions *Applicability (or limitations on applicability) of actions relative to LERF
Basis for assignment of sigma values for the Pc calculations using the HCR model
Basis for extent of cutset review in looking for dependencies among HFEs in cutsets
Basis for assignment of error factors (different than ASEP) for the post-initiator actions Table RAl7: Surry PRA. Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PWROG F-0 3-9 System specific design attributes appear to be Unresolved but tracked per PRA focused 2010 modeled appropriately based on a review of the Configuration Control Program. fault trees. SPS AS.1, AS.2, SC.1, and SC.2 provide the majority of information to properly define the accident sequences.

However, there is not always a specific reference in these notebooks to those procedures used to address the events or to identify the need for additional operator actions. (This F&O oriainated from SR AS-AS) Serial No. 17-016A Docket Nos. 50-280/281 Page 47 of 76 Impact of F&O on Proposed AOT extension This is judged to be a primarily a documentation consideration only. This issue does not affect sequences relevant to this application. Current Peer Peer Review Review Findings PWROG F-0 1-10 focused 2010 Serial No. 17-016A Docket Nos. 50-280/281 Page 48 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension The SPS PRA model is constructed as a linked Unresolved with the exception of item 4, This F&O is fault tree using the CAFTA software. SPS PRA which is considered resolved with the made up of Notebook QU.1 documents the integration following basis: several process followed for the various model elements .. The modeling of the TDAFW pump failure to unresolved items. In general, the integration of the model appears run with a mission time of 24 hours instead of Each of these to account for system dependencies. However, 4 hours is conservative. The basic approach items were the following issues related to linking of the taken for adding different running failure basic integrated model were identified: events with different mission times is that if reviewed and 1. The method of linking the SSIE models into the 24 hour mission time basic event has a assessed for the integrated model is not clear and it is difficult high risk importance, then a new basic event impact on this to trace. with a mission time for the sequence would application.

2. The modeling of loss of RCP seal cooling be developed.

Since the importance of the Cutset and Fault (gate U1-LOSC) was not conditioned to be TDAFW running failure basic events is not Tree review addressed for flooding initiators (i.e., gate U1-significant, then a separate basic event was indicated that LOSC requires input from gate U1-TRANSIENTS not added. accident to make up the AND logic and gate U1-TRANSIENTS does not include flooding initiators sequences as an input). Therefore, flood events that fail one related to this of the sources of RCP Seal Cooling (e.g., F&O are not %FLOOD-AB-SPRAY-U1CCP2AB) are not being significant to this appropriately combined with other random application. failures which could result in loss of all seal cooling. 3. Operator action to isolate a condenser waterbox during maintenance is credited following failure of the isolation valve (e.g., see BE 1CWMOV-FOCW106A combined with REC-FLD-TB-CN-WB in cutsets 379 and 6400 of the SPS MOD A U1-CDF-Avg Maintenance.CUT file). Failure of the isolation valve to close or spurious opening of the valve should be equivalent to failure of the operator action. It appears that the intent was to allow operator *. Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution recovery of level switch failures which would normally close the valve automatically to terminate flooding.

4. The turbine driven AFW pump logic used under gate U1-SGC-BO is based on a 24 hour mission time. This may be somewhat conservative since the turbine driven pump is only credited for 4 hours in SBO. 5. IF.5 details the effects of flood initiating events. Indirect effects due to submergence, spray, etc. seem to be captured.

However, direct effects for many initiators associated with service water expansion joint failures are not captured. For example, IF.5 Table 2.3.4-4 shows that isolable event %FLOOD-TB-SW-XJ-SHD will fail ESGR if not isolated. The fact that CCW HX cooling would be failed, even if the break is isolated, is not shown or apparently modeled. Likewise, the individual CCW HX inlet expansion joint failures are not modeled as failing the associated HX. (This F&O originated from SR QU-A1) Associated SR(s) AS-81 QU-A1 IFQU-A8 IFQU-A9 Serial No. 17-016A Docket Nos. 50-280/281 Page 49 of 76 Impact of F&O on Proposed AOT extension Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PWROG F-0 4-5 Although error factors are provided for individual Unresolved but tracked per PRA focused 2010 HEPs in HR.2 and dependent HEPs in HR.4, Configuration Control Program. there is no discussion in either notebook regarding how the error factors were assigned. PRA guidance document NF-AAA-PRA-101-2052 does provide guidance for an error factor to an individual HPE, but this information is not repeated or referenced in HR.2. (This F&O originated from SR HR-13) Associated SR(s) HR-12 HR-13 Serial No. 17-016A Docket Nos. 50-280/281 Page 50of76 Impact of F&O on Proposed AOT extension The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

1) which multiplies HEPs in the S007 Aa model by 10. This sensitivity demonstrates that this issue does not impact the results of this analysis.

Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PW ROG F-0 3-13 In 2002, an operator survey was completed to Unresolved but tracked per PRA focused 201 O document timing estimates from operators of Configuration Control Program. various experience levels. The timing results from the survey are used in the HRA for the HEPs. Table 6.1 of HR.2 states "the response times for operator actions may be estimated by procedure talk through or operator surveys. Therefore, this is retained as a source of uncertainty." For the SPAR-H HEPs, time available is based on engineering judgment. The delay (TDelay), action (TM) and response times* (T1 /2) are conservative estimates based on a table top review of the procedures as well as input from other HEPs of similar actions and events. (This F&O originated from SR HR-G4) Associated SR(s) HR-G4 PW ROG F-0 3-16 Several documentation issues were identified: Unresolved but per PRA focused 201 O 1. Table 4 of HR.4 contains an error factor for Configuration Control Program. each of the analyzed groupings of dependent operator errors. However, there is no explanation of how the error factor was derived. The equation for deriving the error factor is contained in Attachment.3, HEP Replacements worksheet, in HR.4. 2*. Three HEPs (HEP-C-FWCOND, HEP-C-1AFW, and HEP-C-BAF) related to a loss of feedwater sequence are recalculated in QU.1 based on longer SG dryout times that occur late in the sequence. This also results in the recalculation of four dependencies which contain Serial No. 17-016A Docket Nos. 50-280/281 Page 51 of 76 Impact of F&O on Proposed AOT extension The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

1) which multiplies HEPs in the S007 Aa model by 10. This sensitivity demonstrates that this issue does not impact the results of this analysis.

The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

1) which multiplies HEPs in the S007 Aa model by 10. This sensitivity Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution one or more of the HEPs. The recalculation of these HEPs and the associated dependencies is not discussed or referenced in HR.2. 3. The SPAR-H HEPs recently added to the SPS PRA model are documented in HR.2. Four HEPs

noted in Table A-2 (New HEPs Added) that were evaluated, do not appear in the Fault Tree.

One new HEP listed (HEP-CPORTGENRMP) was not analyzed and is also not in the FT. New HEPs added for the recent model update were not necessarily covered by the 2002 survey results. System Analysis notebooks and review of HR.2 does not indicate that simulator observations or talkthroughs with operators were performed. For the SPAR-H HEPs recently added to the SPS PRA model, the time available to eomplete the actions were not based on applicable generic studies (e.g. thermal I hydraulic analysis or simulations from similar plants). (This F&O originated from SR HR-11) Associated SR(s) HR-11 HR-12 Serial No. 17-016A Docket Nos. 50-280/281 Page 52 of 76 Impact of F&O on Proposed AOT extension demonstrates that this issue does not impact the results of this analysis. Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PW ROG F-0 3-18 SPS GARD NF-AA-PRA-101-2052 states in Unresolved but tracked per PRA focused 201 O Section 3.5, "the SPAR-H model is not Configuration Control Program. recommended where more detailed analysis of diagnosis errors is needed" and references NUREG/CR-1842 for more information. The NU REG states "This approach results in a somewhat 'generic' answer that is sufficient for some of the broad regulatory applications for which SPAR-His intended, but perhaps is insufficient for detailed plant-specific evaluations (a limitation)" and also references NUREG-1792. This NU REG states "detailed assessments of the significant HFE contributors should be performed." (This F&O originated from SR HR-G1) Associated SR(s) HR-G1 PWROG F-0 3-14 SPS HR.2 does not check the consistency of the Unresolved but tracked per PRA focused 2010 post-initiator HEP quantifications. A comparison Configuration Control Program. of previous HEP values with current HEP values is found in the QU.2 notebook supporting files but no relative comparisons are made. (This F&O originated from SR HR-G6) Associated SR(s) HR-G6 Serial No. 17-016A Docket Nos. 50-280/281 Page 53 of 76 Impact of F&O on Proposed AOT extension The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

1) which multiplies HEPs in the S007 Aa model by 10. This sensitivity demonstrates that this issue does not impact the results of this analysis.

The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

1) which multiplies HEPs in the S007 Aa model by 10. This sensitivity Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PWROG F-0 3-11 System Analysis notebooks indicate that "formal Unresolved but tracked per PRA focused 2010 interviews with the plant staff are not Configuration Control Program. documented." There is no documentation to show that the interpretation of the procedures is consistent with plant operational and training practices.

An Operator Survey on HEP timing was performed in 2002 and is the basis for HEP timing. Survey results are not a substitute for face-to-face interviews/discussions or simulations to obtain adequate interpretation of the applicable procedures and sequence of events. New HEPs added for the recent model update using SPAR-H relied primarily on engineering judgment without plant operations and training input to ensure that interpretation of the procedures is consistent with plant observations and training procedures. (This F&O originated from SR HR-E3) ' Serial No. 17-016A Docket Nos. 50-280/281 Page 54of76 Impact of F&O on Proposed AOT extension demonstrates that this issue does not impact the results of this analysis. The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

1) which multiplies HEPs in the S007 Aa model by 10. Additionally information from operator interviews is not expected to impact HEPS by more than an order of magnitude.

This sensitivitv Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding

Dominion Resolution PWROG F-0 3-15 The delay (TOelay), action TM and response Unresolved but tracked per PRA focused 2010 times (T1/2) are conservative estimates based Configuration Control Program. on a table top review of the procedures as well as input from other HEPs of similar actions and events. (This F&O originated from SR HR-G5) Associated SR(s) HR-G5 Serial No. 17-016A Docket Nos. 50-280/281 Page 55 of 76 Impact of F&O on Proposed AOT extension demonstrates that this issue does not impact the results of this analysis.

The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

1) which multiplies HEPs in the S007 Aa model by 10. Additionally information from operator interviews is not expected to impact HEPs by more than an order of Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Summary of Finding , Dominion Resolution Review Findings PWROG F-0 4-7 The development and application of recovery Unresolved but tracked per PRA focused 2010 action HEPs for cutsets in the Surry PRA are Configuration Control Program. discussed in Notebooks HR.3 and HR.10. The development, application and documentation of these recovery HEPs are generally consistent with industry practice.

Not discussed, however, are those recovery HEPs that were added to the cutsets after initial quantification. There is no indication that the post-processing recovery HEPs have been examined to determine whether dependencies exist between them and other HEPs in the respective cutsets. (This F&O originated from SR HR-H3) Associated SR(s) HR-12 HR-H3 Serial No. 17-016A Docket Nos. 50-280/281 Page 56 of 76 Impact of F&O on Proposed AOT extension magnitude. This sensitivity demonstrates that this issue does not impact the results of this analysis. This is judged to be a primarily a documentation consideration only. This issue does not affect sequences relevant to this application. Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PWROG F-0 3-19 The plant's approach to analyzing HEPs is more Unresolved but tracked per PRA focused 2010 involved than the Category I requirements (it is Configuration Control Program. actually closer to Category 111111), but it does not address all of the PSFs identified for the Category 11/111 requirements (a limitation of the SPAR-H method); therefore, MET was selected for Category I. (This F&O originated from SR HR-G3) Associated SR(s) HR-G3 Serial No. 17-016A Docket Nos. 50-280/281 Page 57 of 76 Impact of F&O on Proposed AOT extension The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

1) which multiplies HEPS in the S007 Aa model by 10. This sensitivity demonstrates that this issue does not impact the results of this analysis.

Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PW ROG F-0 2-3 A detailed review of SSIE cutsets identified some Unresolved but tracked per PRA focused 201 O problems: Configuration Control Program. 1. The cutsets do not include all possible combinations, for example: a. Train A CCW pump fails-to-run and Train B CCW pump fails-to-start is in the SSIE cutsets, but other failure events that could lead to Train B CCW pump fails-to-start such as AC failure and actuation signal failure are not included;

b. There is an inconsistency in the modeling of the above combination when compared to the mitigating system fault tree. In the mitigating system fault tree, failure to start of the Pump B is conditioned by its standby failure (see gate 1-CC-P-1 B-FTS). In the SSIE fault tree, the standby status of Pump B is not considered (see gate U1-CC-INIT-AB);

c. Relief valve failure does not appear in the cutsets for the loss of CCW SSIE. 2. Cutsets including both PROB-xxxxxB-STDBY (Train B) and -PROBxxxxxA-STDBY (Train A) events may be underestimating the impact. (fhis F&O originated from SR IE-C10) Associated SR(s) IE-C10 Serial No. 17-016A Docket Nos. 50-280/281 Page 58 of 76 Impact of F&O on Proposed AOT extension Cutset and Fault Tree review indicated that accident sequences related to this F&O are not significant to this application.

Any impact to this application of findings related to initiating events are bounded by the results of Sensitivity

3 which increases LOOP frequencies in the SOO? Aa model by a factor of 10. Therefore, this F&O does not impact the results of this analvsis.

Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PW ROG F-0 3-4 Comparison of IE frequency to industry mean Unresolved but tracked per PRA focused 201 O values is performed in SPS PRA Notebook Part Configuration Control Program. Ill, Volume IE.3, Revision 1, Table 2-4 by comparing 7 modeled Initiating Events with 5 other unit results and to NUREG/CR-5750. The remaining 12 Initiating Events (with Fault Trees) are not compared. Other Initiating Events are also not compared. (This F&O originated from SR IE-C12) Serial No. 17-016A Docket Nos. 50-280/281 Page 59 of 76 Impact of F&O on Proposed AOT extension Comparison with sources and similar plants renders similar results to the I Es compared. Any impact to this application of findings related to initiating events are bounded by the results of Sensitivity

3 which increases LOOP frequencies in the S007 Aa model by a factor of 10.

Current Peer Peer Review Review Findings PWROG F-0 3-5 focused 201 O PW ROG F-0 2-2 focused 2010 Serial No. 17-016A Docket Nos. 50-280/281 Page 60 of 76 Table RAl7: Surry PRA Model Peer Review F&Os Impact of F&O Summary of Finding Dominion Resolution on Proposed AOT extension Recovery events are added to cutsets based on The two recovery actions that the reviewer F&O is resolved. post-processing with QRECOVER and plant-identified as not being in the Surry PRA specific rule file as discussed in SPS HR.3 model but were calculated in the HR.3 model notebook, Section 2.2, and the QU.1 notebook. notebook were removed from the model Some recovery actions (e.g., REC-FTSCC and during the transition from the WinNUPRA REC-FTSBC) should be modeled as HEPs in the model to the CAFT A model. Since the FT so all pertinent cutsets are generated and standby pumps get an auto-start signal if the dependency assessed. running pump fails, these recoveries were (This F&O originated from SR IE-C11) AND'd with the failure of the pressure switch. Since these were not showing up in the cutsets, it was determine that credit for the operator recovery would not be included. If these pressure switch failure basic events had a high importance, then adding the recovery credit would be considered. Some failure modes such as passive failures Unresolved but tracked per PRA Inclusion of these (piping failure, relief valve failure, etc.) are not Configuration Control Program. low probability included in Surry SSIE models. Surry SY:2 passive failure Notebook Table 1 states that passive failure of modes would not piping is assumed to have a negligible probability impact the and is not included in the models. Yet as described in EPRI 1016741, passive failures that accident may be excluded from the post-initiator model scenarios that may be important in the SSIE model. are significant to (This F&O originated from SR IE-C10) this application Associated SR(s) (i.e., LOOP). Any IE-C10 impact to this application of findings related to initiatinq events Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PWROG F-0 3-3 The SPS IE analysis uses the "multiplier method Unresolved but tracked per PRA focused 2010 as stated in SPS IE.3, section 2.2, and Configuration Control Program. references EPRI TR-1013490 to describe the approach used for converting the operating train failure rates into an annual frequency. A later EPRI report, EPRI TR-1016741, provides reasons why this method is not the consensus method in the industry. (This F&O originated from SR IE-C9) Serial No. 17-016A Docket Nos. 50-280/281 Page 61 of 76 Impact of F&O on Proposed AOT extension are bounded by the results of Sensitivity

3 which increases LOOP frequencies in the S007 Aa model by a factor of 10. Therefore, this F&O does not impact the result of this analvsis.

Use of the multiplier method interferes with calculation of importance metrics from the PRA model and impedes visibility and traceability of the model. Cutset and Fault Tree review indicated that accident sequences related to this F&O are not significant to this application. Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PWROG F-0 4-1 Criteria for screening flood areas, sources, Unresolved but tracked per PRA focused 2010 failure mechanisms, and scenarios from further Configuration Control Program. consideration are not clear. Examples include: 1. The next to last paragraph of Section 2.1 in IF.3, Revision 2, states that an area can be screened if it has "no equipment relevant to the PRA and no significant flood sources." It is not clear whether spray is considered a "significant flood source" and whether rooms that posed a potential spray risk were screened.

2. Other criteria that are not listed are being used to screen areas. In Table 5.2-1, the Emergency Diesel Generator Rooms are screened on the basis of their failure not causing a reactor trip or requiring a shutdown.

In the same table, several rooms are screened because the tanks in the room do not have sufficient volume to cause critical flood height. In addition, in several places it is noted that sources are screened because they consist of "small bore lines in this flood area which are too small to have a flood break frequency." It is not clear what constitutes "a line too small to have a break frequency." 3. NF-M-PRA-101-2071, Section 3.2.1 says that "Information regarding the susceptibility of components to failure by spray and other physical phenomena such as jet impingement and pipe whip should be obtained." However, no notes regarding such impacts were found in the walkdown notes for areas screened from detailed analysis (e.g., area FLA5) and in other areas the relationship of flood sources to targets for spray is not defined (e.Q. IF.3 Section 5.2.8 does not Serial No. 17-016A Docket Nos. 50-280/281 Page 62 of 76 Impact of F&O on Proposed AOT extension Cutset and Fault Tree review indicated that accident sequences related to this F&O are not significant to this application. Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution discuss potential spray of the reactor trip breakers by chilled water lines in the room or the proximity of junctions boxes, if any, to flood sources). (This F&O originated from SR IFPP-82) Associated SR(s) IFPP-82 IFSO-A3 IFS0-82 IFSN-A12 IFSN-82 IFEV-82 IFQU-82 Serial No. 17-016A Docket Nos. 50-280/281 Page 63 of 76 Impact of F&O on Proposed AOT extension Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PWROG F-0 1-5 SPS PRA Notebook IF.3 documents the flood Unresolved but tracked per PRA focused 2010 sources but does not provide sufficient Configuration Control Program. information in the following areas: 1. The flood source walkdown is documented in SPS PRA Notebook IF.1, but the walkdown sheets do not contain all requested information. For example, the columns for recording the spatial relationship between flood sources and PRA equipment are not typically completed.

2. IF.3 Tables 5.2-2 through 5.2-10 do not contain all expected information on flood sources. For example, the system pressure and temperature is not included to allow determination of which sources have the potential for pipe whip. In addition, it is not clear whether capacities listed for tanks are related to the tank volume or system volume. 3. Spray/leakage impacts on equipment in the area are not clearly considered for screened sources. 4. Piping connected to tanks such as the chilled

water surge tanks were screened based on the capacity of the tank itself. There appears to have been no consideration of the capacity of attached makeup sources which could exceed the critical capacity. (This F&O originated from SR IFS0-82) Associated SR(s) IFSO-A1 IFSO-A5 IFS0-82 IFSN-A15 Serial No. 17-016A Docket Nos. 50-280/281 Page 64of76 Impact of F&O on Proposed AOT extension Cutset and Fault Tree review indicated that accident sequences related to this F&O are not significant to this application.

Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PW ROG F-0 1-6 There appear to be inconsistencies in the Unresolved but tracked per PRA focused 2010 application of generic failure rates to piping size Configuration Control Program. ranges and some misapplication of information from a previous study for Kewaunee in calculating the flood initiating event frequencies for SPS. For example: 1. The SHP table of the "SPS IF.2 Unit-1 Main Steam Valve House Upper Elevation Piping.xis" spreadsheet uses failure rates of 1.87E-05 for piping of >2" to 6" and 3.47E-05 for piping> 6". However, Reference 6.4.9 of notebook IF.2 indicates that these values are events/year based on the piping lengths for Kewaunee, not failure rates in units of events per Reactor Operating Year-Linear Foot as used in the spreadsheet. In addition, it was noted that it appears that the 1.87E-05/year value in Reference

6.4.9 should

have been 1.87E-04/year. 2. In the "SPS IF.2 Cable Spreading Room.xis" spreadsheet, it is noted that the failure rate applied to 6" fire protection piping is the rate associated with the >6" to 24" size range. In the SPS IF.2 Mechanical Equipment Room 2.xls the failure rate associated with the >4" to 6" range is used for 6" fire protection piping. 3. Piping under 2" has been excluded from the spray frequency calculation. However, Footnote 4 for the table in Section 5 of SPS IF.2 (carried over from EPRI 1013141) notes that for CCW and CST piping, the noted failure rates should be applied to piping under 2". In addition, where EPRI 1013141 does not orovide failure rates Serial No. 17-016A Docket Nos. 50-280/281 Page 65 of 76 Impact of F&O on Proposed AOT extension Cutset and Fault Tree review indicated that accident sequences related to this F&O are not significant to this application. Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution other sources should be considered. (This F&O originated from SR IFEV-AS) Associated SR(s) IF EV-AS PWROG F-0 1-17 Dominion procedure NF-AA-PRA-282 states that Unresolved but tracked per PRA focused 2010 "Based on the results of the sensitivity studies, Configuration Control Program. the analyst should document the insights from the sensitivity analyses. The discussion should also highlight any potential limitations of the use of the PRA model for applications (e.g., as a result of significant sensitivity to particular modeling assumptions, as a result of limitations of the scope or level of detail for the model for certain systems or initiating events, etc.)." No such discussion of model limitations was found in SPS PRA Notebook QU.4. Potential limitations in the quantification process that could impact applications could include such things as: 1. The replacement of independent HEPs by the combination dependent events which may affect the importance measures for the HEPS and evaluation of scenarios in which the failure of one of the replaced events is guaranteed, 2. Assumptions used in the baseline model regarding the probability of equipment being in the standby state which may not be appropriate for all applications, and 3. Limitations of the SPAR-H method of analyzing HEPs. (This F&O originated from SR QU-FS) Associated SR(s) QU-FS Serial No. 17-016A Docket Nos. 50-280/281 Page 66 of 76 Impact of F&O on Proposed AOT extension This is judged to be a primarily a documentation consideration only. This issue does not affect sequences relevant to this application. Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PW ROG F-0 1-16 Significant initiating events, sequences, and Unresolved but tracked per PRA focused 201 O cutsets are documented in SPS PRA Notebook Configuration Control Program. QU.2 Sections 2.3.1, 2.3.2 and 2.3.3. Basic event importance factors are discussed in Section 2.3.4 and included in Attachment

3. However, it is noted that the HEP importance factors are affected by the replacement of the independent HEPs during the quantification process. (This F&O originated from SR QU-D6) Associated SR(s) QU-D6 PWROG F-0 1-8 A global assumption is made that staggered Unresolved but tracked per PRA focused 2012 testing is applicable to all common cause events Configuration Control Program. SPS DA.3 Revision 5, Section 2.2.1, Item 1 ). Typically, some components such as containment isolation valves, HHSI isolation valves, and others may only be tested during the outages. Additional justification for application of the staggered testing assumption to _those components tested on an 18 month basis during outages is needed. -{This F&O originated from SR DA-D5) Associated SR(s) DA-D5 Serial No. 17-016A Docket Nos. 50-280/281 Page 67 of 76 Impact of F&O on AOT extension This is judged to be a primarily a documentation consideration only. This issue does not affect sequences relevant to this application.

Changing CCF events from "staggered basis" to "non-staggered basis" results in an increase in the baseline CDF and LERF values. The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

2) which multiplies CCFs in the S007 Aa model by 10.

Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PWROG F-0 2-9 EPRI generic CCF sources of model uncertainty Unresolved but tracked per PRA focused 2012 are tabulated in Table 1 of the SPS DA.3, Rev. 5 Configuration Control Program. notebook. DA-A-2 notes that component boundaries are not consistent with the failure data, but states that this is a consensus model approach and not a source of uncertainty for Surry. This should be considered a source of model uncertainty and/or be corrected. Missing from the evaluation of sources of model uncertainty are all SPS-specific assumptions, including those tabulated in SPS DA.3 Rev. 5 Section 2.2. (fhis F&O originated from SR DA-E3) Associated SR(s) DA-E3 Serial No. 17-016A Docket Nos. 50-280/281 Page 68 of 76 Impact of F&O on Proposed AOT extension There is no impact on CDF or LERF as this is a documentation enhancement (Ref. PRACC 16425); therefore, this gap has no impact on this application. Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PW ROG F-0 2-3 Regarding component boundaries, Section 3.3.1 Unresolved but tracked per PRA focused 2012 of the CCF GARD (NF-AA-PRA-101-2062, Rev. Configuration Control Program. 4) states, "When defining common cause failure events (and utilizing generic data concerning the probability of these events), the analyst must ensure that the component boundaries assumed for common cause failures are consistent with the boundaries used for the independent failures." DOM.DA.1 Rev. 2 states "To ensure consistency between the generic database and the plant specific database, the component boundary needs to be verified. This notebook documents the generic database with component boundaries defined according to NUREG/CR-6928 (Ref. 6.1). This generic database shall be applicable to all of the Dominion PRA models." However, Assumption 8 in Section 2.2.1 of SPS DA.3 Rev. 5 states "CCF data boundaries were not compared to the boundaries of DOM DA.1. Generic common cause failure factors were used because no plant specific common cause failures were identified. A review of the generic common cause failures indicates that its boundaries were wider than DOM DA.1 boundaries." (This F&O originated from SR DA-D6) Associated SR(s) DA-A2 DA-D6 Serial No. 17-016A Docket Nos. 50-280/281 Page 69 of 76 Impact of F&O on Proposed AOT extension There is no adverse impact on CDF or LERF as a result of this F&O. Leaving this item unresolved adds modeling conservatism; therefore, this gap has no impact on this application. Table RAl7: Surry PRA Model Peer Review F&Os Current Peer

Peer Review Review Findings Summary of Finding Dominion Resolution PWROG F-0 1-2 Scenario 1 in AS.2, Attachment 3, Appendix Unresolved but tracked per PRA focused 2012 ISLOCA F is screened even though the event Configuration Control Program. frequency would be greater than 1.0E-06 (calculated as 3.85E-06).This scenario should be reconsidered to ensure the screening is appropriately justified using the criteria specified in IE-C6. (This F&O originated from SR IE-C6) PWROG F-0 2-2 The issue of ISLOCA flood propagation and Unresolved but tracked per PRA focused 2012 steaming effects in tt)e Safeguards Building is Configuration Control Program. not adequately addressed.

Section 2.4 of the IE.1 notebook states that flooding/spatial effects need not be considered because an unisolated ISLOCA was assumed to go directly to core damage. However, if there is a successful isolation prior to core damage, there is still a question about the effects of the water/steam that was already leaked. For example, AFW pump operation should be shown not to be impacted, as well as potential effects on the credited isolation valve itself. The PRA staff researched the issue during the peer review and provided information that appears to justify the operability of the isolation valve, but additional analysis is required and needs to be documented. (This F&O orii:iinated from SR IE-C3) Serial No. 17-016A Docket Nos. 50-280/281 Page 70 of 76 -Impact of F&O on Proposed ACT extension This is judged to be a primarily a documentation consideration only. This issue does not affect sequences relevant to this aoolication. This is judged to be a primarily a documentation consideration only. This issue does not affect sequences relevant to this application. Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PW ROG F-0 1-6 The SPS PRA model includes mutually exclusive Unresolved but tracked per PRA focused 2012 logic to remove certain CCF combinations. This Configuration Control Program. is supported by a brief description in GARD NF-M-PRA-101-2062, Probabilistic Risk Assessment Procedures and Methods PRA Data Analysis-Common Cause Failure Modeling, Section 3.2.3. This guidance does not seem to be technically supported by NUREG/CR-5485 Section 5.4.4 which only supports removal of combinations of two common cause failure events where the combinations include the same pump (e.g., CCF of Pumps A and 8 in combination with CCF of pumps A and C). Further, NUREG/CR-5485 Section 5.2 notes that NUREG/CR-4780, Volume 1 discusses conditions under which these combinations may be valid (see NUREG/CR-4780, Volume 1, Section 3.3.1 ). Therefore, it may be allowable to remove combinations of Independent Failure of Pump A and CCF of Pumps A and 8 in the same cutset. However, it does not seem valid to remove cutsets involving Independent Failure of Pump A and CCF of Pumps 8 and C. This is especially problematic in the mutually exclusive logic for the diesels where cutsets containing Independent Failure of the MC diesel and common cause failure of two of the emergency diesels. Since there are substantial differences in design between the MC Diesel and the other diesels, it does not seem valid to remove these combinations. (This F&O originated from SR QU-87) Associated SR(s) : QU-87 Serial No. 17-016A Docket Nos. 50-280/281 Page 71 of 76 Impact of F&O on Proposed AOT extension The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

2) which multiplies CCFs in the 8007 Aa model by 10.

.. Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PWROG F&O 1-10 The MC diesel is included in a common cause Unresolved but tracked per PRA focused 2012 group with the other emergency diesel Configuration Control Program. generators even though SPS notebook SY.3.EP states that "The MC diesel has a different manufacturer for the generator and the diesel engine and is unique to both units." SPS OA.3 addresses this in an assumption that states that. "If S80 diesel is modeled as one of the EOG CCF groups, because of the less similarity between the EOG and S80 diesel, the alpha factor of 3 of 3 EOGs CCF to run may be set as 1.06E-2*0.9 = 9.54E-3 and the alpha factor of MC diesel and 2 EOGs CCF to run may be set as 1.06E-2

0.1 = 1.06E-3." However, there is no technical basis for the factor of 10 reduction, only a qualitative discussion, yet this is dispositioned as not being a source of uncertainty. (This F&O originated from SR SY-83) Associated SR(s) OA-06 SY-83 Serial No. 17-016A Docket Nos. 50-280/281 Page 72 of 76 Impact of F&O on Proposed AOT extension The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

2) which multiplies CCFs in the S007 Aa model by 10.

Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution PW ROG F&O 2-5 The CCF grouping appears to have been Unresolved but tracked per PRA focused 2012 performed properly for pumps and some MOVs Configuration Control Program. examined. However, checks of the Electric Power system model and check valves in SI and FW models show CCF combinations that are missing. In the Electric Power system model, the CCF of buses, inverters, breakers and fuel oil pump strainers (possibly other components as well) were modeled for complete failure of all in the group, but not for smaller numbers. For example, Table 3.8-1 shows 1 EETFM-C8-480TFM being comprised of eight transformers. However, failure of a group as small as 2 (e.g., transformer 1 H/1J) could be significant, as these transformers feed the 480V buses that power the 1A/2A and 1 B/2B recirculation spray pumps. While it is acceptable to model CCF of combinations greater than 4 jointly (as is stated in the Section 3.2.2 of the GARD, this means creating a joint probability that sums all the 5/8, 6/8, 7/8 and 8/8 combinations into one), the individual combinations of 2, 3 and 4 still need to be captured. The other logic reviewed that are missing combinations are seen under gates 1-SI-82, 1-Sl-236, 1-FW-27, 1-FW-28, 1-FW-29 and 1-FW-61/1-FW-62. These instances were identified in a short review of the system models, and the review team is concerned the problem is widespread. Another item noted is Section 2.3 of the DOM.DA.3 notebook states "The Supply Breakers that feed the Emergency Buses, if there is a loss of off-site power, should be modeled for a common cause failure to open Serial No. 17-016A Docket Nos. 50-280/281 Page 73 of 76 Impact of F&O on Proposed AOT extension The impact of this gap on this application was evaluated by performing a sensitivity study (Sensitivity

2) which multiplies CCFs in the 8007 Aa model by 10.

Table RAl7: Surry PRA Model Peer Review F&Os Current Peer Peer Review Review Findings Summary of Finding Dominion Resolution when the Emergency Diesel Generators are required to be running and supplying power to the emergency buses." This was not modeled in the EP fault trees (they would be expected under gates 1-EP-8KR-15H8-FTO and 1-EP-8KR-15J8-FTO-LC, etc.). (This F&O originated from SR SY-83) Associated SR(s) SY-83 DA-A1 PWROG F&O 2-8 The DOM.DA.3 R3 notebook Section 2.3 states Unresolved but tracked per PRA focused 2012 that CCF of air-cooled transfonners would not be Configuration Control Program. modeled. There is no mention of this in the EP system notebook. Many of the transformers modeled in the PRA are air-cooled but have CCF modeled. The Surry PRA model would need to be updated to match the assumption in the DOM DA.3 notebook. (This F&O originated from SR Associated SR(s) SY-83 DA-A1 Serial No. 17-016A Docket Nos. 50-280/281 Page 74 of 76 Impact of F&O on Proposed AOT extension There is no adverse impact on CDF or LERF as this is conservative and will be removed from the model; therefore, this gap has no impact on this application. APLA, RA/ #8: Serial No. 17-016A Docket Nos. 50-280/281 Page 75 of 76 The Internal fire Hazard evaluation appears to independently evaluate fire frequency without considering the possible fire induced failures of equipment, cables, and spurious operation of nominally independent equipment in the identified event scenarios.

a. Provide a bounding estimate of the fire induced ICCDP, ICLERP, and subsequent changes in GDF and LERF associated with extending the AOT from 7 to 14 days. Dominion Response:

A bounding estimate for fire risk was generated. The internal events model cutsets related to the ESW pumps were reviewed and four sequences were !dentified that are potentially relevant to fire risk. Four fire scenarios were postulated based on these sequences, which are documented in Table RAl8a1 below. Conditional risk scenarios were constructed by setting the applicable initiating event (unit transient, small LOCA, or loss of Circulating Water) for each scenario to 1. Major equipment relevant to these scenarios was also failed if it could be failed by fire impacting the equipment, cables, or causing spurious equipment operations in the postulated fire scenarios. The internal events model was used to quantify the conditional risk of each scenario. To determine a bounding fire frequency, the total frequency from NUREG/CR-6850, Volume 2 was proportioned equally across the four scenarios, resulting in a frequency for each scenario of 7.5E-2. The risk from all four scenarios was summed to generate the bounding fire risk assessment for the proposed AOT extension. The results are displayed in Table RAl8a2 below and meet the acceptance guidelines for RG 1.177 and RG 1.174. This bounding assessment is conservative because it uses conservative frequencies and does not credit fire suppression or the current configuration which includes low leakage RCP seals. Table RAl8a1: Postulated Fire Scenarios for Bounding Fire Risk Assessment Postulated Fire Initiating Event Additional Failures Modeled Frequency Applied Scenario RCP Seal LOCA Small LOCA 1 of 2 4160V Emergency Buses 7.5E-2 Spurious PORV LOCA Small LOCA 1 of 2 4.160V Emergency Buses 7.SE-2 Transient with no AFW Transient 3 of 3 AFW pumps, 1 of 2 7.SE-2 4160V Emergency Buses Loss of Circ Water LOCW 8 of 8 Circ Water pumps* 7.SE-2 Serial No. 17-016A Docket Nos. 50-280/281 Page 76 of 76 Table RAl8a2: Analysis #2 Results -Bounding Fire Risk Estimate Results U1 U1 ICCDP Single 14 day TS ent 2.12E-10 3.92E-11 4.21 E-15 1.12E-15}}

v t e Letter Sequence Response to RAI
CAC:MF8145, (Approved, Closed)
Initiation Acceptance... Administration Questions Answers 1 March 2017 10 March 2017 Results Approval Other:
MONTHYEAR2017-03-10 [Table View]

ML17066A187: Difference between revisions

Revision as of 05:33, 9 October 2018

Text

Attachment:

6.4.9 should

Navigation menu

ML17066A187: Difference between revisions

Revision as of 05:33, 9 October 2018

Text

Attachment:

6.4.9 should

Navigation menu

Search