ML18096B506

Peach Bottom Atomic Power Station, Units 2 and 3 - Request for Additional Information Exelon Generation Company, LLC Requested Amendment to Adopt 10 CFR 50.69 for Risk-informed Categorization and Treatment of Structures, Systems and Compone

ML18096B506
Person / Time
Site:	Peach Bottom
Issue date:	04/06/2018
From:	Tobin J C Plant Licensing Branch 1
To:	David Helker Exelon Generation Co
Tobin J C, NRR/DORL/LPL1, 415-2328
References
EPID L-2017-LLA-0281
Download: ML18096B506 (15)

Text

From:Tobin, JenniferTo:Helker, David P:(GenCo-Nuc)Cc:Gropp Jr, Richard W:(GenCo-Nuc)

Subject:

Peach Bottom Units 2 and 3 - Request for Additional Information - Adopt 50.69 License Amendment (EPID L-2017-LLA-0281)Date:Friday, April 06, 2018 3:02:00 PMAttachments:APLA Peach Bottom 50-69 Final RAIs.docx

Dear Mr. Helker,

By letter dated August 30, 2017 (Accession No. ML17243A014), as supplemented by letterdated October 24, 2017 (ADAMS Accession No. ML17297B521), Exelon GenerationCompany, LLC requested an amendment to the facility operating license for Peach BottomAtomic Power Station Units 2 and 3 to adopt 10 CFR 50.69 for Risk-informedCategorization and Treatment of Structures, Systems and Components. The Nuclear Regulatory Commission's (NRC) staff is reviewing your submittal and hasdetermined that additional information is needed to complete its review. The specificrequest for additional information (RAI) questions are provided below. The RAI questionswere provided in draft form to you via e-mail on March 21, 2018. The draft questions weresent to ensure that the questions were understandable, the regulatory basis for thequestions was clear, and to determine if the information was previously docketed. Thefinalized questions are both attached and below for your convenience. In a clarification phone call on April 6, 2018, staff discussed the draft RAI questions anddetermined that minor clarifications to RAI-08 would aid in the understanding andresponse. Mr. Gropp confirmed that Exelon would provide a response to the RAI questionswithin 30 days of receipt (by May 7, 2018). If you have any questions, please contact me at (301) 415-2328. A copy of this e-mail willbe made publicly available in ADAMS. Thanks,Jenny -----------------------------------------------------------------------------------------------------------------------------------------------REQUEST FOR ADDITIONAL INFORMATIONAPPLICATION TO ADOPT 10 CFR 50.69 RISK-INFORMED CATEGORIZATION OFSTRUCTURES, SYSTEMS, AND COMPONENTSEXELON GENERATION COMPANYPEACH BOTTOM ATOMIC POWER STATION (PBAPS), UNITS 2 AND 3DOCKET NOS. 50-277 AND 50-278 Title 10, of the Code of Federal Regulations, Part 50, Section 69 (10 CFR 50.69), "Risk-informed categorization and treatment of structures, systems, and components for nuclearpower reactors," allows licensees to use a risk-informed process to categorize systems,structures, and components (SSCs) according to their safety significance in order toremove SSCs of low safety significance from the scope of certain identified special treatment requirements. Regulatory Guide (RG) 1.201, Revision 1, "Guidelines forCategorizing Structures, Systems, and Components in Nuclear Power Plants According totheir Safety Significance" (Agencywide Documents Access and Management System(ADAMS) Accession No. ML061090627) endorses, with regulatory positions andclarifications, the Nuclear Energy Institute (NEI) guidance document NEI 00-04, Revision 0"10 CFR 50.69 SSC Categorization Guideline", (ADAMS Accession No. ML052910035) asone acceptable method for complying with the requirements in 10 CFR 50.69. BothRG 1.201 and NEI 00-04 cite RG 1.200, "An Approach for Determining the TechnicalAdequacy of Probabilistic Risk Assessment Results for Risk-Informed Activities," (revision 1is available at ADAMS Accession No. ML040630078) which endorses industry consensusprobabilistic risk assessment (PRA) standards (i.e., American Society of MechanicalEngineers/American Nuclear Society (ASME/ANS) RA-Sa-2009, "Addenda to ASME/ANSRA-S-2008, Standard for Level 1/Large Early Release Frequency Probabilistic RiskAssessment for Nuclear Power Plant Applications" available at ADAMS Accession No.ML092870592), as the basis against which peer reviews evaluate the technical adequacyof a PRA. Revision 2 of RG 1.200 is available at ADAMS Accession No. ML090410014. By letter dated August 30, 2017 (ADAMS Accession No. ML17243A014), as supplementedby letter dated October 24, 2017 (ADAMS Accession No. ML17297B521), ExelonGeneration Company, LLC (Exelon, the licensee) submitted a license amendment request(LAR) to adopt the regulation in 10 CFR 50.69 at the Peach Bottom Atomic Power Station(PBAPS), Units 2 and 3. Section 3.1.1 of the LAR states that Exelon will implement the riskcategorization process of 10 CFR 50.69 in accordance with NEI 00-04, Revision 0, asendorsed by RG 1.201. However, the licensee's LAR does not contain enough informationfor the Nuclear Regulatory Commission (NRC) staff to determine if the licensee hasimplemented the guidance in NEI 00-04, as endorsed by RG 1.201, appropriately as ameans to demonstrate compliance with all of the requirements in 10 CFR 50.69. Thefollowing requests for additional information (RAIs) outline the specific issues andinformation needed to complete the NRC staff's review: RAI 01 - Scope and Quality of PRA Self-Assessments and Peer Reviews Section 50.69(c)(i) of 10 CFR requires that a licensee's PRA must be of sufficient qualityand level of detail to support the categorization process, and must be subjected to a peerreview process assessed against a standard or set of acceptance criteria that is endorsedby the NRC. Section 3.3 of the LAR states that the licensee conducted a self-assessmentand a full-scope peer review for its internal events PRA (IEPRA) and fire PRA (FPRA) inNovember 2010 and 2011, respectively. However, the licensee did not indicate whatstandard the PRAs were reviewed against or whether internal flooding was included in thescope of review. Address the following: a) Describe the scope and reason for the self-assessments and peer reviewsperformed for the IEPRA and FPRA, and confirm they were performed against theguidance in RG 1.200, Revision 2. b) Confirm that the IEPRA full-scope peer review included internal flooding orotherwise provide the history of peer reviews for the internal flooding PRA. c) Identify which findings presented in Attachment 3 of the LAR were self-assessment findings and which were full-scope peer review findings.

RAI 02 - PRA Upgrades Section 50.69(b)(2)(iii) of 10 CFR requires that the results of the peer review processconducted to meet 10 CFR 50.69 (c)(1)(i) criteria be submitted as part of the LAR. During apeer review, the documentation of differences or deficiencies between the licensee's PRAmodel and the NRC endorsed PRA standard are labeled as facts and observations(F&Os). In the supplement dated October 24, 2017, the licensee stated that the F&Oclosure review team performed an assessment of whether each finding resolutionconstitutes a PRA "upgrade" and identified one upgrade. In addition, the licensee statedthat the independent peer review team did consider whether any other of the findingresolutions should have been considered an upgrade and that the review team concurredwith the assessment that there were no other upgrades. However, the licensee did notdiscuss what the upgrade was or how review of the upgrade was completed. Describe the change identified as an "upgrade," provide a description of the focused-scopepeer review performed on the upgrade, and include all resulting F&Os with completedisposition for the application. RAI 03 - Open F&Os in the Process of Being Resolved Section 50.69(c)(i) of 10 CFR requires that a licensee's PRA must be of sufficient qualityand level of detail to support the categorization process, and must be subjected to a peerreview process assessed against a standard or set of acceptance criteria that is endorsedby the NRC. Section 50.69(b)(2)(iii) of 10 CFR requires that the results of the peer reviewprocess conducted to meet 10 CFR 50.69 (c)(1)(i) criteria be submitted as part of theapplication. During a peer review, the documentation of differences or deficienciesbetween the licensee's PRA model and the NRC endorsed PRA standard are labeled asF&Os. Attachment 3 of the LAR provides F&Os and self-assessment findings that are stillopen or partially resolved following the F&O closure review. The dispositions for several ofthese F&Os do not contain sufficient information to justify the licensee's conclusion that theopen F&O has minimal or no impact on the application. The specific F&Os and requestedinformation are as follows: a) Internal events F&Os 2011 1, 3-4, 3-6, and 5-8 pertaining to modeling of pre-initiators. These F&Os and their corresponding dispositions state that the licenseedid not derive the currently modeled pre-initiators from a formal review ofprocedures and practices at the plant. The licensee stated that a resolution ofthese F&Os would have a minimal impact on this application, and are thereforenot necessary. Provide the following: i. A justification clearly showing why not resolving F&Os 2011 1, 3-4, 3-6,and 5-8 has no impact on the 10 CFR 50.69 categorization results, or ii. A mechanism that ensures a review of the procedures and practices at theplant is conducted, and that any pre-initiators identified from the review areincluded in the PRA models prior to implementing the 10 CFR 50.69categorization process. b) Fire F&O 2012 33 pertaining to adjustments to floor-area-ratios. This F&O and the corresponding disposition states that the licensee did not consider correctlythe treatment of obstructed floor space and consideration of maintenance practicewhen determining the floor-area-ratios (FARs) used to allocate the transientignition frequency across a Physical Analysis Unit (PAU). The F&O indicates thatthe licensee did not remove the area-obstructed locations from the denominator inthe FARs equations. In addition, the disposition indicates that the licensee did notuse maintenance practice to inform the likelihood of transient locations. Thedisposition states that the licensee will resolve the resulting adjustment to thetransient fire scenario frequencies in the current model update and that theresolution has minimal impact on this application. Provide the following: i. A justification clearly showing why not resolving F&O 2012 33 prior toimplementing the 10 CFR 50.69 categorization process has no impact onthe 10 CFR 50.69 categorization results or, ii. A mechanism that ensures the FARs are adjusted as discussed in the F&Oresolution prior to implementing the 10 CFR 50.69 categorization process. c) Fire F&O 2012 40 pertaining to lack of detailed fire modeling (FM). The F&Ocites oversimplified FM in risk significant scenarios. The disposition states that theF&O closure team found that the licensee did perform detailed FM for risksignificant scenarios in the modeling update, with some exceptions. Accordingly,the F&O closure team recommended that the licensee apply the detailed two-pointFM approach to those exceptions as described in NUREG-6850, "Electric PowerResearch Institute (EPRI)/NRC-RES Fire PRA Methodology for Nuclear PowerFacilities," (ADAMS Accession Nos. ML052580075, ML052580118, andML103090242) in which models are developed at the 98th and 75th percentiles. The disposition states that the licensee will perform the recommended two-pointFM and that the resolution has minimal impact on this application. Provide thefollowing: i. A justification clearly showing why not resolving F&O 2012 40 prior toimplementing the 10 CFR 50.69 categorization process has no impact onthe 10 CFR 50.69 categorization results or, ii. A mechanism that ensures the two-point FM is applied to the risk-significantfire scenarios prior to implementing the 10 CFR 50.69 categorizationprocess. d) Fire F&O 2012 17 pertaining to credit for fire wrapping. The disposition to thisF&O states that the FPRA currently credits fire wrap for protecting enclosedcables though there may be locations potentially vulnerable to mechanicaldamage. Accordingly, the disposition states that the licensee will conduct a reviewof potentially vulnerable fire wrap configurations and, if vulnerabilities areconfirmed, then the licensee will remove credit for the fire wrap from the FPRA. Provide the following: i. A justification clearly showing why not resolving F&O 2012 17 prior toimplementing the 10 CFR 50.69 categorization process has no impact onthe 10 CFR 50.69 categorization results or, ii. A mechanism that ensures a review is conducted of potentially vulnerablefire wrap configurations and credit removed from the PRA for any fire wrapsusceptible to mechanical damage prior to implementing the 10 CFR 50.69categorization process. e) Fire F&O 2012 37 pertaining to review of excluded electrical panels. The F&Ostates that the licensee did not consider failure of electrical panels and all targetsterminating at the panel for all non-propagating electrical panel fires that aremodeled in the FPRA. The disposition states that the licensee will confirm that theexcluded panels lead to a single failure, and that the licensee will incorporate anyexcluded panels into the FPRA if the licensee determines that multiple failuresfrom an excluded panel fire are possible. The licensee states that this F&O isbeing resolved in the current model update and has minimal impact on thisapplication. In addition, this resolution does not appear to be consistent with theNRC guidance in NUREG/CR-6850, "Electric Power Research Institute(EPRI)/NRC-RES Fire PRA Methodology for Nuclear Power Facilities," (ADAMSAccession Nos. ML052580075, ML052580118, and ML103090242), for excludingelectrical cabinets from being counted if they are simple wall-mounted panels withless than four switches or they are well-sealed and robustly secured cabinetscontaining circuits below 440 V. Explain how the resolution to F&O 2012 37 is consistent with the guidancecontained in NUREG/CR-6850, or provide a description and justification for theproposed resolution. f) Fire F&O 2012 1 pertaining to resolution of internal events findings. Thedisposition to this F&O states that internal events F&O 6-11 (regarding supportingrequirement (SR) DA-C11) and F&O 3-6 (regarding SR HR-A1 and HR-A2)remained open for the FPRA. The F&O states for internal events F&O FPIE 6-11(mistakenly referred to as F&O 6-1 in PB-PRA-021.01 and the FPIE PR report)that validation is needed for use of maintenance rule unavailability data in thePRA. Internal events F&O 3-6 pertaining to identification of pre-initiators isalready addressed in this question in part (a). The licensee states that F&O 2012 1 has not been resolved but has minimal impact on the 10 CFR 50.69categorization process. i. Confirm that the internal events F&O 6-11 cited in Fire F&O 2012 1 wasclosed by the 2016 F&O Closure process, or alternatively providedescription of the F&O and its resolution or disposition for the10 CFR 50.69 application. ii. Provide a justification clearly showing why not resolving F&O 2012 1 hasno impact on the 10 CFR 50.69 categorization results. RAI 04 - Open/Partially Open Findings Resolved Using Sensitivity Studies Section 50.69(c)(i) of 10 CFR requires that a licensee's PRA must be of sufficient qualityand level of detail to support the categorization process, and must be subjected to a peerreview process assessed against a standard or set of acceptance criteria that is endorsed by the NRC. Section 50.69(b)(2)(iii) of 10 CFR requires that the results of the peer reviewprocess conducted to meet 10 CFR 50.69 (c)(1)(i) criteria be submitted as part of theapplication. During a peer review, the documentation of differences or deficienciesbetween the licensee's PRA model and the NRC endorsed PRA standard are labeled asF&Os. The resolutions to a number of F&Os presented in Attachment 3 of the LAR involvesensitivity studies that the licensee will perform to validate the treatments used in the PRA. The dispositions for these cases do not explain how the licensee will resolve the F&Os ifthe sensitivity study results do not validate that the F&O has only a minimal impact on theapplication. The specific F&Os and requested information are as follows: a) Fire F&O 2012 6 pertaining to the uncertainty of FM parameters. The dispositionfor this F&O states that to resolve this F&O the licensee will evaluate FM parameteruncertainties for risk significant scenarios and that the licensee will performsensitivity studies for parameters identified to be key sources of uncertainty. Thelicensee states that this F&O is being resolved in the current model update and hasminimal impact on this application. Provide the results of the sensitivity studies onthe FM parameters to demonstrate no impact on the 10 CFR 50.69 categorizationresults, or if any impact is identified, describe the actions that will be taken toaddress it. b) Fire F&O 2016 1 pertaining to the assumption that control cables would be thecritical target cable in most scenarios. The disposition states that to resolve theF&O the licensee will use the weighted average of cable parameters in a PAU forthermally-induced electrical failure (THIEF) calculations (in some cases specifictarget parameters were used). The disposition states that the licensee is performingsensitivity studies by using the upper and lower bounds of these cable parametersas input to THIEF (for calculating critical time to cable insulation failure) to compareto the results of running THIEF using the mean parameters. The licensee statesthat this F&O is being resolved in the current model update and has minimal impacton this application. Provide the results of the sensitivity studies on the cableparameters to demonstrate no impact on the 10 CFR 50.69 categorization results,or if any impact is identified, describe the actions that will be taken to address it. c) Fire F&O 2016 2 pertaining to the radial increments used in THIEF. Thedisposition to this F&O states that to resolve this F&O the licensee will perform asensitivity study on the "radial increment" used in the THIEF modeling, which islarger than the radial increment recommended in NUREG/CR-6931, Volume 3,"Cable Response to Live Fire (CAROLFIRE) Volume 3: Thermally-Induced ElectricalFailure (THIEF) Model," (ADAMS Accession No. ML081190261) to demonstrate thatthe increment used yields a reasonable but conservative result. In addition, thelicensee states that this F&O is being resolved in the current model update and hasminimal impact on this application. Provide the results of the sensitivity studies onthe radial increments to demonstrate no impact on the 10 CFR 50.69 categorizationresults, or if any impact is identified, describe the actions that will be taken toaddress it. d) Fire F&O 2016 6 pertaining to the assumed flame spread over horizontal cabletrays (FLASH-CAT) and THIEF modeling parameters. The disposition to this F&Ostates that the licensee will perform sensitivity studies that vary the parameters usedin FLASH-CAT and THIEF between reasonable upper and lower bounds in order to determine the impact on analysis. The licensee states that this F&O is beingresolved in the current model update and has minimal impact on this application. Provide the results of the sensitivity studies on the fire modeling parameters (bothFLASH-CAT and THIEF) to demonstrate no impact on the 10 CFR 50.69categorization results, or if any impact is identified, describe the actions that will betaken to address it. RAI 05 - Open F&Os with Incomplete Dispositions Section 50.69(c)(i) of 10 CFR requires that a licensee's PRA must be of sufficient qualityand level of detail to support the categorization process, and must be subjected to a peerreview process assessed against a standard or set of acceptance criteria that is endorsedby the NRC. Section 50.69(b)(2)(iii) of 10 CFR requires that the results of the peer reviewprocess conducted to meet 10 CFR 50.69 (c)(1)(i) criteria be submitted as part of theapplication. During a peer review, the documentation of differences or deficienciesbetween the licensee's PRA model and the NRC endorsed PRA standard are labeled asF&Os. The following dispositions for open or partially resolved F&Os presented inAttachment 3 of the LAR seemed incomplete and require further justification and/orinformation: a) Undesired Operator Actions in FPRA The dispositions for F&Os 2012 6 and 2012 7, pertaining to undesired operatoractions, state that the licensee will perform a procedure-by-procedure review tojustify why consideration of instrumentation that could potentially mislead operatorsor cause them to perform a harmful action did not need to be modelled in theFPRA. The licensee states that if a justification cannot be provided, then thelicensee will incorporate the undesired operator actions into the PRA models asapplicable. However, the disposition does not explain how the licensee willincorporate these undesired operator actions, which could propagate errors ofcommission and errors of omission into the PRA model. In addition, the dispositiondoes not indicate whether this update would meet the criteria for a PRA "upgrade"as defined in the PRA Standard (i.e., American Society of MechanicalEngineers/American Nuclear Society (ASME/ANS) RA-Sa-2009, "Addenda toASME/ANS RA-S-2008, Standard for Level 1/Large Early Release FrequencyProbabilistic Risk Assessment for Nuclear Power Plant Applications"), as endorsedby RG 1.200. Provide the following: i. A detailed discussion describing how the licensee will incorporate anyapplicable undesired operator actions into the PRA model. The discussionshould include a description of the modeling of failures to respond tofire-induced damage of PRA modeled components (i.e., failure ofomission), and undesired actions that could to lead to undesired equipmentactuations (i.e., errors of commission). ii. A discussion of whether the PRA model update to include any applicableundesired operator actions constitutes a PRA "upgrade" as defined in thePRA Standard (i.e., ASME/ANS RA-Sa-2009), as endorsed by RG 1.200. If the update is considered a PRA "upgrade", then propose a mechanism that ensures a focused-scope peer review of the upgrade is performed andall resulting F&Os are resolved prior to implementing the 10 CFR 50.69categorization process. b) Uncoordinated Breakers in FPRA The disposition for F&O 2012 1 states that circuits modeled in the FPRAdetermined to be uncoordinated will be modelled as "non-coordinated" in the FPRA. However, the disposition does not describe how the licensee will model the non-coordinated circuits. Failures caused by these inadequate circuits could represent anew kind of failure not currently modeled in the FPRA. The disposition does notindicate whether this update would meet the criteria for a PRA "upgrade" as definedin the PRA Standard (i.e., ASME/ANS RA-Sa-2009), as endorsed in RG 1.200. Provide the following: i. A discussion of the non-coordinated circuits and the extent of the modelingnecessary to address this F&O. ii. A detailed discussion describing how the licensee will incorporate anynon-coordinated circuits in the PRA model. The discussion should include adescription of any modeled component failures associated with theuncoordinated circuit, including failures that could occur upstream anddownstream of possible fault locations along the circuit. iii. A discussion of whether the PRA model update to include non-coordinatedcircuits constitutes a PRA "upgrade" as defined in the PRA Standard (i.e.,ASME/ANS RA-Sa-2009), as endorsed by RG 1.200. If the update isconsidered a PRA "upgrade", then provide a mechanism that ensures afocused-scope peer review of the upgrade is performed and all resultingF&Os are resolved prior to implementing the 10 CFR 50.69 categorizationprocess. RAI 06 - Open/Partially Open F&Os Associated with Fire Modeling Codes Section 50.69(c)(i) of 10 CFR requires that a licensee's PRA must be of sufficient qualityand level of detail to support the categorization process, and must be subjected to a peerreview process assessed against a standard or set of acceptance criteria that is endorsedby the NRC. Section 50.69(b)(2)(iii) of 10 CFR requires that the results of the peer reviewprocess conducted to meet 10 CFR 50.69 (c)(1)(i) criteria be submitted as part of theapplication. During a peer review, the documentation of differences or deficienciesbetween the licensee's PRA model and the NRC endorsed PRA standard are labeled asF&Os. The dispositions to F&Os 2016 3, 2016 4, and 2016 9, indicate that thefindings will be resolved by showing that the THIEF, FLASH-CAT, and Time-to-Automatic-Detection models and related model parameters have been applied within their knownlimits of applicability. Provide demonstration that the THIEF, FLASH-CAT, and Time-to-Automatic-Detection models and related model parameters have been used within knownlimits of their applicability. For each of the F&Os 2016 3, 2016 4, and 2016 9,provide a complete description of the F&O, and a discussion of its resolution. RAI 07 - Implementation Items Section 50.69(b)(2)(ii) requires that a licensee's application contain a description of themeasures taken to assure that the quality and level of detail of the systematic processesthat evaluate the plant for internal and external events during normal operation, low power,and shutdown are adequate for the categorization of SSCs. Attachment 6 of the LARpresents dispositions for assumptions and modeling uncertainties that include plannedupdates to the PRA models after the NRC approves the 10 CFR 50.69 implementationamendment request but before the licensee implements its 10 CFR 50.69 categorizationprogram. These updates include updating the internal flooding pipe break frequencies, andremoving credit for core melt arrest in vessel. However, the licensee did not discusswhether these future updates can have an impact on the 10 CFR 50.69 categorizationprocess. a) Justify why the updates included in Attachment 6 of the LAR (i.e., updating theinternal flooding pipe break frequencies, and removing credit for core melt arrest invessel) have no impact on the 10 CFR 50.69 categorization results or include themin response to item b. below. b) Provide a list of activities and PRA modeling changes including any items that willnot be completed prior to issuing the amendment but must be completed prior toimplementing the 10 CFR 50.69 categorization process. Propose a mechanism thatensures these activities and changes will be completed and appropriately reviewedand any issues resolved prior to implementing the 10 CFR 50.69 categorizationprocess (for example, a license condition that includes all applicable implementationitems and a statement that they will be completed prior to implementation of the10 CFR 50.69 categorization process). RAI 08 - Key Assumptions and Uncertainties that could Impact the Application Sections 50.69(c)(1)(i) and (ii) require that a licensee's PRA be of sufficient quality andlevel of detail to support the SSC categorization process, and that all aspects of theintegrated, systematic process used to characterize SSC importance must reasonablyreflect the current plant configuration and operating practices, and applicable plant andindustry operational experience. The dispositions presented in Attachment 6 of the LAR forkey assumptions and modeling uncertainties state in each case that they do not represent akey source of uncertainty and will not be an issue for the 10 CFR 50.69 categorization. However, in a number of instances the licensee did not provide sufficient information forNRC to conclude that the uncertainty will not impact the 10 CFR 50.69 categorizationresults. The specific F&Os and requested information are as follows: a) Diesel Generator Cooling Fan Success Criteria Attachment 6 (page 71) of the LAR states that the outdoor air temperature at theplant does not often exceed the design basis temperature for the diesel generators(DG) (i.e., >80° F) which would require two DG cooling fans. The licensee statesthat it assumes just one-of-two DG cooling fans to be an adequate success criterionin the PRA for the entire year. Based on publically available information, theaverage high temperatures at the plant location appears to be above 80° F in thesummer months. Provide the following:

i. A basis for the adequacy of the assumed one-out-of-two DG cooling fanssuccess criteria in the summer months, or ii. Justify quantitatively that the 10 CFR 50.69 categorization results are notsensitive to the assumption that one-of-two DG cooling fans provideadequate cooling when the temperature at the plant exceeds the designbasis temperature(i.e., >80° F), or iii. A mechanism that ensures the model is adjusted to account for highersummer temperatures prior to implementing the 10 CFR 50.69 categorizationprocess. b) Low Pressure Injection after Core Damage Attachment 6 (page 74) of the LAR indicates that the licensee takes credit in thePRA for "timely" low pressure injection after core damage to avoid a large earlyrelease. The disposition states that this assumption provides a reasonable best-estimate approach. However, the licensee does not provide a basis for theassumption. Provide the following: i. A discussion which describes the thermohydraulic basis or rationalesupporting the validity of the assumption that injection of low pressurecooling after core damage can prevent a large early release. The discussionshould include a description of the term "timely low pressure injection." ii. Alternatively, justify quantitatively that the 10 CFR 50.69 categorizationresults are not sensitive to this assumption. c) Safety Relief Valve (SRV) Success Criteria after Passing Liquid Attachment 6 (page 75) of the LAR states that although the SRVs are designed topass water, and 10 CFR Part 50, Appendix R, "Fire protection program for nuclearpower facilities operating prior to January 1, 1979," models the RPV being floodedwith water returning to the suppression pool via the SRVs, the SRVs are nevertested in this fashion. The disposition also states that the licensee assigns anominal failure probability to the SRVs for failure to open following flooding of thesteam lines that causes liquid to pass through the SRVs. However, the NRC staffnotes that the NUREG/CR-6928, "Industry Average Performance for Componentsand Initiating Events at U.S. Commercial Nuclear Power Plants," (ADAMSAccession No. ML070650650) Component Reliability 2010 update provides aprobability estimate for an SRV to fail to close (passing liquid) of 0.1. Provide thefollowing: i. Confirm that the nominal failure probability used for the SRVs to fail to open isconsistent with the 2010 update of NUREG/CR-6928 or justify the nominalfailure probability used, including a justification for assuming the SRVs willwork absent of testing. ii. Given that the licensee has identified a scenario where SRVs couldpotentially pass liquid following steam-line flooding, discuss how the licensee addresses the failure of the SRVs to close after passing liquid. Thediscussion should include consequences of the failure, how/if the failure ismodeled, and a justification for the failure to close probabilities used,including a justification for the assumption that the SRVs will work absent oftesting, or iii. As an alternative to ii., justify, quantitatively, that the 10 CFR 50.69categorization results are not sensitive to the assumption or SRV failuremodes addressed in items i and ii above. d) Turbine success while passing liquid Attachment 6 (page 76) of the LAR states that a steam turbine similar to the reactorcore isolation cooling (RCIC) and high-pressure coolant injection (HPCI) turbineswas tested with a slug of water and found to run successfully. The LAR states thatbased on this information, the licensee assigned a nominal failure to provide a slightconservative bias for this failure mode. However, the licensee does not provide ajustification showing why a single test on another turbine is sufficient to support theassumption that RCIC and HPCI turbines may continue running after passingliquid. In addition, the licensee does not provide information describing whatnominal failure probability value the licensee assigned to this failure mode and whythe licensee considered the nominal probability to be conservative. Provide thefollowing: i. A discussion describing whether the RCIC and HPCI turbines were designedto continue running after passing liquid. If the RCIC and HPCI turbines weredesigned to continue running while and after passing liquid, then discusswhether these conditions encompass the conditions that would occur in theaccident scenarios for which these turbines were credited (which appears toinclude instances in which there is continuous flow of water for the scenario). ii. A justification for the nominal failure probability assigned to the failure of theRCIC and HPCI turbines to run after passing liquid, or iii. As an alternative to ii., if the RCIC and HPCI turbines were not designed tocontinue running while and after passing liquid for the conditions that wouldoccur in the accident scenarios they were credited for in the PRA,quantitatively justify that the 10 CFR 50.69 categorization results are notsensitive to this assumption. e) Loss of Net Positive Suction Head (NPSH) Attachment 6 (page 77) of the LAR states that loss of NPSH during certain events(e.g. loss-of-coolant accidents (LOCAs)) leads to the loss of the suppression poolas a cooling inventory source. The LAR explains, however, that this inventory isonly needed if the residual heat removal (RHR) system cross-tie fails. The LARalso explains that the licensee assigned a nominal failure probability to provide aslight conservative bias. However, the licensee does not provide informationdescribing the RHR cross-tie or the specific failure mode of the cross-tie. Inaddition, the licensee does not provide information describing what nominal failure probability value is assigned to this failure mode and why the licensee consideredthe nominal probability to be conservative. Provide the following: i. A discussion describing how the RHR is cross-tied to provide a cooling optiongiven loss of the suppression pool as a source of cooling inventory. Identifythe specific failure mode that was assigned a nominal failure probability. ii. A justification for the nominal failure probability value, and why this value isvalid for this failure mode, or iii. As an alternative to ii., justify, quantitatively, that the 10 CFR 50.69categorization results are not sensitive to this assumption. f) Low Intake Pond Level Attachment 6 (page 78) of the LAR states that in the unlikely event of low intakepond level, risk can be averted by reducing power levels prior to a plant trip or bytripping the circulating water pumps following a plant trip. The LAR states thatnominal failure probabilities are assigned to derive the overall likelihood that theprecursor events (based on plant experience) proceed to a totally unrecoverabletype of event. Provide the following: i. Description and justification on how the nominal failure probabilities arederived, and the specific failure mode associated with the assignedprobabilities. ii. Confirmation and description of the two operator actions cited above areproceduralized steps required to be taken when the intake pond level is lowenough to fail this inventory source, or iii. As an alternative to ii., justify, quantitatively, that the 10 CFR 50.69categorization results are not sensitive to this assumption. g) HEP Dependency Analysis Attachment 6 (page 80) of the LAR identifies dependent Human Error Probability(HEP) values as a source of uncertainty and cites the sensitivity study on HEPsperformed in accordance with Section 5 of NEI 00-04 as the disposition. Clarifyhow the sensitivity studies performed for HEPs will address HEP dependencyuncertainty. h) Pipe Rupture due to Water Hammer Attachment 6 (page 82) of the LAR states that there is uncertainty associated withthe success of RHR when the reactor is in suppression pool cooling mode becauseof the potential for a pipe rupture following a water hammer. The LAR states thatthe water hammer events and values utilized provide a reasonable approach. However the licensee does not describe nor justify the approach. Provide thefollowing:

i. A description of and a basis for the approach used to model the potential forpipe rupture in the RHR system following a water hammer while insuppression pool cooling mode or, ii. Justify, quantitatively, that the 10 CFR 50.69 categorization results are notsensitive to this treatment. RAI 09 - Overall Categorization Process Section 50.69(b)(2)(i) requires that a licensee's application contain a description of theprocess for SSC categorization. The discussion in Section 3.1.1 of the LAR does notprovide enough information for the NRC staff to clearly understand the sequence ofevaluations in the categorization process, what information the licensee will develop anduse, and what guidance the licensee will follow for the Integrated Decision-making Panel(IDP) decision-making process during the categorization of each system. Information onthe training and expertise of the IDP team is provided in the LAR and need not be repeatedin response to this RAI. Provide the following: a) Summarize, in the order they will be performed, the sequence of elements orsteps that the licensee will follow for each system that will be categorized. A flowchart, such as that provided in the September 6, 2017 public meeting with NuclearEnergy Institute (NEI) (ADAMS Accession No. ML17249A072) regarding10 CFR 50.69 License Amendment Requests may be provided instead of adescription. The public meeting summary is available at ADAMS Accession No.ML17265A020. The steps should include: i. The input from all PRA evaluations such as use of the results from the IEPRA,internal flooding, and FPRAs; ii. The input from non-PRA approaches (seismic, other external events, andshutdown); iii. The input from the responses to the seven qualitative questions in Section 9.2of NEI 00-04; iv. The input from the defense-in-depth (DID) matrix; v. The input from the passive categorization methodology. b) In description to item (a) above, please clarify the difference between "preliminaryhigh-safety significance (HSS)" and "assigned HSS" and identify which inputs can,and which cannot, be changed from preliminary HSS to low-safety significance(LSS) by the IDP. Confirm that the proposed approach is consistent with theguidance in NEI 00-04, as endorsed by RG 1.201.c) In description to item (a) above, please clarify which steps of the process areperformed at the function level and which steps are performed at the componentlevel. Describe how the categorization of the component impacts thecategorization of the function, and vice-versa. Describe instances in which thefinal safety significance of the function would differ from the safety significance of the component(s) that support the function, and confirm that the proposedapproach is consistent with the guidance in NEI 00-04 as endorsed by RG 1.201.d) Section 7 of NEI 00-04 states that if any SSC is safety significant, from either thePRA-based component safety significance assessment (Section 5) or the DIDassessment (Section 6), then the associated system function is of preliminarysafety significance. The cited guidance applies to all aspects identified inSections 5 and 6 of NEI 00-04, including Section 5.3 through 5.5 dedicated toseismic, external hazards, and shutdown risk. If the licensee's categorizationprocess differs from the guidance in Section 7 of NEI 00-04 cited above, wherefunctions supported by any HSS component(s) will be assigned HSS, describeand justify the approach. e) The industry flow chart presented by NEI at the September 6, 2017 public meetingshows that the passive categorization would be undertaken separately from theactive categorization. i. Explain how the results from the passive categorization will be integratedwith the overall categorization results. ii. If the results from the passive categorization can be changed by the IDP,explain and justify the proposed approach.RAI 10 - SSCs Categorization based on External Hazards Sections 50.69(c)(1)(i) and (ii) require that a licensee's SSC categorization processconsider results and insights from a plant-specific PRA that is of sufficient quality and levelof detail to support the SSC categorization process, as well as determine SSC functionalimportance using an integrated, systematic process for addressing initiating events (internaland external), SSCs, and plant operating modes, including those not modeled in the plant-specific PRA. The guidance in NEI 00-04 provides one acceptable method for includingexternal events in the categorization of each SSC. The information provided in the LAR isnot sufficient for the NRC to determine if the licensee's proposed categorization process isconsistent with the guidance in NEI 00-04. The specific hazards and requested informationare as follows: a) "Other" External Hazards Section 3.2.4 of the LAR states that the categorization process will use screeningresults from the Individual Plant Evaluation of External Events (IPEEE), performedin response to GL 88-20, "Individual Plant Examination for Severe AccidentVulnerabilities," for evaluation of safety significance related to the other externalhazards and that all SSCs credited in other IPEEE external hazards are consideredHSS. The use of "other" instead of a more precise description does not allow theNRC staff to compare the license's proposed process with the guidance inNEI 00-04. Provide the following: i. Identify the external hazards that will be evaluated according to the flow chartin Figure 5-6 of NEI 00-04. ii. Identify which hazards will have all SSCs credited and considered HSSinstead of using the flow chart.

iii. A description of, and justification for, any additional method(s) different from(i) or (ii) above that the licensee will use to evaluate individual SSCs against"other" external hazards and identify the hazards that the licensee willevaluate with these methods. iv. Confirm that all hazards not included in the categorization process inresponse to (i), (ii), or (iii) above will be considered insignificant for everySSC and therefore will not be considered during the categorization process. b) External Flooding Figure 5-6 of NEI 00-04 shows that if a component is included in a screenedscenario, then in order for that component to be considered a candidate LSS, thelicensee has to show that if the component was removed, the screened scenariowould not become unscreened. Attachment 4 of the LAR indicates that the licenseescreens external flooding hazards out of the 10 CFR 50.69 process. The LARstates that flooding from rivers and streams (precipitation based), and dam failureare bounded by the current licensing basis (CLB) and do not pose a challenge to theplant, and that flooding from local intense precipitation is not bounded by the currentlicensing basis. However, the licensee does not provide any information describingSSCs, if any, that are included in the screening. Provide the following: i. A discussion identifying SSCs, if any, that are credited in the screening ofexternal flooding, including passive and/or active components ii. A discussion explaining how the licensee will apply the guidance in Figure 5-6of NEI 00-04 to any SSCs that are credited for screening of external flooding. c) High Winds Figure 5-6 of NEI 00-04 shows that if a component is included in a screenedscenario, then in order for that component to be considered a candidate LSS, thelicensee has to show that if the component was removed, the screened scenariowould not become unscreened. Attachment 4 of the LAR states that the licenseescreens out the extreme wind or tornado hazard on the basis that the frequency ofdamage to the exposed components is estimated to be less than 1E-6/year. However, the licensee does not provide any information describing the SSCs, if any,that are included in the screening. Provide the following: i. A discussion identifying the SSCs, if any, that are credited in the screening ofextreme wind and tornados, including passive and/or active components. ii. A discussion explaining how the licensee will apply the guidance in Figure 5-6of NEI 00-04 to any SSCs that are credited for screening of extreme windand tornados. iii. Explain how the discussion in items (i) and (ii) above would be impacted bythe current effort to assess tornado missile protection hazard in response toRIS 2015-06 "Tornado Missile Protection," (ADAMS Accession No.ML15020A419).

RAI 11 - Shutdown Risk Section 50.69(c)(1)(ii) requires that a licensee's SSC categorization process determineSSC functional importance using an integrated, systematic process for addressing initiatingevents (internal and external), SSCs, and plant operating modes, including those notmodeled in the plant-specific PRA. Section 3.2.5 of the LAR states that the 10 CFR 50.69categorization process will use the shutdown safety management plan described inNUMARC 91-06, "Guidelines for Industry Actions to Assess Shutdown Management,"(ADAMS Accession No. ML14365A203), for categorization of safety significance related tolow power and shutdown conditions. However, the LAR does not cite the othercategorization criteria listed in NEI 00-04 Section 5.5 pertaining to shutdown events (e.g.,DID attributes, failures that would initiate a shutdown event). Clarify and provide a basis forhow the categorization of SSCs will be performed for shutdown events, and how it isconsistent with the guidance in NEI 00-04, as endorsed by RG 1.201. RAI 12 - Qualitative Assessments When classifying SSCs according to risk significance, 10 CFR 50.69(c)(1)(ii) states that alicensee must determine SSC functional importance using an integrated, systematicprocess for addressing initiating events (internal and external), SSCs, and plant operatingmodes, including those not modeled in the plant-specific PRA. The functions to beidentified and considered include design bases functions and functions credited formitigation and prevention of severe accidents. The systematic process outlined inNEI 00-04 includes qualitative assessment questions to assist the IDP in determiningwhether functions/SSCs identified as candidate LSS that are implicitly depended upon tomaintain safe shutdown capability, prevention of core damage and maintenance ofcontainment integrity. The staff notes that, unlike the DID assessment outlined in Section 6of NEI 00-04, the qualitative questions for the IDP in Section 9 are not limited to design-basis events. To address the requirements in 10 CFR 50.69(c)(1)(ii), describe and justify, with anexample, such as FLEX functions, how the severe accident functions will be consideredduring the categorization process, including discussion on how each of the sevenqualitative questions in Section 9 of NEI 00-04 would be applied to those functions. RAI 13 - Interfacing SSCs On January 31, 2018, the NRC staff observed the deliberations of the IDP for the RHRSystem. The staff observed that the RHR categorization process appeared to categorizeRHR functions/SSCs that interfaced with other systems that have not yet been categorized,such as the Automatic Depressurization System (ADS) and Core Spray System. Theguidance in Section 4 of NEI 00-04 states that a candidate LSS SSC that supports aninterfacing system should remain uncategorized until all interfacing systems arecategorized. a. Confirm that the guidance in NEI 00-04 will be followed and that any functions/SSCsthat serve as an interface between two or more systems will not be categorized untilthe categorization for all of the systems that they support is completed, or otherwisejustify your proposed approach.

b. Confirm that the item in a. above will be captured in the 50.69 categorizationprocedures. RAI 14 - F&O Closeout Process Section 50.69(b)(2)(iii) of 10 CFR requires that the results of the peer review processconducted to meet 10 CFR 50.69 (c)(1)(i) criteria be submitted as part of the LAR. On January 31 and February 1, 2018, the NRC staff performed an audit of the 50.69categorization process, including a review of the revised F&O Closure Review Reportand the revised PRA Technical Adequacy Evaluation Report. The staff noticedseveral inconsistencies between the information provided in the revised F&O closurereport and the revised PRA technical adequacy evaluation, as compared to theinformation provided in the LAR and LAR supplement. Examples of thediscrepancies include1) Peer review F&Os not included in the LAR that were notaddressed in the F&O closure report, and 2) open F&Os listed in the F&O closurereport that were not provided in the LAR. Provide the following: a. An explanation of how the F&Os were selected for the F&O closure process andhow/why that differs from those provided in the LAR. b. Because the scope of the 10 CFR 50.69 categorization process is so broad, all openF&Os may potentially be relevant to this LAR and could have an impact on the10 CFR 50.69 categorization results. Provide any outstanding F&Os that have notbeen closed by the F&O closure review and have not been provided in the LAR. Foreach F&O include: i. The complete F&O description, the peer review team recommendation, thelicensee's resolution, or disposition for the application, and the F&O closureteam assessment, if applicable. ii. The licensee's documented justification supporting the classification of eachF&O finding resolution for open F&Os as either a PRA "upgrade" or PRA"maintenance update", as defined in the ASME/ANS RA-Sa-2009 PRAStandard endorsed by RG 1.200, Revision 2. RAI 15 - Passive Component Categorization LAR Section 3.1.2 stated that for the categorization of passive components and the passivefunction of active components, the licensee will use the method for risk-informedrepair/replacement activities consistent with the safety evaluation issued by the Office ofNuclear Reactor Regulation, "Request for Alternative ANO2-R&R-004, Revision 1, Requestto Use Risk-informed Safety Classification and Treatment for Repair/Replacement Activitiesin Class 2 and 3 Moderate and High Energy Systems, Third and Fourth 10-Year InserviceInspection Intervals," for Arkansas Nuclear One, Unit 2, dated April 22, 2009 (ADAMSAccession No. ML090930246). The LAR further states that this methodology will be applied to determine the safety significance of Class 1 SSCs. The NRC staff notes that this methodology has been approved for Class 2 and Class 3SSCs. Because Class 1 SSCs constitute principal fission product barriers as part of thereactor coolant system or containment, the consequence of pressure boundary failure forClass 1 SSCs may be different than for Class 2 and Class 3, and therefore the criteria inthe ANO-2 methodology cannot automatically be generalized to Class 1 SSCs withoutfurther justification. The LAR does not justify how the ANO-2 methodology can be applied to Class 1 SSCs andhow sufficient DID and safety margins are maintained. A technical justification for Class 1SSCs should address how the methodology is sufficiently robust to assess the safetysignificance of Class 1 SSCs, including, but not limited to: justification of theappropriateness of the conditional core damage probability (CCDP) numerical criteria usedto assign 'High', 'Medium' and 'Low' safety significance to these loss of coolant initiatingevents; identification and justification of the adequacy of the additional qualitativeconsiderations to assign 'Medium' safety significance (based on the CCDP) to 'High' safetysignificance; justification for crediting operator actions for success and failure of pressureboundary; guidelines and justification for selecting the appropriate break size (e.g. doubleended guillotine break or smaller break); and include supporting examples of types of Class1 SSCs that would be assigned low safety significance, etc. As mentioned in the meeting summary from the February 20, 2018 Risk-Informed SteeringCommittee (RISC) meeting (ADAMS Accession No. ML18072A301), the NRC staffunderstands that the industry is planning to limit the scope to Class 2 and Class 3 SSCs,consistent with the pilot Vogtle license amendment (ADAMS Accession No.ML14237A034). Please provide the requested technical justification or confirm the intent to apply the ANO-2passive categorization methodology only to Class 2 and Class 3 equipment.----------------------------------------------------------------------------------------------------------------------------------------------- Jenny TobinProject ManagerNRR/DORL/LPL-1Office O9-C12 Phone 301-415-2328