ML13224A246: Difference between revisions
StriderTol (talk | contribs) (Created page by program invented by StriderTol) |
StriderTol (talk | contribs) (Created page by program invented by StriderTol) |
||
Line 18: | Line 18: | ||
=Text= | =Text= | ||
{{#Wiki_filter:INDIANAMICHIGANPOWERA unit of American Electric PowerAugust 2, 2013Docket Nos.: 50-31550-316Indiana Michigan PowerCook Nuclear PlantOne Cook PlaceBridgman, MI 49106Indiana Michigan Power.comAEP-NRC-2013-5310 CFR 2.201U.S. Nuclear Regulatory CommissionAttn: Document Control DeskWashington, DC, 20555-0001Donald C. Cook Nuclear Plant Units 1 and 2Response to the Non-Cited Violations Resulting from ComponentDesign Bases Inspection 05000315/2013010; 05000316/2013010References:1. Letter from W. Hodge, Indiana Michigan Power Company (I&M), to C. Tilton, U.S. NuclearRegulatory Commission (NRC), "D. C. Cook CDBI Response to Question 2012-CDBI-298,"dated November 15, 2012, (ADAMS Accession No. ML12320A544).2. Letter from K. O'Brien, NRC, to S. Bahadur, NRC, "Task Interface Agreement -LicensingBasis for Donald C. Cook Nuclear Power Plant, Units 1 and 2, During a Steam GeneratorTube Rupture Event Coincident with a Loss of Offsite Power (TIA 2012-11)," datedDecember 7, 2012, (ADAMS Accession No. ML13011A382).3. Letter from A. M. Stone, NRC, to L. J. Weber, I&M, "D. C. Cook Nuclear Power Plant, Units1 and 2, Component Design Bases Inspection 05000315/2012007; 05000316/2012007,"dated January 11, 2013 (ADAMS Accession No. ML13011A401).4. Letter from G. Shear, NRC, to L. J. Weber, I&M, "Donald C. Cook Nuclear Power Plant,Units 1 and 2, Component Design Bases Inspection 05000315/2013010;05000316/2013010," dated July 8, 2013, (ADAMS Accession No. ML13189A243).This letter provides Indiana Michigan Power Company's (l&M's),Nuclear Plant (CNP) Units 1 and 2, response contesting thedocumented by Reference 4, Component Design Bases05000315/2013010; 05000316/2013010.licensee for Donald C. CookNon-Cited Violations (NCVs)Inspection (CDBI) ReportIn Reference 1, I&M identified docketed correspondence supporting I&M's understanding of CNP'slicensing basis to assume only a single-unit loss of offsite power (LOOP) coincident with a designbasis Steam Generator Tube Rupture (SGTR) accident. In Reference 2, the Nuclear RegulatoryCommission (NRC) Region III Staff issued a Task Interface Agreement Report documenting U.S. Nuclear Regulatory Commission AEP-NRC-2013-53Page 2the results of its consultation with the NRC Office of Nuclear Reactor Regulation regarding the NRCStaff's understanding of CNP's licensing basis to assume a multi-unit LOOP as an initial conditionof a design basis SGTR accident. In Reference 3, the NRC Staff notified I&M that two potentialfindings relating to the operability of steam generator power operated relief valves (SG PORVs)during a design basis SGTR accident identified by the NRC Staff during a CDBI performed at CNPbetween July 23, 2012, and December 31, 2012, would remain unresolved items (URIs) pendingthe NRC Staffs resolution of questions regarding the scope of a LOOP assumed within CNP'sSGTR accident analysis. In Reference 4, the NRC Staff resolved the URIs issued by Reference 3and issued NCVs of CNP Technical Specifications 5.4.1 (prescribing emergency operatingprocedures (EOPs) to mitigate the consequences of a design basis SGTR accident) and 3.7.4(governing the operability of SG PORVs). Reference 4 states that I&M had violated TechnicalSpecification 5.4.1 because CNP EOPs could not ensure that personnel would be able to operateSG PORVs as required by CNP's licensing basis during an SGTR accident accompanied by aLOOP affecting both units at CNP. Reference 4 also states that I&M had violated TechnicalSpecification 3.7.4 because it had failed on several occasions to declare the SG PORVsunavailable after taking a control air compressor out of service for maintenance. Reference 4characterized the NCVs as representing a more-than-minor performance deficiency with cross-cutting aspects.I&M contests the NCVs identified in Reference 4 because those NCVs lack technical justificationand are inconsistent with NRC regulations and guidance. Specific bases for I&M's contest of theNCVs include the following:* The NCVs are based on an erroneous understanding of CNP's licensing basis. Contrary tothe NCVs, CNP's licensing basis assumptions regarding the initial conditions for a SGTRaccident have never considered a coincident LOOP involving both units. Further, the NRCStaff's understanding of CNP's licensing basis underlying the NCVs does not acknowledgedocketed correspondence between I&M and NRC Staff supporting I&M's position, does notrepresent a fair reading of CNP's Updated Final Safety Analysis Report (UFSAR), and isinconsistent with the NRC's current regulatory position regarding the loss of offsite power tonon-safety related auxiliary systems at other multi-unit sites.* The NRC Staff has not demonstrated that I&M's understanding of CNP's licensing basis failsto provide adequate protection of public health and safety from either design basis events orbeyond-design basis external events. Further, the NRC Staff has not demonstrated that itsown position would provide a meaningful improvement in the protection of public health andsafety.* The NRC Staff's determination that the NCVs represent a more-than-minor performancedeficiency with cross-cutting aspects is based on an erroneous understanding of the scopeof a LOOP assumed within CNP's design basis SGTR accident analysis, is inconsistent withthe NRC Staffs statements in docketed correspondence, and is unrepresentative of presentlicensee performance.Enclosure 1 to this letter contains an affirmation statement. Enclosure 2 to this letter lays out indetail the regulatory and factual support for I&M's response contesting the | {{#Wiki_filter:INDIANAMICHIGANPOWERA unit of American Electric PowerAugust 2, 2013Docket Nos.: 50-31550-316Indiana Michigan PowerCook Nuclear PlantOne Cook PlaceBridgman, MI 49106Indiana Michigan Power.comAEP-NRC-2013-5310 CFR 2.201U.S. Nuclear Regulatory CommissionAttn: Document Control DeskWashington, DC, 20555-0001Donald C. Cook Nuclear Plant Units 1 and 2Response to the Non-Cited Violations Resulting from ComponentDesign Bases Inspection 05000315/2013010; 05000316/2013010References:1. Letter from W. Hodge, Indiana Michigan Power Company (I&M), to C. Tilton, U.S. NuclearRegulatory Commission (NRC), "D. C. Cook CDBI Response to Question 2012-CDBI-298,"dated November 15, 2012, (ADAMS Accession No. ML12320A544).2. Letter from K. O'Brien, NRC, to S. Bahadur, NRC, "Task Interface Agreement -LicensingBasis for Donald C. Cook Nuclear Power Plant, Units 1 and 2, During a Steam GeneratorTube Rupture Event Coincident with a Loss of Offsite Power (TIA 2012-11)," datedDecember 7, 2012, (ADAMS Accession No. ML13011A382).3. Letter from A. M. Stone, NRC, to L. J. Weber, I&M, "D. C. Cook Nuclear Power Plant, Units1 and 2, Component Design Bases Inspection 05000315/2012007; 05000316/2012007,"dated January 11, 2013 (ADAMS Accession No. ML13011A401).4. Letter from G. Shear, NRC, to L. J. Weber, I&M, "Donald C. Cook Nuclear Power Plant,Units 1 and 2, Component Design Bases Inspection 05000315/2013010;05000316/2013010," dated July 8, 2013, (ADAMS Accession No. ML13189A243).This letter provides Indiana Michigan Power Company's (l&M's),Nuclear Plant (CNP) Units 1 and 2, response contesting thedocumented by Reference 4, Component Design Bases05000315/2013010; 05000316/2013010.licensee for Donald C. CookNon-Cited Violations (NCVs)Inspection (CDBI) ReportIn Reference 1, I&M identified docketed correspondence supporting I&M's understanding of CNP'slicensing basis to assume only a single-unit loss of offsite power (LOOP) coincident with a designbasis Steam Generator Tube Rupture (SGTR) accident. In Reference 2, the Nuclear RegulatoryCommission (NRC) Region III Staff issued a Task Interface Agreement Report documenting | ||
U.S. Nuclear Regulatory Commission AEP-NRC-2013-53Page 2the results of its consultation with the NRC Office of Nuclear Reactor Regulation regarding the NRCStaff's understanding of CNP's licensing basis to assume a multi-unit LOOP as an initial conditionof a design basis SGTR accident. In Reference 3, the NRC Staff notified I&M that two potentialfindings relating to the operability of steam generator power operated relief valves (SG PORVs)during a design basis SGTR accident identified by the NRC Staff during a CDBI performed at CNPbetween July 23, 2012, and December 31, 2012, would remain unresolved items (URIs) pendingthe NRC Staffs resolution of questions regarding the scope of a LOOP assumed within CNP'sSGTR accident analysis. In Reference 4, the NRC Staff resolved the URIs issued by Reference 3and issued NCVs of CNP Technical Specifications 5.4.1 (prescribing emergency operatingprocedures (EOPs) to mitigate the consequences of a design basis SGTR accident) and 3.7.4(governing the operability of SG PORVs). Reference 4 states that I&M had violated TechnicalSpecification 5.4.1 because CNP EOPs could not ensure that personnel would be able to operateSG PORVs as required by CNP's licensing basis during an SGTR accident accompanied by aLOOP affecting both units at CNP. Reference 4 also states that I&M had violated TechnicalSpecification 3.7.4 because it had failed on several occasions to declare the SG PORVsunavailable after taking a control air compressor out of service for maintenance. Reference 4characterized the NCVs as representing a more-than-minor performance deficiency with cross-cutting aspects.I&M contests the NCVs identified in Reference 4 because those NCVs lack technical justificationand are inconsistent with NRC regulations and guidance. Specific bases for I&M's contest of theNCVs include the following:* The NCVs are based on an erroneous understanding of CNP's licensing basis. Contrary tothe NCVs, CNP's licensing basis assumptions regarding the initial conditions for a SGTRaccident have never considered a coincident LOOP involving both units. Further, the NRCStaff's understanding of CNP's licensing basis underlying the NCVs does not acknowledgedocketed correspondence between I&M and NRC Staff supporting I&M's position, does notrepresent a fair reading of CNP's Updated Final Safety Analysis Report (UFSAR), and isinconsistent with the NRC's current regulatory position regarding the loss of offsite power tonon-safety related auxiliary systems at other multi-unit sites.* The NRC Staff has not demonstrated that I&M's understanding of CNP's licensing basis failsto provide adequate protection of public health and safety from either design basis events orbeyond-design basis external events. Further, the NRC Staff has not demonstrated that itsown position would provide a meaningful improvement in the protection of public health andsafety.* The NRC Staff's determination that the NCVs represent a more-than-minor performancedeficiency with cross-cutting aspects is based on an erroneous understanding of the scopeof a LOOP assumed within CNP's design basis SGTR accident analysis, is inconsistent withthe NRC Staffs statements in docketed correspondence, and is unrepresentative of presentlicensee performance.Enclosure 1 to this letter contains an affirmation statement. Enclosure 2 to this letter lays out indetail the regulatory and factual support for I&M's response contesting the NCVs. | |||
U.S. Nuclear Regulatory Commission AEP-NRC-2013-53Page 3Regardless of the outcome of I&M's contest of the NCVs, I&M will continue to evaluate cost-effective measures for the improvement of safety margins against SGTR accidents.Following the 2012 CDBI, I&M revised CNP procedures and implemented plant modifications toprovide additional defense-in-depth and improved safety margins during an SGTR accident. InMarch 2013, I&M completed installation of a plant modification and revised CNP operatingprocedures to ensure that backup nitrogen tanks are immediately and automatically available duringan SGTR accident for operation of SG PORVs without the need for manual valve manipulationoutside the control room. I&M has also revised CNP Work Control processes to provide additionaldefense-in-depth from a loss of control air pressure by restricting removal for maintenance of theoperating unit's control air compressor when the opposite unit is shutdown and the shutdown unit'splant air compressor is aligned to preferred offsite power.This letter contains no new or revised commitments. If you have any questions, please contactMr. Michael K. Scarpello, Regulatory Affairs Manager, at (269) 466-2649.Sincerely,Joel P. GebbieSite Vice PresidentDMB/kmhEnclosures:1. Affirmation2. Indiana Michigan Power Company's Response to "Donald C. Cook Nuclear Power Plant,Units 1 and 2, Component Design Bases Inspection 05000315/2013010;05000316/2013010," dated July 8,2013c: C. A. Casto, NRC Region IIIJ.T. King, MPSCS. M. Krawec, AEP Ft. Wayne, w/o enclosureE. Leeds, NRC NRRMDEQ-RMD/RPSNRC Resident InspectorA. M. Stone, NRC Region IIIC. Tilton, NRC Region IIIT. J. Wengert, NRC Washington, DCR.P. Zimmerman, NRC Washington, DC | |||
ENCLOSURE I TO AEP-NRC-2013-53AFFI RMATIONI, Joel P. Gebbie, being duly sworn, state that I am Site Vice President of Indiana Michigan PowerCompany (I&M), that I am authorized to sign and file this request with the Nuclear RegulatoryCommission on behalf of I&M, and that the statements made and the matters set forth hereinpertaining to I&M are true and correct to the best of my knowledge, information, and belief.Indiana Michigan Power CompanyJoel P. GebbieSite Vice PresidentSWORN TO AND SUBSCRIBED BEFORE METHIS____ DAY OF ,A)ws 2013My Commission Expires ( I 2 IO{ | |||
ENCLOSURE 2 TO AEP-NRC-2013-53Indiana Michigan Power Company's Response to "Donald C. Cook Nuclear PowerPlant, Units 1 and 2, Component Design Bases Inspection 05000315/2013010;05000316/2013010," dated July 8, 20131. IntroductionThe Non-Cited Violations (NCVs) within the Nuclear Regulatory Commission (NRC) StaffsJuly 8, 2013, letter (Reference 1) to Indiana Michigan Power Company (I&M) are based on anerroneous understanding of the licensing basis of Donald C. Cook Nuclear Plant (CNP). TheNRC Staff's position that CNP's design basis Steam Generator Tube Rupture (SGTR) accidentassumes a coincident loss of offsite power (LOOP) that can involve both units at CNP isinconsistent with pertinent, docketed correspondence between the NRC Staff and I&M. Further,the NRC Staff's position is unsupported by a fair reading of CNP's Updated Final SafetyAnalysis Report (UFSAR), and is likewise inconsistent with relevant historical and currentregulatory positions of the NRC. Additionally, the NRC Staff has not demonstrated that I&M'sunderstanding of CNP's licensing basis fails to provide adequate protection of public health andsafety from either design basis events or beyond-design basis external events. Lastly, the NRCStaff's determination that the NCVs represent a more-than-minor performance deficiency withcross-cutting aspects relies on an erroneous understanding of the scope of a LOOP assumedwithin CNP's design basis SGTR accident analysis, is inconsistent with the NRC Staff'sstatements in docketed correspondence, and is unrepresentative of present licenseeperformance.Documents referenced herein are listed as references at the end of this Enclosure.2. History of the Non-Cited ViolationsThe NCVs contested by I&M result from findings by the NRC Staff during the ComponentDesign Bases Inspection (CDBI) conducted at CNP between July 23, 2012, andDecember 31, 2012. As described in Reference 2, the CDBI entailed a review of licensing basisdocumentation and drawings of the CNP compressed air system to verify that support functionsprovided to the steam generator power operated relief valves (SG PORVs) were consistent withCNP's licensing basis requirements for SGTR accidents.As stated in Reference 2, the NRC Staff contended during the CDBI that CNP was not inconformance with Technical Specifications 5.4.1 (prescribing emergency operating procedures(EOPs) to mitigate the consequences of a design basis SGTR accident) and 3.7.4 (governingthe operability of SG PORVs). Based on its belief that CNP's licensing basis assumptions for aSGTR accident included a coincident LOOP affecting both units at CNP, the NRC Staffreasoned that the only available source of control air pressure during the most limiting SGTRaccident would be the affected unit's dedicated control air compressor (CAC) receiving powerfrom one of the two emergency diesel generators (EDG). However, if the affected unit's CACwere unavailable as a result of emergent or planned maintenance, then the NRC Staff reasonedthat control air pressure would be unavailable to operate the affected unit's SG PORVs. Inreviewing CNP operating records, the NRC Staff identified several occasions in which CACs at | |||
Enclosure 2 to AEP-NRC-2013-53Page 2CNP would have been unavailable due to maintenance, but I&M had not declared the SGPORVs inoperable.I&M disagreed with the NRC Staff's characterization of CNP's licensing basis assumptions for aSGTR event. Noting that the CNP licensing basis for an SGTR event did not consider acoincident multi-unit LOOP, I&M contended that the NRC Staffs finding was based on a beyonddesign basis accident scenario. The NRC Staff requested assistance from the NRC Office ofNuclear Reactor Regulation (NRR) in resolving the disagreement regarding CNP's licensingbasis assumptions. On November 15, 2012, I&M submitted Reference 3 to NRC Staff,containing information identifying the technical and regulatory bases supporting I&M's positionand providing docketed correspondence. Reference 3 in particular identified a SafetyEvaluation Report (SER, Reference 4) dated October 24, 2001, explicitly discussing CNP'sassumptions for SGTR accident initial conditions, and revealing the NRC Staff's evaluation andendorsement of I&M's understanding of the CNP licensing basis assumptions for an SGTRaccident.On December 7, 2012, NRC Region III Staff issued Reference 5 after consulting with NRR,contradicting I&M's understanding of CNP's licensing basis assumptions for SGTR accidents.Reference 5 cited only three passages within CNP's UFSAR (Reference 6) in support of itsposition, interpreting a handful of references to the terms "LOOP" and "station" in descriptions ofCNP electrical systems to mean that CNP's licensing basis assumed a LOOP would affect bothunits at CNP in an SGTR accident. Reference 5 suggests that it did not examine the technicaland regulatory bases and docketed correspondence supporting a contrary position referencedwithin Reference 3 submitted by I&M.On January 11, 2013, the NRC Staff issued Reference 2, identifying the CDBI findings at issueas unresolved items (URIs) pending submission of additional information from I&M regardingCNP's licensing basis assumptions for SGTR accidents. Reference 2 repeated Reference 5'sconclusions regarding CNP's licensing basis assumptions for SGTR accidents without furtherexplanation or analysis; further, Reference 2 again did not address the technical and regulatorybases and docketed correspondence identified in Reference 3 forwarded by I&M. OnFebruary 8, 2013, I&M provided Reference 7 to the NRC Staff, refuting Reference 5'sinterpretation of CNP's UFSAR and providing additional detail regarding the technical andregulatory bases supporting I&M's understanding of the CNP licensing basis assumptions for anSGTR accident. During a May 20, 2013, technical debrief of the CDBI findings, the NRC Staffrepeated its understanding of the scope of the LOOP assumed within SGTR's accident analysis,again without addressing the technical and regulatory bases and docketed correspondencesupporting I&M's position. In a re-exit teleconference for the URIs conducted on May 24, 2013,the NRC Staff informed I&M that the NRC Staff planned to issue an NCV for violation ofTechnical Specification 3.7.4 requirements regarding the operability of SG PORVs.On July 8, 2013, the NRC Staff issued Reference 1. In Reference 1, the NRC Staff identifiedNCVs of CNP Technical Specifications 5.4.1 (prescribing EOPs to mitigate the consequences ofa design basis SGTR accident) and 3.7.4 (governing the operability of SG PORVs). Reference1 states that I&M had violated Technical Specification 5.4.1 because CNP EOPs could notensure that personnel would be able to operate SG PORVs as required by CNP's licensingbasis during an SGTR accident accompanied by a LOOP affecting both units at CNP.Reference 1 also states that I&M had violated Technical Specification 3.7.4 because it had | |||
Enclosure 2 to AEP-NRC-2013-53Page 3failed on several occasions to declare the SG PORVs unavailable after taking a CAC out ofservice for maintenance.Reference 1 characterized the NCVs as representing a more-than-minor, cross-cuttingperformance deficiency involving areas of human performance, the component ofdecisionmaking, and the aspect of conservative assumptions because I&M had incorrectlyassumed that control air pressure to the SG PORVs of a unit experiencing an SGTR accidentaccompanied by a LOOP would remain available from the unaffected unit's plant air compressor(PAC).Reference 1 also attempted to refute I&M's explanation within Reference 7 of its understandingof CNP's licensing basis assumptions for SGTR accidents. Acknowledging I&M's position thatCNP's licensing basis did not assume a single failure of a non-safety-related component (inparticular, the unaffected unit's PAC), during an SGTR event, Reference 1 contends that I&Mhad nevertheless failed to demonstrate that control air would reasonably be available during anSGTR event accompanied by a multi-unit LOOP. Similarly, Reference 1 asserts that even if theunaffected unit's PAC would be available during a design basis SGTR accident, I&M had failedto identify that assumption within its SGTR accident analysis, and the NRC Staff had neverexplicitly approved that assumption. Further, Reference 1 endorsed Reference 5'sinterpretation of the UFSAR's use of the term LOOP to refer to multi-unit events, adding that theabsence of CNP operating procedures preventing alignment of the same offsite power sourcesto both units made a multi-unit LOOP a credible event within CNP's licensing basis.3. Overview of Pertinent CNP Systems and Operatinq Proceduresa. CNP Steam Generator Power Operated Relief ValvesIn accordance with Reference 6 (at Sections 10.2.2 and 14.2.4), the SG PORVs preventoverpressure conditions in the steam generators by releasing secondary system steam toatmosphere following a loss of condenser vacuum. The SG PORVs form part of the mainsteam system pressure boundary, and thus are safety-related equipment for main steam systempressure retention.CNP operating procedures prescribe operator actions in the event of a SGTR accident. CNPoperating procedures allow SG PORVs to be operated using motive force provided by controlair supplied by either the compressed air system shared between the two units, control airpressure supplied by a unit-specific CAC, or installed backup nitrogen tanks that can be alignedto the SG PORVs. In March 2013, I&M completed installation of a plant modification andrevised its operating procedures to ensure that the backup nitrogen tanks are immediately andautomatically available during an SGTR accident without the need for manual valvemanipulation outside the control room.b. CNP Compressed Air SystemSection 9.8.2 of Reference 6 describes the control air provided by CNP's compressed airsystem as the ordinary source of motive force for operation of SG PORVs for both units at CNP.Per Reference 6, Section 1.3.9.h, CNP's compressed air system is a single system sharedbetween both units at CNP. Each unit at CNP contains one CAC capable of providing control | |||
Enclosure 2 to AEP-NRC-2013-53Page 4air only within that unit, as well as a PAC capable of providing control air to both units via ashared header. Both units share a single backup air compressor capable of providing control airto loads within either unit.During normal operations, control air pressure for operating both units' SG PORVs is providedby one of the two PACs. Low pressure in the shared plant compressed air header will result inthe automatic start and loading of the other unit's PAC. Low control air header pressure in oneof the unit-specific control air headers will cause that unit's CAC to start.During normal operations, the operating PAC receives power from its unit's auxiliarytransformers, which are in turn powered by that unit's main generator or preferred offsite powertransformers. The CAC associated with each unit at CNP can be powered by either offsitepower source in normal operations, but can only receive power from its unit's CD EDG afteroffsite power has been lost to that unit. The CACs and PACs are both non-safety relatedequipment governed by the Maintenance Rule at 10 CFR 50.65.CNP Work Control processes impose a series of administrative controls to maximize availabilityof control air pressure when a CAC or PAC is taken out of service for maintenance:* In the event a CAC is taken out of service for maintenance, bothPACs and the installed backup nitrogen tanks must be guarded; and* In the event that a PAC is taken out of service, the followingequipment is guarded: (1) the opposite unit's PAC, (2) both CACs, (3)the opposite unit's CD EDG, and (4) the backup air compressor.Following the 2012 CDBI, I&M revised CNP Work Control processes to provide additionaldefense-in-depth from a loss of control air pressure by restricting removal for maintenance ofthe operating unit's CAC when the opposite unit is shutdown and the shutdown unit's PAC isaligned to preferred offsite power.4. Regulatory Basis for the Assumption of Only a Single-Unit LOOP within CNP's SGTRAccident Analysisa. CNP's Licensing Basis Has from the Beginning Assumed that an SGTR AccidentWould Involve a Coincident, Single-Unit LOOPCNP's original licensing basis explicitly assumed that SG PORVs would remain availablethroughout an SGTR accident. As described in the Preliminary Safety Analysis Report (PSAR,Reference 9) for Units 1 and 2 submitted on December 18, 1967, and repeated in Sections14.2.4 and 14.2.7 of the FSAR for Units 1 and 2 dated February 2, 1971 (Reference 10), CNP'soriginal licensing basis evaluated the radiological consequences of an SGTR accident byconservatively estimating the mass release of radioactivity to the environment over the30-minute time span between SGTR accident initiation and subsequent termination of primaryto secondary mass transfer from the completion of mitigation measures taken by operators.I&M's analytical assumption of 30 minutes' mass release before termination of the event wasconsidered inherently conservative because it neglected the reduction in mass flow that wouldoccur during this same time period. | |||
Enclosure 2 to AEP-NRC-2013-53Page 5Inherent in that postulated 30-minute mass release was an assumption of the success ofoperator actions such as the operation of SG PORVs to mitigate the event. Section 14.2.4 ofReference 10 in several places explicitly credited the availability of SG PORVs during a designbasis SGTR regardless of conditions.Reference 10's evaluation of SGTR accidents omits any mention of the possibility thatcompressed air system components could be unavailable as a result of a single failure ormaintenance, as it prefaced its elaboration of the sequence of events initiated by an SGTRevent by stating that its analysis had "assum[ed] normal operation of the various plant controlsystems ....... Reference 10 at Section 14.2.4. Further, Reference 10 assumed that SGPORVs would remain available regardless of the status of offsite power, stating that when a unitwas "without offsite power":Condenser bypass valves will automatically close and the steamgenerator pressure will rapidly increase resulting in steam discharge tothe atmosphere through the steam generator safety valves and/or thepower operated relief valves.Reference 10 at Section 14.2.4. Elsewhere, Reference 10 noted that:In the event of a co-incident station blackout, the steam dump valveswould automatically close to protect the condenser. The steam generatorpressure would rapidly increase resulting in steam discharge to theatmosphere through the steam generator safety and/or power operatedrelief valves.Reference 10 at Section 14.2.4 (emphasis added).I&M's assumption that SG PORVs remained available for mitigation of an SGTR accident isconsistent with the description of the compressed air system elsewhere within CNP's originalFSAR. Among the design bases for CNP's compressed air system within Reference 10 is arequirement for continued availability of control air:The [compressed air system] must provide a continuous supply ofcompressed air to vital systems under both normal and abnormalconditions.Reference 10 at Section 9.8.2 (emphasis added). With this in mind, each of CNP's PACs weredesigned to be "capable of supplying the entire demand of both plant and control-instrument airrequirements for both units," as the offline PAC automatically started on low pressure in the(shared) plant air header. Reference 10 at Section 9.8.2.3.Although CNP's original FSAR accounted for the availability of compressed air systemcomponents within the opposite plant, the staggered construction and licensing of CNP Units 1and 2 resulted in a more unit-specific design and function for other CNP systems. For example,Unit l's construction and licensing (1974) several years before Unit 2 (1977) meant that thedesign bases of the electrical systems for each of the two units at CNP were, as a practicalmatter, unit-specific. For example, although each EDG shares a fuel oil tank with an EDG in the | |||
Enclosure 2 to AEP-NRC-2013-53Page 6other unit, the fuel oil tank's capacity is based on the design operational requirements of asingle EDG. Reference 6 at Section 8.4. Consequently, references within Reference 10'sSGTR accident analysis to a "loss of offsite power" or a "station blackout" referred to an eventinvolving only a single unit.The analysis of a design basis SGTR accident in the revised FSAR evaluating Unit 2 as-built(Reference 11) used nearly identical language to that used within the SGTR accident analysis inthe original Units 1 and 2 FSAR (Reference 10). Further, subsequent versions of both units'UFSAR analyses for SGTR accidents retained the CNP's original assumptions regarding theavailability of SG PORVs -and, in fact, arguably placed even greater emphasis on thecontinued availability of those components in their SGTR accident analysis. In particular,July 1997 revisions to the UFSAR for both units were revised to better track CNP EOPsidentifying the SG PORVs (and not the steam generator safety valves) as the initial means ofpreventing steam generator overpressure after loss of offsite power:In the event of a coincident station blackout, the steam dump valveswould automatically close to protect the condenser. The steam generatorpressure would rapidly increase, resulting in steam discharge to theatmosphere through the steam generator power operated relief valves(and the steam generator safety valves if their setpoint had beenreached).Reference 12 at Section 14.2.4 (emphasis added). Later UFSAR revisions to CNP's SGTRaccident analysis also incorporated the original FSAR's language describing the continuedavailability of SG PORVs despite a LOOP or station blackout virtually unchanged. Reference 6at Section 14.2.4. Further, I&M's review of pertinent docketed correspondence with the NRCStaff has discovered no evidence of a departure from CNP's original assumption of a unit-specific LOOP coincident with an SGTR accident.b. The NRC Staff Has Reviewed and Endorsed CNP's Design Basis Assumptions forSGTR Accidents in Docketed CorrespondenceOn October 24, 2000, I&M submitted a license amendment request (LAR, Reference 10) torevise the methodology used in designing CNP EOPs during a design basis SGTR accident.The Westinghouse Owners Group methodology (WCAP-10698-P-A ("SGTR AnalysisMethodology to Determine Margin to Steam Generator Overfill")) that I&M proposed to adapt foruse within its SGTR accident analysis incorporated lessons learned from operationalexperience, plant simulator studies, and advances in computer modeling techniques to bettercharacterize steam generator fill conditions during an SGTR accident. Of particular importanceto CNP was that the LOFTTR2 computer program used in the WCAP-10698-P-A methodologysimulated the effects of operator actions on margin to steam generator overfill during an SGTRaccident. By incorporating elements of the WCAP-10698-P-A methodology for the simplifiedcalculations of margin to steam generator overfill within its original SGTR accident analysisassumptions, I&M could revise CNP EOPs to assure margins to steam generator overfill whileremaining within the conservative margins to radiological consequences described in its originalSGTR accident analysis. | |||
Enclosure 2 to AEP-NRC-2013-53Page 7Although the NRC had previously accepted WCAP-10698-P-A for use by licensees, the NRCStaff had to evaluate its application within CNP's SGTR accident analysis. In a series ofdocketed correspondence with the NRC Staff detailing how the WCAP-10698-P-A would beused within CNP's SGTR accident analysis, I&M repeatedly emphasized that the newmethodology would not disturb existing license basis assumptions in its SGTR accidentanalysis. Specifically, the safety analysis for I&M's LAR noted that:The proposed change ...does not affect any accident initiators orprecursors .... The proposed change also does not affect the ability ofoperators to mitigate the consequences of an accident.Reference 13, Attachment 1 at Page 4 (emphasis added). I&M repeated this claim in the LAR'sevaluation of significant hazards required by 10 CFR 50.92(c):[T]he new methodology does not affect equipment malfunctionprobability .... The proposed change does not impact the design ofaffected plant systems, involve a physical alteration to the systems, orchange the way in which systems are currently operated, such thatpreviously unanalyzed SGTRs would not occur. The change toincorporate the WCAP-10698-P-A methodology does not introduce anynew malfunctions ....Reference 13, Attachment 2 at Pages 2-3 (emphasis added).Subsequent docketed correspondence between I&M and the NRC Staff was even more explicitin describing the retention of existing license basis assumptions for SGTR accidents. In aJune 29, 2001, response (Reference 14) to a May 7, 2001, letter from the NRC Staff requestingadditional information (RAI) regarding how I&M intended to use the WCAP-10698-P-A within itsSGTR accident analysis, I&M emphasized that its use of the WCAP-10698-P-A methodologywas "limited", and that, by-and-large, "CNP's present methodology would be retained forcalculating the radiological consequences of the postulated SGTR .... ." Reference 14,Attachment 1 at Page 1. In particular, I&M noted that its analysis retained existing licensingbasis assumptions regarding the availability of certain systems, components, and instruments(listed in a table within Reference 14) credited for accident mitigation in an SGTR. Among theitems listed in that table were the "air-operated" SG PORVs, which the notes accompanying thetable stated were themselves safety-grade components because they "form part of the mainsteam system pressure boundary upstream of the SG stop valves," even though their "electricaland control air appurtenances [were] not safety-grade." Reference 14, Attachment 1 at Pages3-4. Reference 14 also noted that I&M's limited use of the WCAP-10698-P-A methodologywould not disturb CNP's existing licensing basis assumption that an SGTR accident would notinvolve a single failure. Reference 14, Attachment 1 at Page 6.Reference 14 also communicated I&M's intention to retain CNP's existing assumptionsregarding the availability of offsite power. Acknowledging that the WCAP-10698-P-Amethodology assumes that "the most challenging SGTR scenario with respect to SG fill includesa coincident loss of offsite power", Reference 14 noted that the modified SGTR analysis wouldretain CNP's original licensing assumption that SG PORVs would remain available despite thefact that "offsite power [was] not ...available." Reference 14, Attachment 1 at Page 4. | |||
Enclosure 2 to AEP-NRC-2013-53Page 8Reference 14 contained no suggestion of a change in the scope of the LOOP assumed withinCNP's SGTR accident analysis.By letter dated October 24, 2001 (Reference 4), the NRC Staff approved I&M's LAR in modifiedform to accommodate CNP's existing licensing basis assumptions for SGTR accidents. In theSER submitted with its approval of I&M's LAR, the NRC Staff acknowledged that licensees likeI&M could not incorporate the WCAP-10698-P-A methodology within their SGTR accidentanalysis in a uniform fashion because "variations in plant designs prevent a single model fromadequately representing all Westinghouse Plants." Reference 4, SER at Page 2.Consequently, the NRC Staff devoted much of the SER to evaluating the differences betweenthe generic WCAP-1 0698-P-A methodology and I&M's proposed approach for incorporating thatmethodology within its licensing basis.The NRC Staff noted that in the immediate case, those differences included I&M's intention ofretaining CNP's existing assumptions for SGTR accidents:To implement the WCAP, the licensee used the LOFTTR2 computer codeand the plant-specific current licensing basis assumptions.Reference 4, SER at Page 2 (emphasis added). The NRC Staff explicitly acknowledged thatCNP's licensing basis assumptions credited certain systems and components, including the SGPORVs and their control air appurtenances, as remaining available for mitigation of an SGTRaccident:The licensee provided a list of systems, components, and instrumentationthat are used for SGTR accident mitigation. They also specified thesafety classification of the systems and power sources. However, thelicensee listed several systems used for SGTR mitigation that are notsafety related and do not have safety related backups. The licenseejustified the use of the non-safety-related equipment by stating that thesesystems are credited in the current UFSAR Section 14.2.4 accidentanalysis. Upon review of Section 14.2.4, the staff concludes that thelicensing basis SGTR analysis does credit limited use of non-safety gradeequipment for mitigating the SGTR.Reference 4, SER at Page 3. Similarly, the NRC Staff acknowledged that CNP's licensing basisdid not assume a worst single failure during an SGTR accident as the WCAP-10698-P-Amethodology did:[T]he licensee did not assume the worst single failure as prescribed bythe WCAP-10698-P-A safety analysis, and did not provide it's [sic] effecton the margin to overfill. The licensee based their decision not to assumethe worst single failure on the fact that their current licensing basis doesnot include a single failure.Reference 4, SER at Page 4. Further, the SER nowhere mentions that I&M intended to discardCNP's existing assumption of a coincident single-unit LOOP during an SGTR accident, or that | |||
Enclosure 2 to AEP-NRC-2013-53Page 9the LOOP assumed within the WCAP-10698-P-A methodology supplanted CNP's existinglicensing basis assumptions for SGTR accidents.Although I&M's proposed retention of CNP's existing licensing basis assumptions for SGTRaccidents "varied significantly" from the assumptions underlying the WCAP-10698-P-Amethodology, the NRC Staff approved I&M's use of some elements of the WCAP-10698-P-Amethodology identified in the LAR and related correspondence:[T]he NRC staff concludes that the licensee can incorporate theLOFTTR2 code into its licensing bases for CNP and can use theLOFTTR2 code, with the current licensing basis assumptions as inputs forthe overfill analysis of steam generator tube rupture accidents. Thischange to the licensing basis does not affect accident initiators orprecursors. This change also does not ...decrease the ability of theoperators to mitigate the consequences of an accident.Reference 4, SER at Page 5 (emphasis added). In justifying its approval of a modifiedWCAP-10698-P-A methodology for use at CNP, the NRC Staff noted that I&M's adaptation ofthe WCAP-10698-P-A methodology to CNP's existing licensing basis assumptions for SGTRaccidents did not affect conservative estimates of the radiological consequences of a designbasis SGTR at CNP. Reference 4, SER at Page 3.I&M's subsequent review of docketed correspondence with the NRC Staff has identified nofurther changes to CNP's licensing basis assumptions regarding the availability of SG PORVs inan SGTR accident, the absence of a single failure assumption within CNP's SGTR accidentanalysis, or the scope of a LOOP assumed in the SGTR analysis.5. The NRC Staff's Understanding of CNP's Licensing Basis Assumptions for SGTR AccidentsDoes Not Address Pertinent Docketed Correspondence, Is Unsupported by a Fair Readingof the UFSAR, and is Inconsistent with the NRC's Historical and Current RegulatoryPositionsa. The NRC Staff's Reading of CNP's Licensing Basis Assumptions for SGTRAccidents Does Not Address Pertinent Docketed CorrespondenceAs noted earlier, the NCVs within Reference 1 are based on the NRC Staffs contention that thecoincident LOOP assumed within CNP's licensing basis SGTR accident analysis involves a lossof offsite power to both units at CNP. The NRC Staff's position is based on a single argumentwithin Reference 5: that it follows from the use of the terms "LOOP" and "station" in a handful ofCNP UFSAR sections, some of which are unrelated to SGTR accident analysis, that a LOOPcan refer to the denial of offsite power to one or both units at CNP.In support of this argument, Reference 5 advances only a handful of UFSAR passages. Thefirst UFSAR passage referenced in Reference 5 comes from Section 1.3.7 describing theauxiliary electrical system for each of the two units at CNP:Donald C. Cook's UFSAR Section 1.3.7, "Electrical System" states, "Themain generators are 1800 rpm, Phase III, 60 cycle, hydrogen and water | |||
Enclosure 2 to AEP-NRC-2013-53Page 10cooled units. The main transformers deliver generator power to the345kV and 765 kV switchyards. The station auxiliary power systemconsists of auxiliary transformers, 4160V and 600 V switchgear, 600Vmotor control centers, 120 V A-C vital instrument buses and 250 V D-Cbuses."Reference 5 at Page 3 (emphasis supplied by NRC Staff). Based on the fact that UFSARSection 1.3.7 described the identical electrical systems for both units, Reference 5 concludedthat the UFSAR passage's reference to "station" must refer to both units at CNP, rather than toeach unit individually. In the same vein, Reference 5 cites a passage from Section 1.3.8 of theUFSAR describing the Safety Features associated with each unit at CNP:Also, Section 1.3.8, "Safety Features," describes the safety featuresincorporated into the design of the plant, including the fact that "even ifexternal auxiliary power to the station is lost concurrent with an accident,power is available for the engineered safeguards from on-site dieselgenerator power to assure protection of the public health and safety forany loss of coolant accident."Reference 5 at Page 3 (emphasis supplied by NRC Staff). Here, too, Reference 5 concludesthe fact that Section 1.3.8 describes identical safety features at each unit means that thepassage's reference to "station" must refer to both units at CNP, rather than only one unit.Lastly, Reference 5 points to language within a passage from the accident analysis (atSection 14.1.12) for "Loss of All AC Power to the Plant Auxiliaries" at Unit 1:"A complete loss of all (non-emergency) AC Power (e.g., offsite power)may result in the loss of all power to the plant auxiliaries, i.e., the RCPs,condensate pumps, etc. The loss of power may be caused by a completeloss of the offsite grid accompanied by a turbine trip at the station, or by aloss of the on-site AC distribution system."Reference 5 at Page 4. The NRC Staff read this reference to a "complete loss of offsite gridaccompanied by a turbine trip at the station" associated with the design basis event postulatedwithin Section 14.1.12 to mean that a LOOP affecting both units is within CNP's licensing basisfor every event evaluated in UFSAR Section 14. Reference 5 at Page 4. Based on theseexamples, Reference 5 reports that NRR concurred with NRC Staff that had performed theCDBI that the LOOP assumed in CNP's SGTR analysis was a "station event, not a unit specificevent." Reference 5 at Page 4.The NRC Staff's position and the UFSAR passages described above represent the only basisidentified by the NRC Staff for its position throughout the multiple docketed communications andmeetings with I&M since the CDBI began in July 2012. The NRC Staff has identified noregulatory provisions or policy guidance requiring the assumption of a LOOP affecting both unitsfor a design basis SGTR accident. The NRC Staff has advanced no docketed correspondencein support of its understanding of CNP's licensing basis for SGTR accidents, and has identifiedno additional passages within CNP's UFSAR supporting its position. | |||
Enclosure 2 to AEP-NRC-2013-53Page 11Further, the NRC Staff has yet to provide a meaningful response to the analysis provided byI&M in References 3 and 7 in support of its understanding of CNP's licensing basisassumptions. Reference 5 does not specifically address the SGTR accident analysisassumptions identified within docketed correspondence highlighted within Reference 3:The scope of this TIA was limited to the licensing basis as related tooffsite power only. The staff did not evaluate other assertions in thelicensee's white paper.Reference 5 at Page 4.1 Reference 2 merely repeated Reference 5's claims regarding CNP'slicensing basis, rather than address the detailed licensing basis interpretation within Reference7 provided by I&M.Further, although Reference 1 suggests that it addresses the understanding of CNP's SGTRaccident licensing basis assumptions advanced by I&M in References 3 and 7, a careful readingof the bases identified in Reference 1 indicates that the NRC Staff's reasoning is circular in thatit depends on, rather than proves the assumption of a multi-unit LOOP in CNP's SGTR accidentanalysis. Specifically, in acknowledging I&M's position that CNP's licensing basis had neverassumed a single failure of a non-safety-related component (specifically the unaffected unit'sPAC) during an SGTR event, Reference 1 contends that I&M had nevertheless failed todemonstrate that an unaffected unit's PAC would reasonably be available during an SGTRaccident affecting one unit:The inspectors agreed that certain older operating plants arecredited with the use of non-safety related equipment to mitigateevents. In these cases, the licensee was required to demonstratethe non-safety-related equipment would reasonably be availableand use of the equipment was bound by a safety-related path.Reference 1, Enclosure at Pages 4 and 5. Similarly, the NRC Staff in Reference 1 agrees withI&M's observation in Reference 7 that the original SER for Unit 1 did not consider that a CACwould be out of service for maintenance pursuant to an assumed single failure, claiming thatthis demonstrates that a CAC would have to be available to supply control air pressure during adesign basis SGTR accident, as its availability would be a limiting condition in CNP's SGTRaccident analysis.However, the above arguments do not prove the NRC's Staff understanding of the scope of theLOOP assumed in CNP's SGTR accident analysis. Because the unaffected unit's non-safety-related PAC would remain available during a single-unit LOOP, control air pressure would bereasonably available and bounded by a safety-related path for main steam system pressureretention purposes, regardless of the status of the CAC on the affected unit. Similarly, theavailability of the affected unit's CAC is not a limiting condition for CNP's SGTR accidentanalysis if the coincident LOOP affects only the unit experiencing the SGTR event such that the1 The NRC Staff has not docketed correspondence between Region III personnel and NRRpersonnel defining the scope of NRR personnel's review of the competing interpretations ofCNP's licensing basis assumptions for the LOOP assumed within CNP's SGTR design basisaccident analysis. | |||
Enclosure 2 to AEP-NRC-2013-53Page 12PAC on the unaffected unit remains available to provide control air pressure to the affectedunit's SG PORVs. Lastly, the NRC Staff statement quoted above is inconsistent with the NRCStaff's statements within Reference 4 endorsing CNP licensing basis assumptions crediting theavailability of SG PORVs and compressed air system components during an SGTR accident.b. The NRC Staff's Position Is Unsupported by a Fair Reading of the UFSARThe NRC Staff's categorical statement that every reference to a LOOP within CNP's UFSARcan be understood to refer to an event denying offsite power to one or both units at CNP isunsupported by a careful reading of that document. The UFSAR contains no generic,controlling definition of the term LOOP requiring it to be understood as referring to either asingle or multi-unit event at every use within the UFSAR. Similarly, the NRC Staff has identifiedno regulatory requirement, policy guidance, or docketed correspondence with I&M requiring anyreference to a LOOP to refer to either a single or multi-unit event. Consequently, whether aparticular reference to a LOOP within CNP's UFSAR refers to a LOOP affecting one or bothunits at CNP must be determined by reference to a number of factors such as the textsurrounding the UFSAR's reference to the LOOP, the larger structure of CNP's UFSAR, as wellas the relevant historical and regulatory background.i. The NRC Staff's Understanding of the Scope of a LOOP Is Not Supported bythe Surroundinq TextA comparison of the different contexts in which the term LOOP appears within CNP's SGTR andLoss of All AC Power to the Plant Auxiliaries accident analyses, respectively, does not supportthe NRC's generic interpretation of the term. As noted earlier, the NRC Staff's understanding ofCNP's licensing basis is based on the potentially broad scope of the LOOP within UFSAR Unit 1Section 14.1.12, "Loss of All AC Power to the Plant Auxiliaries." The UFSAR's description ofthe particular LOOP at issue could involve:A complete loss of all (non-emergency) AC power (e.g., offsite power) ...result[ing] in the loss of all power to the plant auxiliaries .... The loss ofpower may be caused by a complete loss of the offsite grid accompaniedby a turbine generator trip at the station, or by a loss of the on-site ACdistribution system.Reference 5 at Page 4 (quoting UFSAR Unit 1, Section 14.1.12.1) (emphasis added). Becausethe context of the UFSAR cited above passage is on its face ambiguous regarding the numberof units at CNP affected by the LOOP, the NRC Staff contends that it could, based only on agenerous reading of the cited text alone, be read to refer to a LOOP to one or both units atCNP.The context surrounding the use of the term LOOP within the SGTR accident analysis inUFSAR Units 1 and 2 Section 14.2.4 demands an entirely different conclusion regarding thenumber of units losing offsite power in a LOOP. Here, the UFSAR's use of the term LOOP isnot qualified by the broad adjectives, complete loss, all power, the offsite grid, etc., used in theearlier accident analyses in a way that could arguably suggest a LOOP denying power to bothunits; rather, CNP's SGTR accident analysis refers only to "offsite power", or "a loss of offsitepower" or "a coincident loss of offsite power." Reference 6 at Section 14.2.4. | |||
Enclosure 2 to AEP-NRC-2013-53Page 13ii. The NRC Staffs Understandinq of the Meaninq of a LOOP Is Inconsistentwith the Structure of CNP's UFSARThe structure of the UFSAR also undercuts the generic meaning attached to the term LOOP bythe NRC Staff. According to Reference 5, the potentially broad scope of the LOOP described inUFSAR Section 14.1.12 defines the meaning of the term throughout the UFSAR. Reference 5at Page 4. However, the NRC Staff provides no justification for why the particular (broad)meaning it assigns to the term LOOP within UFSAR Section 14.1.12 is more appropriate forgeneric application throughout the UFSAR than the more limited-scope LOOP described withinother sections of the UFSAR such as Section 14.2.4.The NRC Staff's position is also not supported by the NRC and industry guidance regarding theform and content of CNP's UFSAR. Consistent with the scheme laid out in Regulatory Guide1.70 (Reference 15), CNP's UFSAR evaluates transient events and accidents satisfying aminimal threshold for best-estimate frequency of occurrence, which are then assigned afrequency grouping based on criteria established by the American Nuclear Society (ANS). Asstated in UFSAR Sections 14.0, ANS Condition 1 (normal operational transients) are omittedfrom CNP's UFSAR, while Condition 2 events (moderate frequency) appear mostly in UFSARSections 14.1, Condition 3 (infrequent) events in UFSAR Section 14.2, and Condition 4 (unlikelybut limiting) events mostly appear in UFSAR Section 14.3. Consistent with Regulatory Guide1.70, CNP's UFSAR analyzes each of the events within the UFSAR individually and for eachunit, to include a description of the initial assumptions, sequence of events, and radiologicalconsequences specific to each event. Reference 15 at Pages 15-4 to 15-7.The NRC Staff's position does not account for this structure. ANS guidance identifying thethreshold for consideration of transient events and accidents within an FSAR requires a minimalbest-estimate frequency of occurrence of >l.OE-6/yr. Reference 16 at 6. However, when theNRC Staff used its Donald C. Cook Nuclear Plant Standardized Plant Analysis Risk (SPAR)Model to calculate a best-estimate frequency of occurrence for an SGTR with a coincident,multi-unit LOOP, it obtained a value (2.12E-6/yr) not much greater than the threshold in ANSguidance; further, when accounting for the risk that a CAC would be unavailable formaintenance for 30 days, the best-estimate frequency of occurrence fell below (1.75E-7/yr) theANS threshold. Reference 1 at Enclosure Page 7. Informal calculations by I&M incorporatingmore recent industry data on the frequency of multi-unit LOOPs provide more reason toconclude that a multi-unit LOOP is too remote an event to be considered in CNP's design basisSGTR analysis. According to Reference 17, there was not one reactor trip coincident with amulti-unit LOOP reported by the U.S. commercial nuclear power industry between 1986-2004.Reference 17 at Page 51. Using this data, I&M's informal calculation of the probability of anSGTR with a coincident, multi-unit LOOP yields a best-estimate frequency of occurrence of6.33E-7/yr -below the ANS threshold for consideration within CNP's UFSAR. Further, thebest-estimate frequency of occurrence is even lower (1.91 E-8) when accounting for the risk thata CAC would be unavailable for any reason, including maintenance.Further, although Regulatory Guide 1.70 states that the input parameters and initial conditionsfor each accident should be "clearly identified" within its analysis, the NRC Staff's contentionassumes that the assumptions regarding the potential scope of one UFSAR Section 14 analysis | |||
Enclosure 2 to AEP-NRC-2013-53Page 14(Loss of All AC Power to the Plant Auxiliaries) automatically carry over wholesale to subsequentaccident analyses (SGTR). Reference 15 at Page 15-5.Additionally, the NRC Staff's contention that its reading of the scope of the LOOP within UFSARSection 14.1.12 should apply to the LOOP assumed in CNP's Section 14.2.4 SGTR analysis.compares accidents with very different frequencies. The Loss of All AC Power to the PlantAuxiliaries is an ANS Condition II event, while the SGTR accident is a Condition III event.Reference 6 at Section 14.0. Further, because a dual-unit LOOP can be expected to occurmuch less frequently than a single-unit LOOP, application of the NRC Staff's reading of thescope of the term LOOP within CNP's SGTR analysis represents a significant change in theinitial assumptions and anticipated frequency for that particular accident. That revisedfrequency of CNP's design basis SGTR accident could conceivably require the assignment ofnew ANS Conditions to either the UFSAR Loss of All AC Power to the Plant Auxiliaries analysis(Reference 6 at Section 14.1.12), or its SGTR accident analysis (Reference 6 at Section14.2.4), which in turn would require the re-organization of CNP's UFSAR. Consequently, theNRC Staff's position does not account for the significance attached by NRC guidance to thedistinction between different ANS Conditions and (by extension) types of design basis events oraccidents.The NRC Staff's references to the use of the word "station" within the UFSAR's description ofCNP systems is similarly not helpful for determining the scope of the LOOP assumed in CNP'sSGTR accident analysis. In support of its contention that every use of the term LOOP refers toeither a single or multi-unit event, Reference 5 points to a handful of examples of the UFSAR'suse of the word "station" in descriptions of CNP Electrical System (at Section 1.3.7) and SafetyFeatures (at Section 1.3.8) that the NRC Staff understands to refer to both units at CNP.However, the NRC Staff nowhere explains why a handful of references to the word "station"within the system descriptions in Sections 1.3.7 and 1.3.8 define the use of that and otherterms (e.g., LOOP) throughout the UFSAR. Regulatory Guide 1.70 understood the systemdescriptions within the first section of a licensee's UFSAR to be distinct from the accidentanalyses described in a later section of the UFSAR:The first chapter of the SAR should present an introduction to the reportand a general description of the plant. This chapter should enable thereader to obtain a basic understanding of the overall facility withouthaving to refer to the subsequent chapters.Reference 15 at Page 1-1 (emphasis added). In contrast, the NRC Staff's position determinesthe meaning of ambiguous terms ("station", "LOOP") in the UFSAR's SGTR accident analysisassumptions not by reference to surrounding text, but by reference to language in an entirelydifferent UFSAR section. The NRC Staff's more fluid distinction between UFSAR sections isdifficult to reconcile with the approach endorsed within Regulatory Guide 1.70.Although the NRC Staff in Reference 1 states that the difference between UFSAR sectionsidentified above supports its understanding of CNP's licensing basis, the NRC Staffs position iserroneous. Conceding that high-level system descriptions within Section 1 of CNP's UFSAR donot prescribe accident analyses assumptions within subsequent UFSAR sections, the NRC Staffincorrectly asserts that: | |||
Enclosure 2 to AEP-NRC-2013-53Page 15This argument supports the inspectors' position that the licenseecannot take credit for the unaffected unit's non-safety-related PACunless explicitly approved by the NRC and described in the SGTRanalysis.Reference 1, Enclosure at Page 5 (emphasis added). Notwithstanding the fact the languagewithin Section 1 of CNP's UFSAR is unhelpful for interpreting language describing UFSARaccident analysis assumptions, it does not follow that Section l's high-level description of thecomponents comprising CNP systems would not control throughout the UFSAR. RegulatoryGuide 1.70 states that Section 1 of CNP's UFSAR exists precisely so that I&M would not haveto describe CNP systems and components multiple times. Reference 15 at Page 1-1. BecauseSection 1.3.9.h of CNP's UFSAR describes CNP's compressed air system as a shared systemof which both units' PACs and CACs are components, the NRC Staffs explicit endorsementwithin the SER in Reference 4 of the continued availability of motive force to the SG PORVsfrom CNP's control air appurtenances and equipment permits I&M to take credit for theunaffected unit's PAC in CNP's SGTR accident analysis. Further, by the NRC Staff's logic, I&Mwould not be able to take credit for the operation of any CAC or PAC within CNP's SGTRaccident analysis, as neither of those components is explicitly mentioned in the UFSAR's SGTRaccident analysis.Additionally, even if the NRC Staff's approach were appropriate, the cited examples of the term"station" within Section 1 of the UFSAR do not support its position. Reference 6 Section 1.3.7states:"The station auxiliary power system consists of auxiliary transformers,4160 v and 600 v switchgear, 600 v motor control centers, 120 v-a-c vitalinstrument buses and 250 v d-c buses."However, the NRC Staffs suggestion that the term "station" in this context necessarily refers toboth units at CNP is incorrect. Indeed, each unit at CNP has the components (redundantauxiliary transformers, multiple 600 v switchgear, independent 120 v-a-c vital instrument busesand 250 v-d-c buses, and 4160 v and 600 v switchgear) the NRC Staff suggests represents ashared system between CNP units. Similarly, both units have the EDGs and turbinesmentioned in the cited passage from UFSAR Section 1.3.8. Further, the NRC Staff's claim thatthe use of the term "station" within Section 1.3.8's description of CNP Safety Features provesthat there is only one, shared auxiliary power system at CNP is at odds with surrounding text notexamined by the NRC Staff. Specifically, UFSAR Section 1.3.9, "Shared Facilities andEquipment," begins by noting that:Separate and similar systems and equipment are provided for each unit,except as noted below.Reference 6 at Section 1.3.9 (emphasis added). The auxiliary power system is absent fromSection 1.3.9's list of shared systems and equipment.iii. The NRC Staff's Understanding of the Term LOOP Is at Odds with theReaulatorv History of CNP and Similarlv-Situated Facilities | |||
Enclosure 2 to AEP-NRC-2013-53Page 16The NRC Staff's understanding of the term LOOP also does not account for docketedcorrespondence acknowledging the retention of the assumptions within CNP's original SGTRaccident analysis. As explained at length earlier, the NRC Staff in 2001 reviewed and explicitlyapproved I&M's retention of CNP's original licensing basis assumptions for SGTR accidents,including the assumption of a single-unit LOOP only. Consequently, the NRC Staff'sunderstanding of the scope of the term LOOP assumed within CNP's SGTR accident analysisnot only re-writes CNP's UFSAR, but also re-writes nearly forty years' worth of pertinentdocketed correspondence.Further, as explained earlier, the NRC Staffs reading of the term LOOP within CNP's SGTRaccident analysis is also inconsistent with the regulatory history of CNP and other multi-unitfacilities of similar vintage. The two units at CNP were licensed and constructed on a staggeredschedule, with construction on Unit 1 beginning before Unit 2 such that Unit 1 received itsoperating license several years before Unit 2 (1974 as opposed to 1977). Consequently, theSGTR accident analysis within CNP's original licensing basis did not, as a practical matter,assume a multi-unit LOOP.Further, the CNP is not the only licensee that assumes only a single-unit LOOP within thedesign basis accident analyses for the units at its facility. I&M's informal polling of other multi-unit facilities licensed in approximately the same timeframe as CNP reveals that many of thoselicensees understand the licensing basis assumptions for units at their facility to assume only asingle-unit LOOP during SGTRs and other accidents. Further, among those licensees whoselicensing basis currently assumes multi-unit LOOPs were some who acknowledged that theircurrent licensing basis assumptions are a departure from original licensing basis assumptionsthat understood LOOPs to affect only a single unit at their facility.Lastly, the Commission's current regulations and guidance governing the availability of offsitepower reflect the unit-specific approach to electric system design within licensing basis accidentassumptions at CNP and other similarly-situated facilities. Most prominently, the current StationBlackout Rule at 10 CFR 50.63 (Reference 8) is unit-specific in its approach to the availability ofAC power, including offsite power. Although the NRC has recently published a Federal Registernotice (Reference 18 at 16179) indicating a desire to revise its Station Blackout Rule and otherregulations and guidance to adopt a facility-wide perspective on continuity of electrical power,interpreting the language within CNP's licensing basis against that proposed approach would bepremature, regardless of whether the NRC Staff can (as Reference 1 asserts) conceive ofscenarios in which plant configuration would make a multi-unit LOOP a credible event at CNP.6. The NRC Staffs Position Is Unnecessary for Assuring Adequate Protection Against EitherDesign Basis Events or Beyond-Design Basis External EventsNRC Orders issued following the earthquake and tsunami at the Fukushima Dai-ichi nuclearpower plant in March 2011 acknowledge that existing defense-in-depth approaches at licensedfacilities provide adequate protection of public health and safety against design basis accidents.Specifically, EA-12-049 states:To protect public health and safety... the NRC's defense-in-depthstrategy includes multiple layers of protection: (1) prevention of accidentsby virtue of the design, construction, and operation of the plant; (2) | |||
Enclosure 2 to AEP-NRC-2013-53Page 17mitigation features to prevent radioactive releases should an accidentoccur; and (3) emergency preparedness programs that include measuressuch as sheltering and evacuation .... These defense-in-depth featuresare embodied in the existing regulatory requirements and thereby provideadequate protection of the public health and safety.Reference 19 at Page 5 (emphasis added). Compliance with those NRC requirements, theNRC concluded, "presumptively assures adequate protection" of public health and safety frominadvertent release of radioactive materials during a design basis accident. Reference 19 atPages 4-5.As explained at length earlier, the NRC Staff's contention within Reference 1 that CNP is not incompliance with licensing basis requirements for a design basis SGTR accident is incorrect.CNP's licensing basis has never assumed that the LOOP coincident with a design basis SGTRaccident involves both units at CNP, and the NRC Staff has presented no meaningful evidencein support of a contrary position. Further, as recently as 2001, the NRC Staff endorsed themeasures (including the crediting of the continued availability of SG PORVs and supportingcompressed air system components) I&M employs for mitigating the risk of inadvertent releaseof radioactive materials during a design basis SGTR accident at CNP. Reference 4 concludesthat I&M's approach to mitigating the consequences of a design basis SGTR provides"reasonable assurance" of protection of public health and safety, and "will be conducted incompliance with the Commission's regulations. ... "Further, as noted earlier, I&M has supplemented the mitigation measures for SGTR accidentsevaluated within Reference 4 to provide additional defense-in-depth from design basis SGTRaccidents. Specifically, I&M in March 2013, completed installation of a plant modification andrevised CNP operating procedures to ensure that backup nitrogen tanks are immediately andautomatically available during an SGTR for operation of SG PORVs without the need for manualvalve manipulation outside the control room. I&M has also revised CNP Work Controlprocesses to provide additional defense-in-depth from a loss of control air pressure byrestricting removal for maintenance of the operating unit's CAC when the opposite unit isshutdown and the shutdown unit's PAC is aligned to preferred offsite power.In contrast, the NRC Staff has not demonstrated that its position would result in any meaningfulcontribution to adequate protection of public health and safety from design basis SGTRaccidents at CNP. As noted earlier, the most recent published industry data on the frequency ofLOOPs within Reference 17 indicates that the best-estimate frequency of occurrence for a multi-unit LOOP coincident with an SGTR would fall well below the minimal threshold within ANSguidance (Reference 16) for consideration within CNP's design basis. Moreover, the differencein core damage frequency from adopting the NRC Staff's position regarding the scope of theLOOP accompanying a design basis SGTR accident is so small (2.4E-8/yr) as to provide nomeaningful advantage over I&M's understanding of CNP's licensing basis for assuring adequateprotection of public health and safety. Reference 1, Enclosure at Page 1. Further, even thismarginal difference in core damage frequency between I&M's and the NRC Staff's positions islikely overstated, as the core damage frequency calculation within Reference 1 (Enclosure atPages 6-7) does not account for the additional defense-in-depth measures implemented at CNPsince the 2012 CDBI. | |||
Enclosure 2 to AEP-NRC-2013-53Page 18Lastly, the NRC Staff has provided no basis to conclude that I&M has failed to provide adequateprotection against beyond-design basis scenarios involving an SGTR accompanied by acoincident, multi-unit LOOP. As explained in Order EA-12-049, the events at FukushimaDai-ichi demonstrated the need for licensees to adopt additional defense-in-depth measures tomitigate the consequences of beyond-design basis external events, such as those resulting inthe extended loss of electrical power at multiple units at a facility. Reference 19 at Pages 4-6.Subsequent NRC guidance (Reference 20 at Page 4) endorsed licensees' use of the NuclearEnergy Institute's (NEI's) Diverse and Flexible Mitigation Capability (FLEX) strategy (Reference21) to satisfy Order EA-12-049's requirements for assuring adequate protection against beyond-design basis external events resulting in extended loss of electrical power (including offsitepower) at both units at a multi-unit facility. As required by Order EA-1 2-049, I&M has submittedan Overall Integrated Plan (Reference 22) for mitigation of beyond-design basis external eventsat CNP. I&M's Overall Integrated Plan incorporates the FLEX strategy endorsed by the NRCStaff in Reference 20 for use by licensees in satisfying the requirements within Order EA-12-049for mitigation measures providing adequate protection from beyond-design basis events suchas a multi-unit LOOP accompanying an SGTR.7. The NRC Staff's Determination that the NCVs Represent a More-than-Minor PerformanceDeficiency Involving Cross-Cutting Aspects Lacks MeritIn Reference 1, the NRC Staff contends that the NCVs represent a more-than-minorperformance deficiency involving cross-cutting areas of human performance, the component ofdecision making, and the aspect of conservative assumptions. Reference 1 Enclosure, atPages 1 and 2. The NRC Staff stated that the NCVs involved cross-cutting aspects becauseI&M's plant procedures assumed that the unaffected unit's compressed air system equipmentwould be available during an SGTR accident, despite the fact that the NRC Staff nowunderstands CNP's licensing basis to assume that an SGTR accident would be accompanied bya multi-unit LOOP. Reference 1 Enclosure, at Pages 1 and 2.The NRC Staff's conclusion that the NCVs involve cross-cutting aspects, however, incorrectlyassumes the validity of NCVs identified within Reference 1. As explained at length above, thoseNCVs are based on an erroneous understanding of the scope of the coincident LOOP withinCNP's design basis SGTR accident analysis: contrary to the NRC Staffs current position,CNP's licensing basis has only ever assumed a single-unit LOOP as an initial condition in anSGTR event. Consequently, the unaffected unit's PAC will remain available to provide controlair pressure to operate SG PORVs in the affected unit in the event of an SGTR event,regardless of the status of the CAC of the affected unit. Further, the NRC Staff in the 2001 SERwithin Reference 4 endorsed I&M's claims regarding the continued availability of control air tooperate an affected unit's SG PORVs during an SGTR accident, notwithstanding a coincidentLOOP. Because the NCVs within Reference 1 are incorrect, the NRC Staff's conclusion thatthose NCVs involve cross-cutting aspects is similarly incorrect.Additionally, even if the NRC Staff's current understanding of CNP's licensing basis werecorrect, the NCVs identified within Reference 1 would not involve cross-cutting aspects.Although Reference 1 (Enclosure, Page 7) criticizes I&M for not having adopted requirements,EOPs, and work control procedures positively demonstrating safety, the NRC Staff nowhereexplains how I&M's requirements were inconsistent with reactor safety and public health. Asnoted earlier, the NRC Staff concluded in the SER (Pages 3 to 5) within Reference 4 that the | |||
Enclosure 2 to AEP-NRC-2013-53Page 19changes to CNP's licensing basis proposed by I&M in its 2000 LAR would not increase the riskor consequences of an SGTR accident beyond the conservative estimates within CNP's originallicensing basis. In arriving at this conclusion, the NRC Staff explicitly noted that I&M hadrevised its EOPs for SGTR accidents to improve margin to steam generator overfill.Reference 4, SER at 4. Further, the core damage frequency data provided by the NRC Staff inReference 1 (Enclosure at Page 1) is consistent with the NRC Staffs conclusions withinReference 4, as the difference in core damage frequency from assuming a dual-unit LOOP isonly marginally different (2.4E-8/yr) from scenarios involving a single-unit LOOP.Further, the NRC Inspection Manual states that for an NCV to have cross-cutting aspects, theperformance deficiency at issue must be "recent (i.e., nominally within the last three years)."Reference 23, at Page 3. However, as explained at length above, the NCVs in Reference 1 arebased on an understanding of CNP's licensing basis that has been in place since the originallicensing of Unit 1 at CNP around forty years ago, and which was endorsed by the NRC Staff asrecently as 2001. Consequently, the NCVs within Reference 1 do not satisfy NRC InspectionManual standards for determining whether NCVs have cross-cutting aspects.Nor can the NRC Staff claim that I&M's failure to correct the longstanding performancedeficiency until recently is indicative of present performance. Although the NRC InspectionManual allows for a cross-cutting determination if "the performance deficiency occurred morethan three years ago, but the performance characteristic has not been corrected or eliminated",it severely limits the application of this exception to "some rare or unusual cases". Reference 23at Page 3. Reference 1 provides no justification for why the NCVs represent a "rare or unusualcase" warranting application of this exception. Further, as explained above, I&M'sunderstanding of its licensing basis is not rare or unusual; in fact, multiple plants of similarvintage and configuration have the same licensing basis assumptions regarding the scope of aLOOP during an SGTR or other accident.8. ConclusionFor the reasons identified above, both the NCVs identified within Reference 1 and the NRCStaff's determination that those NCVs involve cross-cutting aspects are incorrect. | |||
Enclosure 2 to AEP-NRC-2013-53Page 20REFERENCES:1. Letter from G. Shear, NRC, to L. J. Weber, I&M, "Donald C. Cook Nuclear Power Plant,Units 1 and 2, Component Design Basis Inspection 05000315/2013010;05000316/2013030," dated July 8, 2013.2. Letter from A. M. Stone, NRC, to L. J. Weber, I&M, "D. C. Cook Nuclear Power Plant,Units 1 and 2, Component Design Bases Inspection 05000315/2012007;05000316/2012007," dated January 11, 2013.3. Letter from W. Hodge, I&M, to C. Tilton, NRC, "D. C. Cook CDBI Response to Question2012-CDBI-298," dated November 15, 2012.4. Letter from J. F. Stang, NRC, to R. P. Powers, I&M, "Donald C. Cook Nuclear Plant,Units 1 and 2 -Issuance of Amendments (TAC Nos. MB0739 and MB0740)," datedOctober 24, 2001.5. Letter from K. O'Brien, NRC, to S. Bahadur, NRC, "Task Interface Agreement -Licensing Basis for Donald C. Cook Nuclear Power Plant, Units 1 and 2, During a SteamGenerator Tube Rupture Event Coincident with a Loss of Offsite Power (TIA 2012-11),"dated December 7, 2012.6. Donald C. Cook Nuclear Plant Updated Final Safety Analysis Report Rev. 24, datedMarch 17, 2012.7. Letter from I&M to Ann Marie Stone and Caroline Tilton, NRC, "Response to NRCInspection Report Issued January 11, 2013 Containing the Results of the ComponentDesign Basis Inspection Conducted Between July 23, 2012 and December 3, 2012,"dated February 8, 2013.8. 10 CFR 50.63, "Loss of All Alternating Current Power."9. Donald C. Cook Nuclear Plant Preliminary Safety Analysis Report for Units 1 and 2,dated December 18, 1967.10. Donald C. Cook Nuclear Plant Final Safety Analysis Report for Units 1 and 2, datedFebruary 2, 1971.11. Amendments to Donald C. Cook Nuclear Plant Final Safety Analysis Report for Units 1and 2, dated November 11, 1977.12. Amendments to the Donald C. Cook Nuclear Plant Final Safety Analysis Report for Units1 and 2, dated July 1997.13. Letter from R.P. Powers, I&M, to the NRC Document Control Desk, "Letter C1000-11,Donald C. Cook Nuclear Plant Units 1 and 2 License Amendment Request for Changesin Steam Generator Tube Rupture Analysis Methodology," dated October 24, 2000. | |||
Enclosure 2 to AEP-NRC-2013-53Page 2114. Letter from M. W. Rencheck, I&M, to the NRC Document Control Desk, "Letter C0601-21, Donald C. Cook Nuclear Plant Units 1 and 2 Response to Request for AdditionalInformation Regarding License Amendment for 'Changes in Steam Generator TubeRupture Analysis Methodology (TAC Nos. MB0739 and MB0740)," dated June 29, 2001.15. NRC Regulatory Guide 1.70, "Standard Format and Content of Safety Analysis Reportsfor Nuclear Power Plants, Rev. 3, " dated November 1978.16. American Nuclear Society, ANSI/ANS-51.1-1983, "Nuclear Safety Criteria for the Designof Stationary Pressurized Water Reactor Plants," dated 1983.17. NUREG/CR-6890, "Reevaluation of Station Blackout Risk and Nuclear Power Plants:Analysis of Loss of Offsite Power Events 1986-2004," dated December 2005.18. 77 Federal Register 16175, "NRC Advanced Notice of Proposed Rulemaking: StationBlackout," dated March 19, 2012.19. NRC Order Number EA-12-049, "Order Modifying Licenses with Regard toRequirements for Mitigation Strategies for Beyond-Design-Basis External Events," datedMarch 12, 2012.20. NRC Interim Staff Guidance JLD-ISG-2012-01, "Compliance with Order EA-12-049,Order Modifying Licenses with Regard to Requirements for Mitigation Strategies forBeyond-Design-Basis External Events, Rev. 0," dated August 29, 2012.21. NEI 12-06, "Diverse and Flexible Coping Strategies (FLEX) Implementation Guide, Rev.0," dated August 2012.22. Letter from J. P. Gebbie, I&M, to NRC, "Donald C. Cook Nuclear Plant Unit 1 and Unit 2Overall Integrated Plan In Response to March 12, 2012 Commission Order ModifyingLicenses with Regard to Requirements for Mitigation Strategies for Beyond-Design-Basis External Events (Order Number EA-12-049)," dated February 27, 2013.23. NRC Inspection Manual Chapter 0612, "Power Reactor Inspection Reports," datedJanuary 24, 2013 | |||
}} | }} |
Revision as of 14:26, 22 March 2018
ML13224A246 | |
Person / Time | |
---|---|
Site: | Cook ![]() |
Issue date: | 08/02/2013 |
From: | Gebbie J P Indiana Michigan Power Co |
To: | Document Control Desk, NRC/RGN-III |
References | |
AEP-NRC-2013-53 IR-13-010 | |
Download: ML13224A246 (25) | |
See also: IR 05000315/2013010
Text
INDIANAMICHIGANPOWERA unit of American Electric PowerAugust 2, 2013Docket Nos.: 50-31550-316Indiana Michigan PowerCook Nuclear PlantOne Cook PlaceBridgman, MI 49106Indiana Michigan Power.comAEP-NRC-2013-5310 CFR 2.201U.S. Nuclear Regulatory CommissionAttn: Document Control DeskWashington, DC, 20555-0001Donald C. Cook Nuclear Plant Units 1 and 2Response to the Non-Cited Violations Resulting from ComponentDesign Bases Inspection 05000315/2013010; 05000316/2013010References:1. Letter from W. Hodge, Indiana Michigan Power Company (I&M), to C. Tilton, U.S. NuclearRegulatory Commission (NRC), "D. C. Cook CDBI Response to Question 2012-CDBI-298,"dated November 15, 2012, (ADAMS Accession No. ML12320A544).2. Letter from K. O'Brien, NRC, to S. Bahadur, NRC, "Task Interface Agreement -LicensingBasis for Donald C. Cook Nuclear Power Plant, Units 1 and 2, During a Steam GeneratorTube Rupture Event Coincident with a Loss of Offsite Power (TIA 2012-11)," datedDecember 7, 2012, (ADAMS Accession No. ML13011A382).3. Letter from A. M. Stone, NRC, to L. J. Weber, I&M, "D. C. Cook Nuclear Power Plant, Units1 and 2, Component Design Bases Inspection 05000315/2012007; 05000316/2012007,"dated January 11, 2013 (ADAMS Accession No. ML13011A401).4. Letter from G. Shear, NRC, to L. J. Weber, I&M, "Donald C. Cook Nuclear Power Plant,Units 1 and 2, Component Design Bases Inspection 05000315/2013010;05000316/2013010," dated July 8, 2013, (ADAMS Accession No. ML13189A243).This letter provides Indiana Michigan Power Company's (l&M's),Nuclear Plant (CNP) Units 1 and 2, response contesting thedocumented by Reference 4, Component Design Bases05000315/2013010; 05000316/2013010.licensee for Donald C. CookNon-Cited Violations (NCVs)Inspection (CDBI) ReportIn Reference 1, I&M identified docketed correspondence supporting I&M's understanding of CNP'slicensing basis to assume only a single-unit loss of offsite power (LOOP) coincident with a designbasis Steam Generator Tube Rupture (SGTR) accident. In Reference 2, the Nuclear RegulatoryCommission (NRC) Region III Staff issued a Task Interface Agreement Report documenting
U.S. Nuclear Regulatory Commission AEP-NRC-2013-53Page 2the results of its consultation with the NRC Office of Nuclear Reactor Regulation regarding the NRCStaff's understanding of CNP's licensing basis to assume a multi-unit LOOP as an initial conditionof a design basis SGTR accident. In Reference 3, the NRC Staff notified I&M that two potentialfindings relating to the operability of steam generator power operated relief valves (SG PORVs)during a design basis SGTR accident identified by the NRC Staff during a CDBI performed at CNPbetween July 23, 2012, and December 31, 2012, would remain unresolved items (URIs) pendingthe NRC Staffs resolution of questions regarding the scope of a LOOP assumed within CNP'sSGTR accident analysis. In Reference 4, the NRC Staff resolved the URIs issued by Reference 3and issued NCVs of CNP Technical Specifications 5.4.1 (prescribing emergency operatingprocedures (EOPs) to mitigate the consequences of a design basis SGTR accident) and 3.7.4(governing the operability of SG PORVs). Reference 4 states that I&M had violated TechnicalSpecification 5.4.1 because CNP EOPs could not ensure that personnel would be able to operateSG PORVs as required by CNP's licensing basis during an SGTR accident accompanied by aLOOP affecting both units at CNP. Reference 4 also states that I&M had violated TechnicalSpecification 3.7.4 because it had failed on several occasions to declare the SG PORVsunavailable after taking a control air compressor out of service for maintenance. Reference 4characterized the NCVs as representing a more-than-minor performance deficiency with cross-cutting aspects.I&M contests the NCVs identified in Reference 4 because those NCVs lack technical justificationand are inconsistent with NRC regulations and guidance. Specific bases for I&M's contest of theNCVs include the following:* The NCVs are based on an erroneous understanding of CNP's licensing basis. Contrary tothe NCVs, CNP's licensing basis assumptions regarding the initial conditions for a SGTRaccident have never considered a coincident LOOP involving both units. Further, the NRCStaff's understanding of CNP's licensing basis underlying the NCVs does not acknowledgedocketed correspondence between I&M and NRC Staff supporting I&M's position, does notrepresent a fair reading of CNP's Updated Final Safety Analysis Report (UFSAR), and isinconsistent with the NRC's current regulatory position regarding the loss of offsite power tonon-safety related auxiliary systems at other multi-unit sites.* The NRC Staff has not demonstrated that I&M's understanding of CNP's licensing basis failsto provide adequate protection of public health and safety from either design basis events orbeyond-design basis external events. Further, the NRC Staff has not demonstrated that itsown position would provide a meaningful improvement in the protection of public health andsafety.* The NRC Staff's determination that the NCVs represent a more-than-minor performancedeficiency with cross-cutting aspects is based on an erroneous understanding of the scopeof a LOOP assumed within CNP's design basis SGTR accident analysis, is inconsistent withthe NRC Staffs statements in docketed correspondence, and is unrepresentative of presentlicensee performance.Enclosure 1 to this letter contains an affirmation statement. Enclosure 2 to this letter lays out indetail the regulatory and factual support for I&M's response contesting the NCVs.
U.S. Nuclear Regulatory Commission AEP-NRC-2013-53Page 3Regardless of the outcome of I&M's contest of the NCVs, I&M will continue to evaluate cost-effective measures for the improvement of safety margins against SGTR accidents.Following the 2012 CDBI, I&M revised CNP procedures and implemented plant modifications toprovide additional defense-in-depth and improved safety margins during an SGTR accident. InMarch 2013, I&M completed installation of a plant modification and revised CNP operatingprocedures to ensure that backup nitrogen tanks are immediately and automatically available duringan SGTR accident for operation of SG PORVs without the need for manual valve manipulationoutside the control room. I&M has also revised CNP Work Control processes to provide additionaldefense-in-depth from a loss of control air pressure by restricting removal for maintenance of theoperating unit's control air compressor when the opposite unit is shutdown and the shutdown unit'splant air compressor is aligned to preferred offsite power.This letter contains no new or revised commitments. If you have any questions, please contactMr. Michael K. Scarpello, Regulatory Affairs Manager, at (269) 466-2649.Sincerely,Joel P. GebbieSite Vice PresidentDMB/kmhEnclosures:1. Affirmation2. Indiana Michigan Power Company's Response to "Donald C. Cook Nuclear Power Plant,Units 1 and 2, Component Design Bases Inspection 05000315/2013010;05000316/2013010," dated July 8,2013c: C. A. Casto, NRC Region IIIJ.T. King, MPSCS. M. Krawec, AEP Ft. Wayne, w/o enclosureE. Leeds, NRC NRRMDEQ-RMD/RPSNRC Resident InspectorA. M. Stone, NRC Region IIIC. Tilton, NRC Region IIIT. J. Wengert, NRC Washington, DCR.P. Zimmerman, NRC Washington, DC
ENCLOSURE I TO AEP-NRC-2013-53AFFI RMATIONI, Joel P. Gebbie, being duly sworn, state that I am Site Vice President of Indiana Michigan PowerCompany (I&M), that I am authorized to sign and file this request with the Nuclear RegulatoryCommission on behalf of I&M, and that the statements made and the matters set forth hereinpertaining to I&M are true and correct to the best of my knowledge, information, and belief.Indiana Michigan Power CompanyJoel P. GebbieSite Vice PresidentSWORN TO AND SUBSCRIBED BEFORE METHIS____ DAY OF ,A)ws 2013My Commission Expires ( I 2 IO{
ENCLOSURE 2 TO AEP-NRC-2013-53Indiana Michigan Power Company's Response to "Donald C. Cook Nuclear PowerPlant, Units 1 and 2, Component Design Bases Inspection 05000315/2013010;05000316/2013010," dated July 8, 20131. IntroductionThe Non-Cited Violations (NCVs) within the Nuclear Regulatory Commission (NRC) StaffsJuly 8, 2013, letter (Reference 1) to Indiana Michigan Power Company (I&M) are based on anerroneous understanding of the licensing basis of Donald C. Cook Nuclear Plant (CNP). TheNRC Staff's position that CNP's design basis Steam Generator Tube Rupture (SGTR) accidentassumes a coincident loss of offsite power (LOOP) that can involve both units at CNP isinconsistent with pertinent, docketed correspondence between the NRC Staff and I&M. Further,the NRC Staff's position is unsupported by a fair reading of CNP's Updated Final SafetyAnalysis Report (UFSAR), and is likewise inconsistent with relevant historical and currentregulatory positions of the NRC. Additionally, the NRC Staff has not demonstrated that I&M'sunderstanding of CNP's licensing basis fails to provide adequate protection of public health andsafety from either design basis events or beyond-design basis external events. Lastly, the NRCStaff's determination that the NCVs represent a more-than-minor performance deficiency withcross-cutting aspects relies on an erroneous understanding of the scope of a LOOP assumedwithin CNP's design basis SGTR accident analysis, is inconsistent with the NRC Staff'sstatements in docketed correspondence, and is unrepresentative of present licenseeperformance.Documents referenced herein are listed as references at the end of this Enclosure.2. History of the Non-Cited ViolationsThe NCVs contested by I&M result from findings by the NRC Staff during the ComponentDesign Bases Inspection (CDBI) conducted at CNP between July 23, 2012, andDecember 31, 2012. As described in Reference 2, the CDBI entailed a review of licensing basisdocumentation and drawings of the CNP compressed air system to verify that support functionsprovided to the steam generator power operated relief valves (SG PORVs) were consistent withCNP's licensing basis requirements for SGTR accidents.As stated in Reference 2, the NRC Staff contended during the CDBI that CNP was not inconformance with Technical Specifications 5.4.1 (prescribing emergency operating procedures(EOPs) to mitigate the consequences of a design basis SGTR accident) and 3.7.4 (governingthe operability of SG PORVs). Based on its belief that CNP's licensing basis assumptions for aSGTR accident included a coincident LOOP affecting both units at CNP, the NRC Staffreasoned that the only available source of control air pressure during the most limiting SGTRaccident would be the affected unit's dedicated control air compressor (CAC) receiving powerfrom one of the two emergency diesel generators (EDG). However, if the affected unit's CACwere unavailable as a result of emergent or planned maintenance, then the NRC Staff reasonedthat control air pressure would be unavailable to operate the affected unit's SG PORVs. Inreviewing CNP operating records, the NRC Staff identified several occasions in which CACs at
Enclosure 2 to AEP-NRC-2013-53Page 2CNP would have been unavailable due to maintenance, but I&M had not declared the SGPORVs inoperable.I&M disagreed with the NRC Staff's characterization of CNP's licensing basis assumptions for aSGTR event. Noting that the CNP licensing basis for an SGTR event did not consider acoincident multi-unit LOOP, I&M contended that the NRC Staffs finding was based on a beyonddesign basis accident scenario. The NRC Staff requested assistance from the NRC Office ofNuclear Reactor Regulation (NRR) in resolving the disagreement regarding CNP's licensingbasis assumptions. On November 15, 2012, I&M submitted Reference 3 to NRC Staff,containing information identifying the technical and regulatory bases supporting I&M's positionand providing docketed correspondence. Reference 3 in particular identified a SafetyEvaluation Report (SER, Reference 4) dated October 24, 2001, explicitly discussing CNP'sassumptions for SGTR accident initial conditions, and revealing the NRC Staff's evaluation andendorsement of I&M's understanding of the CNP licensing basis assumptions for an SGTRaccident.On December 7, 2012, NRC Region III Staff issued Reference 5 after consulting with NRR,contradicting I&M's understanding of CNP's licensing basis assumptions for SGTR accidents.Reference 5 cited only three passages within CNP's UFSAR (Reference 6) in support of itsposition, interpreting a handful of references to the terms "LOOP" and "station" in descriptions ofCNP electrical systems to mean that CNP's licensing basis assumed a LOOP would affect bothunits at CNP in an SGTR accident. Reference 5 suggests that it did not examine the technicaland regulatory bases and docketed correspondence supporting a contrary position referencedwithin Reference 3 submitted by I&M.On January 11, 2013, the NRC Staff issued Reference 2, identifying the CDBI findings at issueas unresolved items (URIs) pending submission of additional information from I&M regardingCNP's licensing basis assumptions for SGTR accidents. Reference 2 repeated Reference 5'sconclusions regarding CNP's licensing basis assumptions for SGTR accidents without furtherexplanation or analysis; further, Reference 2 again did not address the technical and regulatorybases and docketed correspondence identified in Reference 3 forwarded by I&M. OnFebruary 8, 2013, I&M provided Reference 7 to the NRC Staff, refuting Reference 5'sinterpretation of CNP's UFSAR and providing additional detail regarding the technical andregulatory bases supporting I&M's understanding of the CNP licensing basis assumptions for anSGTR accident. During a May 20, 2013, technical debrief of the CDBI findings, the NRC Staffrepeated its understanding of the scope of the LOOP assumed within SGTR's accident analysis,again without addressing the technical and regulatory bases and docketed correspondencesupporting I&M's position. In a re-exit teleconference for the URIs conducted on May 24, 2013,the NRC Staff informed I&M that the NRC Staff planned to issue an NCV for violation ofTechnical Specification 3.7.4 requirements regarding the operability of SG PORVs.On July 8, 2013, the NRC Staff issued Reference 1. In Reference 1, the NRC Staff identifiedNCVs of CNP Technical Specifications 5.4.1 (prescribing EOPs to mitigate the consequences ofa design basis SGTR accident) and 3.7.4 (governing the operability of SG PORVs). Reference1 states that I&M had violated Technical Specification 5.4.1 because CNP EOPs could notensure that personnel would be able to operate SG PORVs as required by CNP's licensingbasis during an SGTR accident accompanied by a LOOP affecting both units at CNP.Reference 1 also states that I&M had violated Technical Specification 3.7.4 because it had
Enclosure 2 to AEP-NRC-2013-53Page 3failed on several occasions to declare the SG PORVs unavailable after taking a CAC out ofservice for maintenance.Reference 1 characterized the NCVs as representing a more-than-minor, cross-cuttingperformance deficiency involving areas of human performance, the component ofdecisionmaking, and the aspect of conservative assumptions because I&M had incorrectlyassumed that control air pressure to the SG PORVs of a unit experiencing an SGTR accidentaccompanied by a LOOP would remain available from the unaffected unit's plant air compressor(PAC).Reference 1 also attempted to refute I&M's explanation within Reference 7 of its understandingof CNP's licensing basis assumptions for SGTR accidents. Acknowledging I&M's position thatCNP's licensing basis did not assume a single failure of a non-safety-related component (inparticular, the unaffected unit's PAC), during an SGTR event, Reference 1 contends that I&Mhad nevertheless failed to demonstrate that control air would reasonably be available during anSGTR event accompanied by a multi-unit LOOP. Similarly, Reference 1 asserts that even if theunaffected unit's PAC would be available during a design basis SGTR accident, I&M had failedto identify that assumption within its SGTR accident analysis, and the NRC Staff had neverexplicitly approved that assumption. Further, Reference 1 endorsed Reference 5'sinterpretation of the UFSAR's use of the term LOOP to refer to multi-unit events, adding that theabsence of CNP operating procedures preventing alignment of the same offsite power sourcesto both units made a multi-unit LOOP a credible event within CNP's licensing basis.3. Overview of Pertinent CNP Systems and Operatinq Proceduresa. CNP Steam Generator Power Operated Relief ValvesIn accordance with Reference 6 (at Sections 10.2.2 and 14.2.4), the SG PORVs preventoverpressure conditions in the steam generators by releasing secondary system steam toatmosphere following a loss of condenser vacuum. The SG PORVs form part of the mainsteam system pressure boundary, and thus are safety-related equipment for main steam systempressure retention.CNP operating procedures prescribe operator actions in the event of a SGTR accident. CNPoperating procedures allow SG PORVs to be operated using motive force provided by controlair supplied by either the compressed air system shared between the two units, control airpressure supplied by a unit-specific CAC, or installed backup nitrogen tanks that can be alignedto the SG PORVs. In March 2013, I&M completed installation of a plant modification andrevised its operating procedures to ensure that the backup nitrogen tanks are immediately andautomatically available during an SGTR accident without the need for manual valvemanipulation outside the control room.b. CNP Compressed Air SystemSection 9.8.2 of Reference 6 describes the control air provided by CNP's compressed airsystem as the ordinary source of motive force for operation of SG PORVs for both units at CNP.Per Reference 6, Section 1.3.9.h, CNP's compressed air system is a single system sharedbetween both units at CNP. Each unit at CNP contains one CAC capable of providing control
Enclosure 2 to AEP-NRC-2013-53Page 4air only within that unit, as well as a PAC capable of providing control air to both units via ashared header. Both units share a single backup air compressor capable of providing control airto loads within either unit.During normal operations, control air pressure for operating both units' SG PORVs is providedby one of the two PACs. Low pressure in the shared plant compressed air header will result inthe automatic start and loading of the other unit's PAC. Low control air header pressure in oneof the unit-specific control air headers will cause that unit's CAC to start.During normal operations, the operating PAC receives power from its unit's auxiliarytransformers, which are in turn powered by that unit's main generator or preferred offsite powertransformers. The CAC associated with each unit at CNP can be powered by either offsitepower source in normal operations, but can only receive power from its unit's CD EDG afteroffsite power has been lost to that unit. The CACs and PACs are both non-safety relatedequipment governed by the Maintenance Rule at 10 CFR 50.65.CNP Work Control processes impose a series of administrative controls to maximize availabilityof control air pressure when a CAC or PAC is taken out of service for maintenance:* In the event a CAC is taken out of service for maintenance, bothPACs and the installed backup nitrogen tanks must be guarded; and* In the event that a PAC is taken out of service, the followingequipment is guarded: (1) the opposite unit's PAC, (2) both CACs, (3)the opposite unit's CD EDG, and (4) the backup air compressor.Following the 2012 CDBI, I&M revised CNP Work Control processes to provide additionaldefense-in-depth from a loss of control air pressure by restricting removal for maintenance ofthe operating unit's CAC when the opposite unit is shutdown and the shutdown unit's PAC isaligned to preferred offsite power.4. Regulatory Basis for the Assumption of Only a Single-Unit LOOP within CNP's SGTRAccident Analysisa. CNP's Licensing Basis Has from the Beginning Assumed that an SGTR AccidentWould Involve a Coincident, Single-Unit LOOPCNP's original licensing basis explicitly assumed that SG PORVs would remain availablethroughout an SGTR accident. As described in the Preliminary Safety Analysis Report (PSAR,Reference 9) for Units 1 and 2 submitted on December 18, 1967, and repeated in Sections14.2.4 and 14.2.7 of the FSAR for Units 1 and 2 dated February 2, 1971 (Reference 10), CNP'soriginal licensing basis evaluated the radiological consequences of an SGTR accident byconservatively estimating the mass release of radioactivity to the environment over the30-minute time span between SGTR accident initiation and subsequent termination of primaryto secondary mass transfer from the completion of mitigation measures taken by operators.I&M's analytical assumption of 30 minutes' mass release before termination of the event wasconsidered inherently conservative because it neglected the reduction in mass flow that wouldoccur during this same time period.
Enclosure 2 to AEP-NRC-2013-53Page 5Inherent in that postulated 30-minute mass release was an assumption of the success ofoperator actions such as the operation of SG PORVs to mitigate the event. Section 14.2.4 ofReference 10 in several places explicitly credited the availability of SG PORVs during a designbasis SGTR regardless of conditions.Reference 10's evaluation of SGTR accidents omits any mention of the possibility thatcompressed air system components could be unavailable as a result of a single failure ormaintenance, as it prefaced its elaboration of the sequence of events initiated by an SGTRevent by stating that its analysis had "assum[ed] normal operation of the various plant controlsystems ....... Reference 10 at Section 14.2.4. Further, Reference 10 assumed that SGPORVs would remain available regardless of the status of offsite power, stating that when a unitwas "without offsite power":Condenser bypass valves will automatically close and the steamgenerator pressure will rapidly increase resulting in steam discharge tothe atmosphere through the steam generator safety valves and/or thepower operated relief valves.Reference 10 at Section 14.2.4. Elsewhere, Reference 10 noted that:In the event of a co-incident station blackout, the steam dump valveswould automatically close to protect the condenser. The steam generatorpressure would rapidly increase resulting in steam discharge to theatmosphere through the steam generator safety and/or power operatedrelief valves.Reference 10 at Section 14.2.4 (emphasis added).I&M's assumption that SG PORVs remained available for mitigation of an SGTR accident isconsistent with the description of the compressed air system elsewhere within CNP's originalFSAR. Among the design bases for CNP's compressed air system within Reference 10 is arequirement for continued availability of control air:The [compressed air system] must provide a continuous supply ofcompressed air to vital systems under both normal and abnormalconditions.Reference 10 at Section 9.8.2 (emphasis added). With this in mind, each of CNP's PACs weredesigned to be "capable of supplying the entire demand of both plant and control-instrument airrequirements for both units," as the offline PAC automatically started on low pressure in the(shared) plant air header. Reference 10 at Section 9.8.2.3.Although CNP's original FSAR accounted for the availability of compressed air systemcomponents within the opposite plant, the staggered construction and licensing of CNP Units 1and 2 resulted in a more unit-specific design and function for other CNP systems. For example,Unit l's construction and licensing (1974) several years before Unit 2 (1977) meant that thedesign bases of the electrical systems for each of the two units at CNP were, as a practicalmatter, unit-specific. For example, although each EDG shares a fuel oil tank with an EDG in the
Enclosure 2 to AEP-NRC-2013-53Page 6other unit, the fuel oil tank's capacity is based on the design operational requirements of asingle EDG. Reference 6 at Section 8.4. Consequently, references within Reference 10'sSGTR accident analysis to a "loss of offsite power" or a "station blackout" referred to an eventinvolving only a single unit.The analysis of a design basis SGTR accident in the revised FSAR evaluating Unit 2 as-built(Reference 11) used nearly identical language to that used within the SGTR accident analysis inthe original Units 1 and 2 FSAR (Reference 10). Further, subsequent versions of both units'UFSAR analyses for SGTR accidents retained the CNP's original assumptions regarding theavailability of SG PORVs -and, in fact, arguably placed even greater emphasis on thecontinued availability of those components in their SGTR accident analysis. In particular,July 1997 revisions to the UFSAR for both units were revised to better track CNP EOPsidentifying the SG PORVs (and not the steam generator safety valves) as the initial means ofpreventing steam generator overpressure after loss of offsite power:In the event of a coincident station blackout, the steam dump valveswould automatically close to protect the condenser. The steam generatorpressure would rapidly increase, resulting in steam discharge to theatmosphere through the steam generator power operated relief valves(and the steam generator safety valves if their setpoint had beenreached).Reference 12 at Section 14.2.4 (emphasis added). Later UFSAR revisions to CNP's SGTRaccident analysis also incorporated the original FSAR's language describing the continuedavailability of SG PORVs despite a LOOP or station blackout virtually unchanged. Reference 6at Section 14.2.4. Further, I&M's review of pertinent docketed correspondence with the NRCStaff has discovered no evidence of a departure from CNP's original assumption of a unit-specific LOOP coincident with an SGTR accident.b. The NRC Staff Has Reviewed and Endorsed CNP's Design Basis Assumptions forSGTR Accidents in Docketed CorrespondenceOn October 24, 2000, I&M submitted a license amendment request (LAR, Reference 10) torevise the methodology used in designing CNP EOPs during a design basis SGTR accident.The Westinghouse Owners Group methodology (WCAP-10698-P-A ("SGTR AnalysisMethodology to Determine Margin to Steam Generator Overfill")) that I&M proposed to adapt foruse within its SGTR accident analysis incorporated lessons learned from operationalexperience, plant simulator studies, and advances in computer modeling techniques to bettercharacterize steam generator fill conditions during an SGTR accident. Of particular importanceto CNP was that the LOFTTR2 computer program used in the WCAP-10698-P-A methodologysimulated the effects of operator actions on margin to steam generator overfill during an SGTRaccident. By incorporating elements of the WCAP-10698-P-A methodology for the simplifiedcalculations of margin to steam generator overfill within its original SGTR accident analysisassumptions, I&M could revise CNP EOPs to assure margins to steam generator overfill whileremaining within the conservative margins to radiological consequences described in its originalSGTR accident analysis.
Enclosure 2 to AEP-NRC-2013-53Page 7Although the NRC had previously accepted WCAP-10698-P-A for use by licensees, the NRCStaff had to evaluate its application within CNP's SGTR accident analysis. In a series ofdocketed correspondence with the NRC Staff detailing how the WCAP-10698-P-A would beused within CNP's SGTR accident analysis, I&M repeatedly emphasized that the newmethodology would not disturb existing license basis assumptions in its SGTR accidentanalysis. Specifically, the safety analysis for I&M's LAR noted that:The proposed change ...does not affect any accident initiators orprecursors .... The proposed change also does not affect the ability ofoperators to mitigate the consequences of an accident.Reference 13, Attachment 1 at Page 4 (emphasis added). I&M repeated this claim in the LAR'sevaluation of significant hazards required by 10 CFR 50.92(c):[T]he new methodology does not affect equipment malfunctionprobability .... The proposed change does not impact the design ofaffected plant systems, involve a physical alteration to the systems, orchange the way in which systems are currently operated, such thatpreviously unanalyzed SGTRs would not occur. The change toincorporate the WCAP-10698-P-A methodology does not introduce anynew malfunctions ....Reference 13, Attachment 2 at Pages 2-3 (emphasis added).Subsequent docketed correspondence between I&M and the NRC Staff was even more explicitin describing the retention of existing license basis assumptions for SGTR accidents. In aJune 29, 2001, response (Reference 14) to a May 7, 2001, letter from the NRC Staff requestingadditional information (RAI) regarding how I&M intended to use the WCAP-10698-P-A within itsSGTR accident analysis, I&M emphasized that its use of the WCAP-10698-P-A methodologywas "limited", and that, by-and-large, "CNP's present methodology would be retained forcalculating the radiological consequences of the postulated SGTR .... ." Reference 14,Attachment 1 at Page 1. In particular, I&M noted that its analysis retained existing licensingbasis assumptions regarding the availability of certain systems, components, and instruments(listed in a table within Reference 14) credited for accident mitigation in an SGTR. Among theitems listed in that table were the "air-operated" SG PORVs, which the notes accompanying thetable stated were themselves safety-grade components because they "form part of the mainsteam system pressure boundary upstream of the SG stop valves," even though their "electricaland control air appurtenances [were] not safety-grade." Reference 14, Attachment 1 at Pages3-4. Reference 14 also noted that I&M's limited use of the WCAP-10698-P-A methodologywould not disturb CNP's existing licensing basis assumption that an SGTR accident would notinvolve a single failure. Reference 14, Attachment 1 at Page 6.Reference 14 also communicated I&M's intention to retain CNP's existing assumptionsregarding the availability of offsite power. Acknowledging that the WCAP-10698-P-Amethodology assumes that "the most challenging SGTR scenario with respect to SG fill includesa coincident loss of offsite power", Reference 14 noted that the modified SGTR analysis wouldretain CNP's original licensing assumption that SG PORVs would remain available despite thefact that "offsite power [was] not ...available." Reference 14, Attachment 1 at Page 4.
Enclosure 2 to AEP-NRC-2013-53Page 8Reference 14 contained no suggestion of a change in the scope of the LOOP assumed withinCNP's SGTR accident analysis.By letter dated October 24, 2001 (Reference 4), the NRC Staff approved I&M's LAR in modifiedform to accommodate CNP's existing licensing basis assumptions for SGTR accidents. In theSER submitted with its approval of I&M's LAR, the NRC Staff acknowledged that licensees likeI&M could not incorporate the WCAP-10698-P-A methodology within their SGTR accidentanalysis in a uniform fashion because "variations in plant designs prevent a single model fromadequately representing all Westinghouse Plants." Reference 4, SER at Page 2.Consequently, the NRC Staff devoted much of the SER to evaluating the differences betweenthe generic WCAP-1 0698-P-A methodology and I&M's proposed approach for incorporating thatmethodology within its licensing basis.The NRC Staff noted that in the immediate case, those differences included I&M's intention ofretaining CNP's existing assumptions for SGTR accidents:To implement the WCAP, the licensee used the LOFTTR2 computer codeand the plant-specific current licensing basis assumptions.Reference 4, SER at Page 2 (emphasis added). The NRC Staff explicitly acknowledged thatCNP's licensing basis assumptions credited certain systems and components, including the SGPORVs and their control air appurtenances, as remaining available for mitigation of an SGTRaccident:The licensee provided a list of systems, components, and instrumentationthat are used for SGTR accident mitigation. They also specified thesafety classification of the systems and power sources. However, thelicensee listed several systems used for SGTR mitigation that are notsafety related and do not have safety related backups. The licenseejustified the use of the non-safety-related equipment by stating that thesesystems are credited in the current UFSAR Section 14.2.4 accidentanalysis. Upon review of Section 14.2.4, the staff concludes that thelicensing basis SGTR analysis does credit limited use of non-safety gradeequipment for mitigating the SGTR.Reference 4, SER at Page 3. Similarly, the NRC Staff acknowledged that CNP's licensing basisdid not assume a worst single failure during an SGTR accident as the WCAP-10698-P-Amethodology did:[T]he licensee did not assume the worst single failure as prescribed bythe WCAP-10698-P-A safety analysis, and did not provide it's [sic] effecton the margin to overfill. The licensee based their decision not to assumethe worst single failure on the fact that their current licensing basis doesnot include a single failure.Reference 4, SER at Page 4. Further, the SER nowhere mentions that I&M intended to discardCNP's existing assumption of a coincident single-unit LOOP during an SGTR accident, or that
Enclosure 2 to AEP-NRC-2013-53Page 9the LOOP assumed within the WCAP-10698-P-A methodology supplanted CNP's existinglicensing basis assumptions for SGTR accidents.Although I&M's proposed retention of CNP's existing licensing basis assumptions for SGTRaccidents "varied significantly" from the assumptions underlying the WCAP-10698-P-Amethodology, the NRC Staff approved I&M's use of some elements of the WCAP-10698-P-Amethodology identified in the LAR and related correspondence:[T]he NRC staff concludes that the licensee can incorporate theLOFTTR2 code into its licensing bases for CNP and can use theLOFTTR2 code, with the current licensing basis assumptions as inputs forthe overfill analysis of steam generator tube rupture accidents. Thischange to the licensing basis does not affect accident initiators orprecursors. This change also does not ...decrease the ability of theoperators to mitigate the consequences of an accident.Reference 4, SER at Page 5 (emphasis added). In justifying its approval of a modifiedWCAP-10698-P-A methodology for use at CNP, the NRC Staff noted that I&M's adaptation ofthe WCAP-10698-P-A methodology to CNP's existing licensing basis assumptions for SGTRaccidents did not affect conservative estimates of the radiological consequences of a designbasis SGTR at CNP. Reference 4, SER at Page 3.I&M's subsequent review of docketed correspondence with the NRC Staff has identified nofurther changes to CNP's licensing basis assumptions regarding the availability of SG PORVs inan SGTR accident, the absence of a single failure assumption within CNP's SGTR accidentanalysis, or the scope of a LOOP assumed in the SGTR analysis.5. The NRC Staff's Understanding of CNP's Licensing Basis Assumptions for SGTR AccidentsDoes Not Address Pertinent Docketed Correspondence, Is Unsupported by a Fair Readingof the UFSAR, and is Inconsistent with the NRC's Historical and Current RegulatoryPositionsa. The NRC Staff's Reading of CNP's Licensing Basis Assumptions for SGTRAccidents Does Not Address Pertinent Docketed CorrespondenceAs noted earlier, the NCVs within Reference 1 are based on the NRC Staffs contention that thecoincident LOOP assumed within CNP's licensing basis SGTR accident analysis involves a lossof offsite power to both units at CNP. The NRC Staff's position is based on a single argumentwithin Reference 5: that it follows from the use of the terms "LOOP" and "station" in a handful ofCNP UFSAR sections, some of which are unrelated to SGTR accident analysis, that a LOOPcan refer to the denial of offsite power to one or both units at CNP.In support of this argument, Reference 5 advances only a handful of UFSAR passages. Thefirst UFSAR passage referenced in Reference 5 comes from Section 1.3.7 describing theauxiliary electrical system for each of the two units at CNP:Donald C. Cook's UFSAR Section 1.3.7, "Electrical System" states, "Themain generators are 1800 rpm, Phase III, 60 cycle, hydrogen and water
Enclosure 2 to AEP-NRC-2013-53Page 10cooled units. The main transformers deliver generator power to the345kV and 765 kV switchyards. The station auxiliary power systemconsists of auxiliary transformers, 4160V and 600 V switchgear, 600Vmotor control centers, 120 V A-C vital instrument buses and 250 V D-Cbuses."Reference 5 at Page 3 (emphasis supplied by NRC Staff). Based on the fact that UFSARSection 1.3.7 described the identical electrical systems for both units, Reference 5 concludedthat the UFSAR passage's reference to "station" must refer to both units at CNP, rather than toeach unit individually. In the same vein, Reference 5 cites a passage from Section 1.3.8 of theUFSAR describing the Safety Features associated with each unit at CNP:Also, Section 1.3.8, "Safety Features," describes the safety featuresincorporated into the design of the plant, including the fact that "even ifexternal auxiliary power to the station is lost concurrent with an accident,power is available for the engineered safeguards from on-site dieselgenerator power to assure protection of the public health and safety forany loss of coolant accident."Reference 5 at Page 3 (emphasis supplied by NRC Staff). Here, too, Reference 5 concludesthe fact that Section 1.3.8 describes identical safety features at each unit means that thepassage's reference to "station" must refer to both units at CNP, rather than only one unit.Lastly, Reference 5 points to language within a passage from the accident analysis (atSection 14.1.12) for "Loss of All AC Power to the Plant Auxiliaries" at Unit 1:"A complete loss of all (non-emergency) AC Power (e.g., offsite power)may result in the loss of all power to the plant auxiliaries, i.e., the RCPs,condensate pumps, etc. The loss of power may be caused by a completeloss of the offsite grid accompanied by a turbine trip at the station, or by aloss of the on-site AC distribution system."Reference 5 at Page 4. The NRC Staff read this reference to a "complete loss of offsite gridaccompanied by a turbine trip at the station" associated with the design basis event postulatedwithin Section 14.1.12 to mean that a LOOP affecting both units is within CNP's licensing basisfor every event evaluated in UFSAR Section 14. Reference 5 at Page 4. Based on theseexamples, Reference 5 reports that NRR concurred with NRC Staff that had performed theCDBI that the LOOP assumed in CNP's SGTR analysis was a "station event, not a unit specificevent." Reference 5 at Page 4.The NRC Staff's position and the UFSAR passages described above represent the only basisidentified by the NRC Staff for its position throughout the multiple docketed communications andmeetings with I&M since the CDBI began in July 2012. The NRC Staff has identified noregulatory provisions or policy guidance requiring the assumption of a LOOP affecting both unitsfor a design basis SGTR accident. The NRC Staff has advanced no docketed correspondencein support of its understanding of CNP's licensing basis for SGTR accidents, and has identifiedno additional passages within CNP's UFSAR supporting its position.
Enclosure 2 to AEP-NRC-2013-53Page 11Further, the NRC Staff has yet to provide a meaningful response to the analysis provided byI&M in References 3 and 7 in support of its understanding of CNP's licensing basisassumptions. Reference 5 does not specifically address the SGTR accident analysisassumptions identified within docketed correspondence highlighted within Reference 3:The scope of this TIA was limited to the licensing basis as related tooffsite power only. The staff did not evaluate other assertions in thelicensee's white paper.Reference 5 at Page 4.1 Reference 2 merely repeated Reference 5's claims regarding CNP'slicensing basis, rather than address the detailed licensing basis interpretation within Reference7 provided by I&M.Further, although Reference 1 suggests that it addresses the understanding of CNP's SGTRaccident licensing basis assumptions advanced by I&M in References 3 and 7, a careful readingof the bases identified in Reference 1 indicates that the NRC Staff's reasoning is circular in thatit depends on, rather than proves the assumption of a multi-unit LOOP in CNP's SGTR accidentanalysis. Specifically, in acknowledging I&M's position that CNP's licensing basis had neverassumed a single failure of a non-safety-related component (specifically the unaffected unit'sPAC) during an SGTR event, Reference 1 contends that I&M had nevertheless failed todemonstrate that an unaffected unit's PAC would reasonably be available during an SGTRaccident affecting one unit:The inspectors agreed that certain older operating plants arecredited with the use of non-safety related equipment to mitigateevents. In these cases, the licensee was required to demonstratethe non-safety-related equipment would reasonably be availableand use of the equipment was bound by a safety-related path.Reference 1, Enclosure at Pages 4 and 5. Similarly, the NRC Staff in Reference 1 agrees withI&M's observation in Reference 7 that the original SER for Unit 1 did not consider that a CACwould be out of service for maintenance pursuant to an assumed single failure, claiming thatthis demonstrates that a CAC would have to be available to supply control air pressure during adesign basis SGTR accident, as its availability would be a limiting condition in CNP's SGTRaccident analysis.However, the above arguments do not prove the NRC's Staff understanding of the scope of theLOOP assumed in CNP's SGTR accident analysis. Because the unaffected unit's non-safety-related PAC would remain available during a single-unit LOOP, control air pressure would bereasonably available and bounded by a safety-related path for main steam system pressureretention purposes, regardless of the status of the CAC on the affected unit. Similarly, theavailability of the affected unit's CAC is not a limiting condition for CNP's SGTR accidentanalysis if the coincident LOOP affects only the unit experiencing the SGTR event such that the1 The NRC Staff has not docketed correspondence between Region III personnel and NRRpersonnel defining the scope of NRR personnel's review of the competing interpretations ofCNP's licensing basis assumptions for the LOOP assumed within CNP's SGTR design basisaccident analysis.
Enclosure 2 to AEP-NRC-2013-53Page 12PAC on the unaffected unit remains available to provide control air pressure to the affectedunit's SG PORVs. Lastly, the NRC Staff statement quoted above is inconsistent with the NRCStaff's statements within Reference 4 endorsing CNP licensing basis assumptions crediting theavailability of SG PORVs and compressed air system components during an SGTR accident.b. The NRC Staff's Position Is Unsupported by a Fair Reading of the UFSARThe NRC Staff's categorical statement that every reference to a LOOP within CNP's UFSARcan be understood to refer to an event denying offsite power to one or both units at CNP isunsupported by a careful reading of that document. The UFSAR contains no generic,controlling definition of the term LOOP requiring it to be understood as referring to either asingle or multi-unit event at every use within the UFSAR. Similarly, the NRC Staff has identifiedno regulatory requirement, policy guidance, or docketed correspondence with I&M requiring anyreference to a LOOP to refer to either a single or multi-unit event. Consequently, whether aparticular reference to a LOOP within CNP's UFSAR refers to a LOOP affecting one or bothunits at CNP must be determined by reference to a number of factors such as the textsurrounding the UFSAR's reference to the LOOP, the larger structure of CNP's UFSAR, as wellas the relevant historical and regulatory background.i. The NRC Staff's Understanding of the Scope of a LOOP Is Not Supported bythe Surroundinq TextA comparison of the different contexts in which the term LOOP appears within CNP's SGTR andLoss of All AC Power to the Plant Auxiliaries accident analyses, respectively, does not supportthe NRC's generic interpretation of the term. As noted earlier, the NRC Staff's understanding ofCNP's licensing basis is based on the potentially broad scope of the LOOP within UFSAR Unit 1Section 14.1.12, "Loss of All AC Power to the Plant Auxiliaries." The UFSAR's description ofthe particular LOOP at issue could involve:A complete loss of all (non-emergency) AC power (e.g., offsite power) ...result[ing] in the loss of all power to the plant auxiliaries .... The loss ofpower may be caused by a complete loss of the offsite grid accompaniedby a turbine generator trip at the station, or by a loss of the on-site ACdistribution system.Reference 5 at Page 4 (quoting UFSAR Unit 1, Section 14.1.12.1) (emphasis added). Becausethe context of the UFSAR cited above passage is on its face ambiguous regarding the numberof units at CNP affected by the LOOP, the NRC Staff contends that it could, based only on agenerous reading of the cited text alone, be read to refer to a LOOP to one or both units atCNP.The context surrounding the use of the term LOOP within the SGTR accident analysis inUFSAR Units 1 and 2 Section 14.2.4 demands an entirely different conclusion regarding thenumber of units losing offsite power in a LOOP. Here, the UFSAR's use of the term LOOP isnot qualified by the broad adjectives, complete loss, all power, the offsite grid, etc., used in theearlier accident analyses in a way that could arguably suggest a LOOP denying power to bothunits; rather, CNP's SGTR accident analysis refers only to "offsite power", or "a loss of offsitepower" or "a coincident loss of offsite power." Reference 6 at Section 14.2.4.
Enclosure 2 to AEP-NRC-2013-53Page 13ii. The NRC Staffs Understandinq of the Meaninq of a LOOP Is Inconsistentwith the Structure of CNP's UFSARThe structure of the UFSAR also undercuts the generic meaning attached to the term LOOP bythe NRC Staff. According to Reference 5, the potentially broad scope of the LOOP described inUFSAR Section 14.1.12 defines the meaning of the term throughout the UFSAR. Reference 5at Page 4. However, the NRC Staff provides no justification for why the particular (broad)meaning it assigns to the term LOOP within UFSAR Section 14.1.12 is more appropriate forgeneric application throughout the UFSAR than the more limited-scope LOOP described withinother sections of the UFSAR such as Section 14.2.4.The NRC Staff's position is also not supported by the NRC and industry guidance regarding theform and content of CNP's UFSAR. Consistent with the scheme laid out in Regulatory Guide1.70 (Reference 15), CNP's UFSAR evaluates transient events and accidents satisfying aminimal threshold for best-estimate frequency of occurrence, which are then assigned afrequency grouping based on criteria established by the American Nuclear Society (ANS). Asstated in UFSAR Sections 14.0, ANS Condition 1 (normal operational transients) are omittedfrom CNP's UFSAR, while Condition 2 events (moderate frequency) appear mostly in UFSARSections 14.1, Condition 3 (infrequent) events in UFSAR Section 14.2, and Condition 4 (unlikelybut limiting) events mostly appear in UFSAR Section 14.3. Consistent with Regulatory Guide1.70, CNP's UFSAR analyzes each of the events within the UFSAR individually and for eachunit, to include a description of the initial assumptions, sequence of events, and radiologicalconsequences specific to each event. Reference 15 at Pages 15-4 to 15-7.The NRC Staff's position does not account for this structure. ANS guidance identifying thethreshold for consideration of transient events and accidents within an FSAR requires a minimalbest-estimate frequency of occurrence of >l.OE-6/yr. Reference 16 at 6. However, when theNRC Staff used its Donald C. Cook Nuclear Plant Standardized Plant Analysis Risk (SPAR)Model to calculate a best-estimate frequency of occurrence for an SGTR with a coincident,multi-unit LOOP, it obtained a value (2.12E-6/yr) not much greater than the threshold in ANSguidance; further, when accounting for the risk that a CAC would be unavailable formaintenance for 30 days, the best-estimate frequency of occurrence fell below (1.75E-7/yr) theANS threshold. Reference 1 at Enclosure Page 7. Informal calculations by I&M incorporatingmore recent industry data on the frequency of multi-unit LOOPs provide more reason toconclude that a multi-unit LOOP is too remote an event to be considered in CNP's design basisSGTR analysis. According to Reference 17, there was not one reactor trip coincident with amulti-unit LOOP reported by the U.S. commercial nuclear power industry between 1986-2004.Reference 17 at Page 51. Using this data, I&M's informal calculation of the probability of anSGTR with a coincident, multi-unit LOOP yields a best-estimate frequency of occurrence of6.33E-7/yr -below the ANS threshold for consideration within CNP's UFSAR. Further, thebest-estimate frequency of occurrence is even lower (1.91 E-8) when accounting for the risk thata CAC would be unavailable for any reason, including maintenance.Further, although Regulatory Guide 1.70 states that the input parameters and initial conditionsfor each accident should be "clearly identified" within its analysis, the NRC Staff's contentionassumes that the assumptions regarding the potential scope of one UFSAR Section 14 analysis
Enclosure 2 to AEP-NRC-2013-53Page 14(Loss of All AC Power to the Plant Auxiliaries) automatically carry over wholesale to subsequentaccident analyses (SGTR). Reference 15 at Page 15-5.Additionally, the NRC Staff's contention that its reading of the scope of the LOOP within UFSARSection 14.1.12 should apply to the LOOP assumed in CNP's Section 14.2.4 SGTR analysis.compares accidents with very different frequencies. The Loss of All AC Power to the PlantAuxiliaries is an ANS Condition II event, while the SGTR accident is a Condition III event.Reference 6 at Section 14.0. Further, because a dual-unit LOOP can be expected to occurmuch less frequently than a single-unit LOOP, application of the NRC Staff's reading of thescope of the term LOOP within CNP's SGTR analysis represents a significant change in theinitial assumptions and anticipated frequency for that particular accident. That revisedfrequency of CNP's design basis SGTR accident could conceivably require the assignment ofnew ANS Conditions to either the UFSAR Loss of All AC Power to the Plant Auxiliaries analysis(Reference 6 at Section 14.1.12), or its SGTR accident analysis (Reference 6 at Section14.2.4), which in turn would require the re-organization of CNP's UFSAR. Consequently, theNRC Staff's position does not account for the significance attached by NRC guidance to thedistinction between different ANS Conditions and (by extension) types of design basis events oraccidents.The NRC Staff's references to the use of the word "station" within the UFSAR's description ofCNP systems is similarly not helpful for determining the scope of the LOOP assumed in CNP'sSGTR accident analysis. In support of its contention that every use of the term LOOP refers toeither a single or multi-unit event, Reference 5 points to a handful of examples of the UFSAR'suse of the word "station" in descriptions of CNP Electrical System (at Section 1.3.7) and SafetyFeatures (at Section 1.3.8) that the NRC Staff understands to refer to both units at CNP.However, the NRC Staff nowhere explains why a handful of references to the word "station"within the system descriptions in Sections 1.3.7 and 1.3.8 define the use of that and otherterms (e.g., LOOP) throughout the UFSAR. Regulatory Guide 1.70 understood the systemdescriptions within the first section of a licensee's UFSAR to be distinct from the accidentanalyses described in a later section of the UFSAR:The first chapter of the SAR should present an introduction to the reportand a general description of the plant. This chapter should enable thereader to obtain a basic understanding of the overall facility withouthaving to refer to the subsequent chapters.Reference 15 at Page 1-1 (emphasis added). In contrast, the NRC Staff's position determinesthe meaning of ambiguous terms ("station", "LOOP") in the UFSAR's SGTR accident analysisassumptions not by reference to surrounding text, but by reference to language in an entirelydifferent UFSAR section. The NRC Staff's more fluid distinction between UFSAR sections isdifficult to reconcile with the approach endorsed within Regulatory Guide 1.70.Although the NRC Staff in Reference 1 states that the difference between UFSAR sectionsidentified above supports its understanding of CNP's licensing basis, the NRC Staffs position iserroneous. Conceding that high-level system descriptions within Section 1 of CNP's UFSAR donot prescribe accident analyses assumptions within subsequent UFSAR sections, the NRC Staffincorrectly asserts that:
Enclosure 2 to AEP-NRC-2013-53Page 15This argument supports the inspectors' position that the licenseecannot take credit for the unaffected unit's non-safety-related PACunless explicitly approved by the NRC and described in the SGTRanalysis.Reference 1, Enclosure at Page 5 (emphasis added). Notwithstanding the fact the languagewithin Section 1 of CNP's UFSAR is unhelpful for interpreting language describing UFSARaccident analysis assumptions, it does not follow that Section l's high-level description of thecomponents comprising CNP systems would not control throughout the UFSAR. RegulatoryGuide 1.70 states that Section 1 of CNP's UFSAR exists precisely so that I&M would not haveto describe CNP systems and components multiple times. Reference 15 at Page 1-1. BecauseSection 1.3.9.h of CNP's UFSAR describes CNP's compressed air system as a shared systemof which both units' PACs and CACs are components, the NRC Staffs explicit endorsementwithin the SER in Reference 4 of the continued availability of motive force to the SG PORVsfrom CNP's control air appurtenances and equipment permits I&M to take credit for theunaffected unit's PAC in CNP's SGTR accident analysis. Further, by the NRC Staff's logic, I&Mwould not be able to take credit for the operation of any CAC or PAC within CNP's SGTRaccident analysis, as neither of those components is explicitly mentioned in the UFSAR's SGTRaccident analysis.Additionally, even if the NRC Staff's approach were appropriate, the cited examples of the term"station" within Section 1 of the UFSAR do not support its position. Reference 6 Section 1.3.7states:"The station auxiliary power system consists of auxiliary transformers,4160 v and 600 v switchgear, 600 v motor control centers, 120 v-a-c vitalinstrument buses and 250 v d-c buses."However, the NRC Staffs suggestion that the term "station" in this context necessarily refers toboth units at CNP is incorrect. Indeed, each unit at CNP has the components (redundantauxiliary transformers, multiple 600 v switchgear, independent 120 v-a-c vital instrument busesand 250 v-d-c buses, and 4160 v and 600 v switchgear) the NRC Staff suggests represents ashared system between CNP units. Similarly, both units have the EDGs and turbinesmentioned in the cited passage from UFSAR Section 1.3.8. Further, the NRC Staff's claim thatthe use of the term "station" within Section 1.3.8's description of CNP Safety Features provesthat there is only one, shared auxiliary power system at CNP is at odds with surrounding text notexamined by the NRC Staff. Specifically, UFSAR Section 1.3.9, "Shared Facilities andEquipment," begins by noting that:Separate and similar systems and equipment are provided for each unit,except as noted below.Reference 6 at Section 1.3.9 (emphasis added). The auxiliary power system is absent fromSection 1.3.9's list of shared systems and equipment.iii. The NRC Staff's Understanding of the Term LOOP Is at Odds with theReaulatorv History of CNP and Similarlv-Situated Facilities
Enclosure 2 to AEP-NRC-2013-53Page 16The NRC Staff's understanding of the term LOOP also does not account for docketedcorrespondence acknowledging the retention of the assumptions within CNP's original SGTRaccident analysis. As explained at length earlier, the NRC Staff in 2001 reviewed and explicitlyapproved I&M's retention of CNP's original licensing basis assumptions for SGTR accidents,including the assumption of a single-unit LOOP only. Consequently, the NRC Staff'sunderstanding of the scope of the term LOOP assumed within CNP's SGTR accident analysisnot only re-writes CNP's UFSAR, but also re-writes nearly forty years' worth of pertinentdocketed correspondence.Further, as explained earlier, the NRC Staffs reading of the term LOOP within CNP's SGTRaccident analysis is also inconsistent with the regulatory history of CNP and other multi-unitfacilities of similar vintage. The two units at CNP were licensed and constructed on a staggeredschedule, with construction on Unit 1 beginning before Unit 2 such that Unit 1 received itsoperating license several years before Unit 2 (1974 as opposed to 1977). Consequently, theSGTR accident analysis within CNP's original licensing basis did not, as a practical matter,assume a multi-unit LOOP.Further, the CNP is not the only licensee that assumes only a single-unit LOOP within thedesign basis accident analyses for the units at its facility. I&M's informal polling of other multi-unit facilities licensed in approximately the same timeframe as CNP reveals that many of thoselicensees understand the licensing basis assumptions for units at their facility to assume only asingle-unit LOOP during SGTRs and other accidents. Further, among those licensees whoselicensing basis currently assumes multi-unit LOOPs were some who acknowledged that theircurrent licensing basis assumptions are a departure from original licensing basis assumptionsthat understood LOOPs to affect only a single unit at their facility.Lastly, the Commission's current regulations and guidance governing the availability of offsitepower reflect the unit-specific approach to electric system design within licensing basis accidentassumptions at CNP and other similarly-situated facilities. Most prominently, the current StationBlackout Rule at 10 CFR 50.63 (Reference 8) is unit-specific in its approach to the availability ofAC power, including offsite power. Although the NRC has recently published a Federal Registernotice (Reference 18 at 16179) indicating a desire to revise its Station Blackout Rule and otherregulations and guidance to adopt a facility-wide perspective on continuity of electrical power,interpreting the language within CNP's licensing basis against that proposed approach would bepremature, regardless of whether the NRC Staff can (as Reference 1 asserts) conceive ofscenarios in which plant configuration would make a multi-unit LOOP a credible event at CNP.6. The NRC Staffs Position Is Unnecessary for Assuring Adequate Protection Against EitherDesign Basis Events or Beyond-Design Basis External EventsNRC Orders issued following the earthquake and tsunami at the Fukushima Dai-ichi nuclearpower plant in March 2011 acknowledge that existing defense-in-depth approaches at licensedfacilities provide adequate protection of public health and safety against design basis accidents.Specifically, EA-12-049 states:To protect public health and safety... the NRC's defense-in-depthstrategy includes multiple layers of protection: (1) prevention of accidentsby virtue of the design, construction, and operation of the plant; (2)
Enclosure 2 to AEP-NRC-2013-53Page 17mitigation features to prevent radioactive releases should an accidentoccur; and (3) emergency preparedness programs that include measuressuch as sheltering and evacuation .... These defense-in-depth featuresare embodied in the existing regulatory requirements and thereby provideadequate protection of the public health and safety.Reference 19 at Page 5 (emphasis added). Compliance with those NRC requirements, theNRC concluded, "presumptively assures adequate protection" of public health and safety frominadvertent release of radioactive materials during a design basis accident. Reference 19 atPages 4-5.As explained at length earlier, the NRC Staff's contention within Reference 1 that CNP is not incompliance with licensing basis requirements for a design basis SGTR accident is incorrect.CNP's licensing basis has never assumed that the LOOP coincident with a design basis SGTRaccident involves both units at CNP, and the NRC Staff has presented no meaningful evidencein support of a contrary position. Further, as recently as 2001, the NRC Staff endorsed themeasures (including the crediting of the continued availability of SG PORVs and supportingcompressed air system components) I&M employs for mitigating the risk of inadvertent releaseof radioactive materials during a design basis SGTR accident at CNP. Reference 4 concludesthat I&M's approach to mitigating the consequences of a design basis SGTR provides"reasonable assurance" of protection of public health and safety, and "will be conducted incompliance with the Commission's regulations. ... "Further, as noted earlier, I&M has supplemented the mitigation measures for SGTR accidentsevaluated within Reference 4 to provide additional defense-in-depth from design basis SGTRaccidents. Specifically, I&M in March 2013, completed installation of a plant modification andrevised CNP operating procedures to ensure that backup nitrogen tanks are immediately andautomatically available during an SGTR for operation of SG PORVs without the need for manualvalve manipulation outside the control room. I&M has also revised CNP Work Controlprocesses to provide additional defense-in-depth from a loss of control air pressure byrestricting removal for maintenance of the operating unit's CAC when the opposite unit isshutdown and the shutdown unit's PAC is aligned to preferred offsite power.In contrast, the NRC Staff has not demonstrated that its position would result in any meaningfulcontribution to adequate protection of public health and safety from design basis SGTRaccidents at CNP. As noted earlier, the most recent published industry data on the frequency ofLOOPs within Reference 17 indicates that the best-estimate frequency of occurrence for a multi-unit LOOP coincident with an SGTR would fall well below the minimal threshold within ANSguidance (Reference 16) for consideration within CNP's design basis. Moreover, the differencein core damage frequency from adopting the NRC Staff's position regarding the scope of theLOOP accompanying a design basis SGTR accident is so small (2.4E-8/yr) as to provide nomeaningful advantage over I&M's understanding of CNP's licensing basis for assuring adequateprotection of public health and safety. Reference 1, Enclosure at Page 1. Further, even thismarginal difference in core damage frequency between I&M's and the NRC Staff's positions islikely overstated, as the core damage frequency calculation within Reference 1 (Enclosure atPages 6-7) does not account for the additional defense-in-depth measures implemented at CNPsince the 2012 CDBI.
Enclosure 2 to AEP-NRC-2013-53Page 18Lastly, the NRC Staff has provided no basis to conclude that I&M has failed to provide adequateprotection against beyond-design basis scenarios involving an SGTR accompanied by acoincident, multi-unit LOOP. As explained in Order EA-12-049, the events at FukushimaDai-ichi demonstrated the need for licensees to adopt additional defense-in-depth measures tomitigate the consequences of beyond-design basis external events, such as those resulting inthe extended loss of electrical power at multiple units at a facility. Reference 19 at Pages 4-6.Subsequent NRC guidance (Reference 20 at Page 4) endorsed licensees' use of the NuclearEnergy Institute's (NEI's) Diverse and Flexible Mitigation Capability (FLEX) strategy (Reference21) to satisfy Order EA-12-049's requirements for assuring adequate protection against beyond-design basis external events resulting in extended loss of electrical power (including offsitepower) at both units at a multi-unit facility. As required by Order EA-1 2-049, I&M has submittedan Overall Integrated Plan (Reference 22) for mitigation of beyond-design basis external eventsat CNP. I&M's Overall Integrated Plan incorporates the FLEX strategy endorsed by the NRCStaff in Reference 20 for use by licensees in satisfying the requirements within Order EA-12-049for mitigation measures providing adequate protection from beyond-design basis events suchas a multi-unit LOOP accompanying an SGTR.7. The NRC Staff's Determination that the NCVs Represent a More-than-Minor PerformanceDeficiency Involving Cross-Cutting Aspects Lacks MeritIn Reference 1, the NRC Staff contends that the NCVs represent a more-than-minorperformance deficiency involving cross-cutting areas of human performance, the component ofdecision making, and the aspect of conservative assumptions. Reference 1 Enclosure, atPages 1 and 2. The NRC Staff stated that the NCVs involved cross-cutting aspects becauseI&M's plant procedures assumed that the unaffected unit's compressed air system equipmentwould be available during an SGTR accident, despite the fact that the NRC Staff nowunderstands CNP's licensing basis to assume that an SGTR accident would be accompanied bya multi-unit LOOP. Reference 1 Enclosure, at Pages 1 and 2.The NRC Staff's conclusion that the NCVs involve cross-cutting aspects, however, incorrectlyassumes the validity of NCVs identified within Reference 1. As explained at length above, thoseNCVs are based on an erroneous understanding of the scope of the coincident LOOP withinCNP's design basis SGTR accident analysis: contrary to the NRC Staffs current position,CNP's licensing basis has only ever assumed a single-unit LOOP as an initial condition in anSGTR event. Consequently, the unaffected unit's PAC will remain available to provide controlair pressure to operate SG PORVs in the affected unit in the event of an SGTR event,regardless of the status of the CAC of the affected unit. Further, the NRC Staff in the 2001 SERwithin Reference 4 endorsed I&M's claims regarding the continued availability of control air tooperate an affected unit's SG PORVs during an SGTR accident, notwithstanding a coincidentLOOP. Because the NCVs within Reference 1 are incorrect, the NRC Staff's conclusion thatthose NCVs involve cross-cutting aspects is similarly incorrect.Additionally, even if the NRC Staff's current understanding of CNP's licensing basis werecorrect, the NCVs identified within Reference 1 would not involve cross-cutting aspects.Although Reference 1 (Enclosure, Page 7) criticizes I&M for not having adopted requirements,EOPs, and work control procedures positively demonstrating safety, the NRC Staff nowhereexplains how I&M's requirements were inconsistent with reactor safety and public health. Asnoted earlier, the NRC Staff concluded in the SER (Pages 3 to 5) within Reference 4 that the
Enclosure 2 to AEP-NRC-2013-53Page 19changes to CNP's licensing basis proposed by I&M in its 2000 LAR would not increase the riskor consequences of an SGTR accident beyond the conservative estimates within CNP's originallicensing basis. In arriving at this conclusion, the NRC Staff explicitly noted that I&M hadrevised its EOPs for SGTR accidents to improve margin to steam generator overfill.Reference 4, SER at 4. Further, the core damage frequency data provided by the NRC Staff inReference 1 (Enclosure at Page 1) is consistent with the NRC Staffs conclusions withinReference 4, as the difference in core damage frequency from assuming a dual-unit LOOP isonly marginally different (2.4E-8/yr) from scenarios involving a single-unit LOOP.Further, the NRC Inspection Manual states that for an NCV to have cross-cutting aspects, theperformance deficiency at issue must be "recent (i.e., nominally within the last three years)."Reference 23, at Page 3. However, as explained at length above, the NCVs in Reference 1 arebased on an understanding of CNP's licensing basis that has been in place since the originallicensing of Unit 1 at CNP around forty years ago, and which was endorsed by the NRC Staff asrecently as 2001. Consequently, the NCVs within Reference 1 do not satisfy NRC InspectionManual standards for determining whether NCVs have cross-cutting aspects.Nor can the NRC Staff claim that I&M's failure to correct the longstanding performancedeficiency until recently is indicative of present performance. Although the NRC InspectionManual allows for a cross-cutting determination if "the performance deficiency occurred morethan three years ago, but the performance characteristic has not been corrected or eliminated",it severely limits the application of this exception to "some rare or unusual cases". Reference 23at Page 3. Reference 1 provides no justification for why the NCVs represent a "rare or unusualcase" warranting application of this exception. Further, as explained above, I&M'sunderstanding of its licensing basis is not rare or unusual; in fact, multiple plants of similarvintage and configuration have the same licensing basis assumptions regarding the scope of aLOOP during an SGTR or other accident.8. ConclusionFor the reasons identified above, both the NCVs identified within Reference 1 and the NRCStaff's determination that those NCVs involve cross-cutting aspects are incorrect.
Enclosure 2 to AEP-NRC-2013-53Page 20REFERENCES:1. Letter from G. Shear, NRC, to L. J. Weber, I&M, "Donald C. Cook Nuclear Power Plant,Units 1 and 2, Component Design Basis Inspection 05000315/2013010;05000316/2013030," dated July 8, 2013.2. Letter from A. M. Stone, NRC, to L. J. Weber, I&M, "D. C. Cook Nuclear Power Plant,Units 1 and 2, Component Design Bases Inspection 05000315/2012007;05000316/2012007," dated January 11, 2013.3. Letter from W. Hodge, I&M, to C. Tilton, NRC, "D. C. Cook CDBI Response to Question2012-CDBI-298," dated November 15, 2012.4. Letter from J. F. Stang, NRC, to R. P. Powers, I&M, "Donald C. Cook Nuclear Plant,Units 1 and 2 -Issuance of Amendments (TAC Nos. MB0739 and MB0740)," datedOctober 24, 2001.5. Letter from K. O'Brien, NRC, to S. Bahadur, NRC, "Task Interface Agreement -Licensing Basis for Donald C. Cook Nuclear Power Plant, Units 1 and 2, During a SteamGenerator Tube Rupture Event Coincident with a Loss of Offsite Power (TIA 2012-11),"dated December 7, 2012.6. Donald C. Cook Nuclear Plant Updated Final Safety Analysis Report Rev. 24, datedMarch 17, 2012.7. Letter from I&M to Ann Marie Stone and Caroline Tilton, NRC, "Response to NRCInspection Report Issued January 11, 2013 Containing the Results of the ComponentDesign Basis Inspection Conducted Between July 23, 2012 and December 3, 2012,"dated February 8, 2013.8. 10 CFR 50.63, "Loss of All Alternating Current Power."9. Donald C. Cook Nuclear Plant Preliminary Safety Analysis Report for Units 1 and 2,dated December 18, 1967.10. Donald C. Cook Nuclear Plant Final Safety Analysis Report for Units 1 and 2, datedFebruary 2, 1971.11. Amendments to Donald C. Cook Nuclear Plant Final Safety Analysis Report for Units 1and 2, dated November 11, 1977.12. Amendments to the Donald C. Cook Nuclear Plant Final Safety Analysis Report for Units1 and 2, dated July 1997.13. Letter from R.P. Powers, I&M, to the NRC Document Control Desk, "Letter C1000-11,Donald C. Cook Nuclear Plant Units 1 and 2 License Amendment Request for Changesin Steam Generator Tube Rupture Analysis Methodology," dated October 24, 2000.
Enclosure 2 to AEP-NRC-2013-53Page 2114. Letter from M. W. Rencheck, I&M, to the NRC Document Control Desk, "Letter C0601-21, Donald C. Cook Nuclear Plant Units 1 and 2 Response to Request for AdditionalInformation Regarding License Amendment for 'Changes in Steam Generator TubeRupture Analysis Methodology (TAC Nos. MB0739 and MB0740)," dated June 29, 2001.15. NRC Regulatory Guide 1.70, "Standard Format and Content of Safety Analysis Reportsfor Nuclear Power Plants, Rev. 3, " dated November 1978.16. American Nuclear Society, ANSI/ANS-51.1-1983, "Nuclear Safety Criteria for the Designof Stationary Pressurized Water Reactor Plants," dated 1983.17. NUREG/CR-6890, "Reevaluation of Station Blackout Risk and Nuclear Power Plants:Analysis of Loss of Offsite Power Events 1986-2004," dated December 2005.18. 77 Federal Register 16175, "NRC Advanced Notice of Proposed Rulemaking: StationBlackout," dated March 19, 2012.19. NRC Order Number EA-12-049, "Order Modifying Licenses with Regard toRequirements for Mitigation Strategies for Beyond-Design-Basis External Events," datedMarch 12, 2012.20. NRC Interim Staff Guidance JLD-ISG-2012-01, "Compliance with Order EA-12-049,Order Modifying Licenses with Regard to Requirements for Mitigation Strategies forBeyond-Design-Basis External Events, Rev. 0," dated August 29, 2012.21. NEI 12-06, "Diverse and Flexible Coping Strategies (FLEX) Implementation Guide, Rev.0," dated August 2012.22. Letter from J. P. Gebbie, I&M, to NRC, "Donald C. Cook Nuclear Plant Unit 1 and Unit 2Overall Integrated Plan In Response to March 12, 2012 Commission Order ModifyingLicenses with Regard to Requirements for Mitigation Strategies for Beyond-Design-Basis External Events (Order Number EA-12-049)," dated February 27, 2013.23. NRC Inspection Manual Chapter 0612, "Power Reactor Inspection Reports," datedJanuary 24, 2013