ML12340A450

NRR E-mail Capture - ME6818 - DAEC Adoption of NFPA 805 - Request for Additional Information (RAI) Items - Round 2 (More)

ML12340A450
Person / Time
Site:	Duane Arnold
Issue date:	12/05/2012
From:	Feintuch K Division of Operating Reactor Licensing
To:	Swenzinski L Nextera Energy
References
TAC ME6818
Download: ML12340A450 (15)

Text

1 NRR-PMDAPEm Resource From:

Feintuch, Karl Sent:

Wednesday, December 05, 2012 12:23 PM To:

Swenzinski, Laura Cc:

Robinson, Jay; Fields, Leslie

Subject:

ME6818 - DAEC Adoption of NFPA 805 - Request for Additional Information (RAI) Items -

Round 2 (More)

Attachments:

ME6818 DAEC APLA RAIs NFPA 805 LAR Round 2 (More).docx By letter dated August 5, 2011, NextEra Energy Duane Arnold, LLC, (the Licensee), submitted a license amendment request (LAR) to transition their fire protection licensing basis at the Duane Arnold Energy Center (DAEC), from Title 10 of the Code of Federal Regulations (10CFR), Section 50.48(b), to 10CFR50.48(c),

National Fire Protection Association Standard NFPA 805 (NFPA 805).

A review team, consisting of U.S. Nuclear Regulatory Commission (NRC) staff and contractors from Pacific Northwest National Laboratory (PNNL) and the Center for Nuclear Waste Regulatory Analyses (CNWRA) participated in a regulatory audit of the DAEC in Palo, Iowa during the period December 12-16, 2011. In a message dated January 31, 2012, (ADAMS Accession No. ML12031A112) the NRC issued requests for additional information (RAIs). In letters dated April 23, 2012 (ADAMS Accession No. ML12117A052) and May 23, 2012 (ADAMS Accession No. ML12146A094) the licensee provided responses to the RAIs. In a message dated November 8, 2012, (ADAMS Accession No. ML12318A394) the NRC issued another request for additional information, which is due for response by January 11, 2013.

The Probabilistic Risk Assessment Licensing (APLA) Branch has reviewed the information provided by the licensee and determined that additional information is needed to complete the review. Please note that review efforts on this task (TAC No. ME6818) are being continued and additional RAIs may be forthcoming.

The review branch is proposing a 60 calendar day response time from the date of formal issuance. For reference, 60 calendar days after December 5, 2012 calculates to be Sunday, February 3, 2013. Therefore, the proposed response date would be Monday, February 4, 2013.

Please review the attached items for clarification and request a conference call if needed. Though the items in the attached file are identified as draft in the heading (subject to the need for clarification), they are firm relative to the information being requested.

If you have any questions or need any additional information concerning these RAIs, please contact me at 301-415-3079.

Docket No: 50-331

Hearing Identifier:

NRR_PMDA Email Number:

552 Mail Envelope Properties (Karl.Feintuch@nrc.gov20121205122300)

Subject:

ME6818 - DAEC Adoption of NFPA 805 - Request for Additional Information (RAI) Items - Round 2 (More)

Sent Date:

12/5/2012 12:23:05 PM Received Date:

12/5/2012 12:23:00 PM From:

Feintuch, Karl Created By:

Karl.Feintuch@nrc.gov Recipients:

"Robinson, Jay" <Jay.Robinson@nrc.gov>

Tracking Status: None "Fields, Leslie" <Leslie.Fields@nrc.gov>

Tracking Status: None "Swenzinski, Laura" <Laura.Swenzinski@nexteraenergy.com>

Tracking Status: None Post Office:

Files Size Date & Time MESSAGE 2199 12/5/2012 12:23:00 PM ME6818 DAEC APLA RAIs NFPA 805 LAR Round 2 (More).docx 51399 Options Priority:

Standard Return Notification:

No Reply Requested:

No Sensitivity:

Normal Expiration Date:

Recipients Received:

DRAFT REQUEST FOR ADDITIONAL INFORMATION LICENSE AMENDMENT REQUEST TO ADOPT NATIONAL FIRE PROTECTION ASSOCIATION STANDARD 805 PERFORMANCE-BASED STANDARD FOR FIRE PROTECTION FOR LIGHT WATER REACTOR GENERATING PLANTS DUANE ARNOLD ENERGY CENTER (TAC NO. ME6818)

Office of Nuclear Reactor Regulation Division of Risk Assessment PRA Licensing Branch Probabilistic Risk Assessment RAI 01.01 In your letter dated May 23, 2012 (ADAMS Accession No. ML12146A094) you responded to Probabilistic Risk Assessment (PRA) Request for Additional Information (RAI) 1 and provided information which addressed the correction to the Fire PRA (FPRA) model but additional information is necessary for the staff to complete its review.

a. Verify the new core damage frequency (CDF) and large early release frequency (LERF) results have been reviewed against the applicable American Society of Mechanical Engineers (ASME) PRA Standard supporting requirements for FQ-E1 and the associated Facts and Observations (F&Os) from the 2010 Peer Review. This includes reasonableness reviews identified in the supporting requirements and the F&Os and of the changes to the risk significant scenario results. Updated Tables W-1 and W-2 indicate that the CDF for scenarios 10E F46 and 10E F75 and the LERF for scenarios 10F F12 and 10E F77 have increased compared to the original tables, yet the fire area totals have decreased. Explain why the totals have decreased.

b. Briefly summarize the review process applied to the updated model results to insure that the logic models are valid and no additional erroneous cutsets are included.

c. If the transient heat release rate (HRR) used in the updated results does not comply with NUREG/CR 6850, EPRI/NRC-RES Fire PRA Methodology for Nuclear Power Facilities, April 2005, justify the HRR used using the clarification guidance criteria discussed in Nuclear Energy Institutes (NEIs) letter to NRC dated September 27, 2011 on Recent Fire PRA Methods Review Panel Decision: Clarification for Transient Fires and Alignment Factor for Pump Oil Fires, (ADAMS Accession No. ML113130448, Non-Publicly Available) and as endorsed by the NRC in its letter to NEI dated June 21, 2012 (ADAMS Accession No. ML12171A583 on Recent Fire PRA Methods Review Panel Decisions and EPRI 1022993, Evaluation of Peak Heat Release Rates in Electrical Cabinet Fires. Specifically, the justification should address the specific attributes and considerations applicable to the location, plant administrative controls, and the results of a review of records related to violations of the transient combustible controls. If the HRR cannot be justified using the guidance criteria, provide the results of a sensitivity study using the NUREG/CR 6850 HRR in the updated FPRA model and addressing the total CDF and LERF as well as the delta () CDF and LERF.

d. For Scenario 11A A02 of Table 2 of the RAI response the fire ignition frequency (FIF) is given as 5.70E-07, the conditional large early release probability (CLERP) is given as 1.0, and the LERF is given as 1.71E-07. Provide clarification.

Probabilistic Risk Assessment RAI 02.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 02. Expand on the response for the disposition of F&O 4-17 on supporting requirement (SR) PRM-B9. The disposition of this F&O addresses the modeling initiation and actuation logic for the main steam isolation valves (MSIVs) fail to close. Address the broader finding concerning inclusion of all the systems that are required for initiation and actuation of a system, including the presence of the conditions needed for automatic actuation (e.g., low vessel water level) and permissive and lockout signals that are required to complete actuation logic. Confirm that the initiation and actuation logic for all systems are appropriately included in the FPRA and that this inclusion meets the requirements of SR PRM-B9.

Probabilistic Risk Assessment RAI 06.01 In your letter dated May 23, 2012 (ADAMS Accession No. ML12146A094) you responded to PRA RAI 06. PRA RAI 06 discussed the multi-compartment analysis (MCA) from 02E to 02B, and your response stated the following:

"When considering a hydrogen recombiner fire event, the fire is most likely to propagate to 02B through the east recombiner vault door and heating, ventilation, and air conditioning (HVAC) room door. The target set outside the HVAC room door in 02B does not contain FPRA cables."

a. The response did not address potential impact of fire propagation through the east recombiner door. Discuss the analysis of this potential pathway for this scenario.

b. In addition, since the fire area is a control rod drive (CRD) module area, discuss whether a potential impact on the CRD modules exists and was considered in the risk analysis.

Describe the impact on the results of this MCA analysis.

Probabilistic Risk Assessment RAI 08.01

a. In your letter dated May 23, 2012 (ADAMS Accession No. ML12146A094) you responded to PRA RAI 08 and stated that since all MCA results are less then 1E-07 they screen out and including them in the total and delta CDF and LERF is not appropriate. As stated in the fire scenario report (FSR) Appendix C, the MCA methodology "differs slightly from the guidance in NUREG/CR-6850, Section 11.5.4 but meets the intent of the requirements."

The staff review of the methodology noted one difference from the NUREG/CR-6850 guidance involving your steps 9, 10 and 11 compared with 6850's Step 5.c (Section 11.5.4.5, p. 11-46). The latter requires that: "...failure of all fire PRA components and cables present in the combination of exposing and exposed compartments should be estimated." If not screened out with the above assumption, the next step is the detailed analysis whose results are to be included in the total CDF and LERF.

This is not a stated requirement of your procedure and it is not clear it is necessarily followed. For example in the response to PRA RAI 06, the discussion of impact of failure of the wall between PAUs 02E and 02B states that: "Targets in 02B adjacent to the wall include MCC 1D41." It thus appears that there may be other equipment in 02B, and if so, the NUREG/CR-6850 guidance would not be met.

Provide a sensitivity study that follows the accepted guidance, i.e., where the final step of the screening analysis is in accordance with the guidance of NUREG/CR-6850 for all areas, such as that for PAUs 02E and 02B, where the final screening step did not include a conservative CCDP assuming all targets in the two PAUs failed. The sensitivity study results should include the resulting scenarios in the FPRA and all the total/delta CDF and LERF evaluations.

b. The MCA analysis uses a CDF of <1E-07 for screening purposes. ASME PRA standard SR QNS-C1for CC II requires (as clarified by RG 1.200) that it be verified that:

i.

the quantitative screening process does not screen the highest risk fire areas and ii.

the sum of the CDF contributions for all screened fire compartments is < 10% of the estimated total CDF for fire events and iii.

the sum of the LERF contributions for all screened fire compartments is < 10% of the estimated total LERF for fire events.

Describe how the results of the corrected FPRA model compare to the ASME PRA standard SR QNS-C1for CC II, as clarified by RG 1.200. If greater than or equal to the SR QNS-C1 criteria, provide the sum of all screened CDF and LERF contributions.

Probabilistic Risk Assessment RAI 10.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) your responded to PRA RAI 10. The response to part (b) of PRA RAI 10 states "An examination of the impairment log, which has been electronic since 2008, reveals that the vast majority of impairments on these systems are not for corrective maintenance but for..." There is no clear statement if any of the fire protection systems credited in the license amendment request (LAR) have experienced outlier behavior relative to system unavailability or not. Describe whether the credited systems have experienced outlier behaviors and, if so, provide an assessment of the impact of this behavior on the FPRA total and delta risk results.

Probabilistic Risk Assessment RAI 11.01

a. In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 11 citing Table O-2 of NUREG/CR-6850 and referring to the two values 5E-4 and 1E-5 as conditional probabilities. In the table, these values are in parentheses and given as per year. It is apparent that these values are frequencies and the 5E-4/yr is the frequency of fires for which catastrophic consequences can be prevented by suppression and not a conditional probability. Provide clarification on this discrepancy, and its impact on the assessment and, as appropriate, provide revised results.

b. In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 11 and discussed fire scenario TBO01. The FSR, Rev. 4 discusses a revised catastrophic TG oil fire, identified as fire scenario TBO01. Appendix A of the FSR identifies it as TB1 O01. The updated quantification report does not appear to include the CDF or LERF for this scenario. Please explain this discrepancy and provide updated results as necessary.

Probabilistic Risk Assessment RAI 12.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 12 and provided the results of a sensitivity study of eliminating the credit for the control power transformer (CPT) for Scenarios 02B-F01 and 03A-D12. Credit for CPT is denoted by footnote 2 of Table 4.0-1 of the FSR which then appears to indicate that CPT credit was also used for scenario 02D-B01. Revision 4 of the FSR added footnote 6 to the Table indicating that this later scenario was not used in the final quantification. Explain why this scenario was not used in the final quantification.

Probabilistic Risk Assessment RAI 14.01 In your letter dated May 23, 2012, (ADAMS Accession No. ML12146A094) you responded to PRA RAI 14. The response included added cable spreading room (CSR) analysis using NUREG/CR-6850 FIF and modeling from the CSR Risk evaluation report. This modeling considered prompt suppression, plant trip, offsite power, Division 1 and Division 2 availability, alternate shutdown capability (ASC) availability, and ASC mitigation. Provide further justification for the following modeling considerations:

a. For plant trip where there is no data, provide additional support for why the 50/50 split is assumed to be a point estimate for the case.

b. For offsite power, the point estimate case assumes 13% of the time offsite power will be lost based on offsite power cables routed in approximately 85 routing points of the total of 660 in the nuclear safety capability assessment (NCSA) database. The upper bound estimate is taken as 20%. Provide further support for the validity of these postulated split fractions.

c. The likelihood that Division 1 paths will be available from the control room (CR) depends on the resolution of PRA RAI 20. For the likelihood that the ASC will be impacted by the fire, a 50/50 split is used for both the point estimate and the upper bound. Discuss the reasonableness of this assumption, including fire modeling or fire modeling assumptions if applicable.

d. Furthermore, it appears that the CSR risk analysis credits prompt suppression for transient fires. However, per NUREG/CR-6850, Attachment P, prompt suppression can only be credited for hot work fire scenarios in which a continuous fire watch is present; this credit does not apply for transient fires. Provide justification on the application of credit for prompt suppression or reconsider the analysis following NUREG/CR-6850 guidance.

e. In the CSR Risk Report, the LERF was estimated to be 30% of CDF based on the CDF/LERF ratio from the full power internal events (FPIE) PRA. Use of the FPIE PRA to estimate FPRA LERF for this fire scenario is not applicable; rather the FPRA LERF should consider the FPRA model. Reconsider the LERF analysis for this fire scenario using a justified analysis and provide revised results.

Probabilistic Risk Assessment RAI 16.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 16. F&O 5-37 and 5-38 identify apparent weaknesses in identifying cues and associated instrumentation in the FPRA evaluations. The response to PRA RAI 16 only addressed the specific operator actions identified and stated that credit for these actions was not, in fact, credited. The disposition of related F&O 1-3 cites Table 3.3-1 for a listing of instrumentation relied upon for operator actions. However, this Table does not include the instrumentation specifically cited in the F&O, i.e. hotwell level and filter differential pressure (dp).

It would appear that hot well level is needed for actions where PCS injection is occurring. For example, Table 3.3-1 includes operator action DCNDSTCNOP02----HE-- OPERATOR FAILS TO OPEN HOTWELL MAKEUP BYPASS LINE with RPV level the only cue identified.

a. Provide further information concerning the need for hotwell level and filter dp instrumentation.

b. Does Table 3.3-1 only include cues and instrumentation that appear in the safe-shutdown list because they are relied on to support safe-shutdown paths? If not, what is included in Table 3.3-1?

c. In addition, for several operator actions concerning loss of room cooling, the cue is stated as "Environmental cue given room heatup given loss of room cooling." If instrumentation is being relied on for the cue, describe whether it has been verified to be available for applicable fire scenarios.

d. Cues and instrumentation may be credited in non-safe-shutdown operator actions in the FPRA that do not appear in the safe-shutdown list. Describe how the affects of fire scenarios on this instrumentation are evaluated for credit in the FPRA.

Probabilistic Risk Assessment RAI 20.01 In your letter dated May 23, 2012, (ADAMS Accession No. ML12146A094) you responded to PRA RAI 20 and described an analysis of a fire in the corner of the CSR. Cables for only motor control center (MCC) 1B34 and 1B44 are mentioned and they are not both in the zone of influence (ZOI) of the postulated transient fire.

a. Explain why this situation is not a variance from deterministic requirement (VFDR) for which any retained risk should be summed with other retained VFDR risks.

b. Confirm that there are no other Division 1 cables other than those for MCC 1B34 in the CSR.

c. Provide justification for not considering the fire induced failure of cables for MCC 1B34 and other (than cables for MCC1B44) division 2 cables in the ZOI.

Probabilistic Risk Assessment RAI 31.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 31. The disposition of F&O 1-1 in the LAR states that each multiple spurious operation (MSO) disposition was added to Table G-1 of the fire model development report (FMDR). In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you

responded to PRA RAI 47 and stated that functional failure and MSO were considered in modeling the containment isolation valves (CIVs). However, review of Table G-1 noted that it did not include MSOs for CIVs. Clarify this discrepancy and provide the CIVs which may be missing from Table G-1 based on the disposition of F&O 1-1.

Probabilistic Risk Assessment RAI 32.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 32.

a. The response does not mention MO-2010. Clarify if V19-048 is modeled in the FPRA as open and requiring operator action to close, and, also, describe whether it is a model surrogate for MO-2010.

b. The response states that closure of this valve is not a required recovery action (RA) to meet the risk criteria for CR abandonment, and is therefore only credited for defense-in-depth (DID). The response also stated that closure of V19-048 is prescribed for a fire in RB1 and RB3. Describe the risk impact if V19-048 cannot be closed.

Probabilistic Risk Assessment RAI 35.01 In your letter dated May 23, 2012, (ADAMS Accession No. ML12146A094) you responded to PRA RAI 35 and provided a sensitivity study using the FAQ 08-0046 (ADAMS Accession No. ML093220426) event tree even though this FAQ is not intended to be applied in the main control room (MCR). Given that a FAQ has not yet been established for incipient detection in the MCR and a basis for credit has not been established, provide a sensitivity analysis on the LAR results (CDF, LERF, CDF, and LERF) without crediting incipient detection in the MCR.

Probabilistic Risk Assessment RAI 39.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 39 and stated None of the fire response actions or the FPIE actions retained for the Fire PRA were considered complex enough to justify the use of the upper bound ASEP curve."

ASME PRA standard (ASME/ANS RA-Sa-2009) SR HR-G3 requires that when estimating human error probabilities (HEPs) the impact of ten specifically defined plant and scenario specific shaping factors should be evaluated in order to meet CC-II. The accident sequence evaluation program (ASEP), which was used for analysis of cognition error when time available for recovery is less than 60 minutes, does not consider these shaping factors. It Is not clear whether ASEP, as it was applied, yields more conservative results than cause-based decision tree method (CBDTM), which was used for cognition error when time available for recovery is greater than 60 minutes. Application of ASEP to this specific set of cognitive errors using the lower bound diagnosis curve (Figure 7-1) provided in NUREG/CR-4772, Accident Sequence Evaluation Program Human Reliability Analysis Procedure, appears to be too optimistic. Page 7-1 of NUREG/CR-4772 states that: critical parameter which operating personnel must commit to memory, use the lower bound values in Figure 7-1. only if those parameters can be classified as skill-based behavior per Table 2-1, otherwise use the nominal values. (The ASEP approach defines skill-based actions as being highly practiced). Use of the lower bound rather than the nominal curve to determine these values for FPRA human failure events (HFEs) is questionable as fire RAs are more complex and less practiced than the internal event RAs addressed by ASEP, even when offset with time reductions of 10 and 20 minutes (for CR and ex-CR actions). Either, demonstrate how SR HR-G3 is being met using ASEP to determine the

HEP for this set (i.e., under 60 minutes available) of cognitive errors, show that treatment of these errors would have negligible impact on the FPRA, or determine the impact of this treatment of cognitive error on fire CDF and LERF, and CDF and LERF.

Probabilistic Risk Assessment RAI 41.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 41 and provided a general discussion on dampers. In the LAR it is noted for engineering evaluation ID FPE-B97-052, FIRE BARRIER EVALUATION 10A TO 12B, Attachment C page 18, that some ventilation ducts have had the fire dampers sealed open.

However, Table C-3 of the updated FSR takes credit for a fire damper between PAU 10A and PAU 12B. Discuss the FPRA modeling assumptions and impacts of sealed open versus unsealed fire dampers between these two fire areas.

Probabilistic Risk Assessment RAI 43.01 In your letter dated May 23, 2012, (ADAMS Accession No. ML12146A094) you responded to PRA RAI 43 and noted that high pressure core injection (HPCI) and reactor core isolation cooling (RCIC) are not credited in the NSCA; therefore it is necessary to lower reactor pressure for inventory control.

a. Discuss the assessment performed for fire-related affects, on the depressurization function for essential switchgear room fire MCAs.

b. Discuss the specific reason that HPCI and RCIC failed during these scenarios. Was it the loss of DC power due to battery depletion, fire damage to DC power, fire damage to HPCI and RCIC or other reasons?

c. Discuss how the ability to depressurize the reactor is modeled for these fire scenarios, including dependencies on DC power and operator actions.

Probabilistic Risk Assessment RAI 44.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 44 which included the following text:

In conclusion, the delta risk calculations in the fire risk calculations would only be affected if:

a) the larger zone of influence included additional VFDRs, or b) the larger zone of influence included a VFDR in question that had not been affected in that particular fire scenario because of a smaller zone of influence.

Clarify what b) means.

Probabilistic Risk Assessment RAI 47.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 47 and noted that cables were not traced for the containment isolation signal (CIS) system, but were traced for CIVs. NUREG/CR-6850, Section 3.5.2.2, provides high level guidance on addressing the CIS. With respect to modeling the containment instrumentation and systems which are important for containment modeling, address the following:

a. Describe the failure modes modeled for CIS if fire damages these cables. Describe whether this has an impact on the CIVs failure to receive isolation signal probability modeled for random failures.

b. CIS impact appears to be manually excluded for battery room fires according to the FSR. Explain this modeling assumption.

c. In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 57 which provides a basic event Containment Isolation Signal Fails. Describe whether this is a system-level basic event which models impacts on multiple CIVs as discussed in NUREG/CR-6850, Section 3.5.5.5.

d. In addition, a review of the instrumentation available to operators in Table 3.3-1 of the fire model fevelopment report did not find drywell pressure. Clarify if drywell pressure instrumentation fire impacts are included in the FPRA. If not, discuss why not.

Probabilistic Risk Assessment RAI 48.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 48.

a. The response to the RAI appears to contain a conflict. In the first paragraph it states "Process monitoring of suppression (or torus) pool level is required for the ASCS" while the second paragraph states that since the monitoring of suppression pool level is not called for in AOP 915,"... the Fire PRA does not credit suppression pool level instrumentation for any of the operator actions." Clarify why suppression pool level monitoring is not included in AOP 915, and therefore not included in the PRA, yet is included as required for the ASCS in the updated final safety analysis report (UFSAR).

Review of Table G-1 in the LAR also noted that there is a primary control station action to monitor torus level (CB1 LI 4363A).

b. Further, the torus pressure instrumentation was noted to be monitored in Attachment C of the LAR for process monitoring. However, a review of Table G-1 in the LAR does not show torus pressure as being involved in any RAs or activities occurring at a primary control station. Describe if torus pressure is modeled in the FPRA. If not, describe why not.

Probabilistic Risk Assessment RAI 55.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 55 and stated:

"Given the recovery actions to establish alternate shutdown were treated as being completely dependent, an HRA was not required for those actions not required in the short term given the short term actions are most likely impacted by fire events."

Clarify what is meant by "an HRA". Describe if it is the assessment of the impact of fire on the non-fire human reliability analysis (HRA). Describe whether the HRA is considered for long term RAs for situations where short term RAs are successful.

Probabilistic Risk Assessment RAI 57.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 57 and stated that the HRA for the Level 1 and Level 2 operator action to manually depressurize relies on reactor pressure vessel (RPV) level as the instrument cue which has been verified available in each fire area. However, experience shows (i.e., Fukushima) that if there is boiling occurring in the reference legs of the reactor, which may occur during a severe accident (in Level 1 portion of analysis) that water level instrumentation provides non-conservative water level indication. Describe the basis for relying on RPV level as the instrument cue in the Level 2 model even if the instrumentation is not impacted by the fire, but is potentially providing incorrect information due to the above condition.

Probabilistic Risk Assessment RAI 58.01 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 58. In reviewing the response, PSAG-2 does not appear to require reviewing changes to the FPIE model for appropriate inclusion in the FPRA for the NPFA 805 application. Describe the process for reviewing changes to the FPIE for inclusion in the FPRA. Confirm that such a review has or will be done prior to transition and will be performed after transition on some periodicity.

Probabilistic Risk Assessment RAI 60.01

a. For a number of VFDRs, a PRA quantification had not been performed. Clarify if the criteria used to determine that a VFDR did not require a PRA quantification was the ZOI, or whether there were other considerations in the determination such as multiple concurrent shorts, fire zones, etc.

b. If a fire scenario involves more than one VFDR, describe whether the delta risk for the ignition source is simply the sum of the individual VFDR delta risks, or does it include synergistic affects from cables and equipment which may all be simultaneously failed by one fire.

c. In addition, the Fire Risk Evaluation report notes that scenarios that do not result in CR abandonment were not considered as part of the delta risk calculations. Explain why the non-abandonment scenarios are not included in the delta risk calculation.

d. In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 60 and indicated that RAs required to establish alternate shutdown capability were included in the FPRA, however, in LAR, Attachment G, it indicates that some RAs have been specifically included in the FPRA, while others have not been included. Clarify this apparent discrepancy. In addition, describe how the RAs were included in the delta-risk calculations for the CR abandonment scenario?

Probabilistic Risk Assessment RAI 65 As described in LAR Attachment U, the internal events peer review was originally performed in December 2007 using the combined PRA standard, ASME/ANS RA-Sa-2005, and RG 1.200, Revision 1. The subsequent focused PRA peer review was conducted in March 2011 using the most current combined PRA standard, ASME/ANS RA-Sa-2009, and RG 1.200, Revision 2. As stated in the Oct. 14, 2011 LAR supplement (ADAMS Accession No. ML1128702452), the scope of the 2011 peer review focused on the SR associated with upgrades, updates, or previous F&Os and not all the SRs previously assessed as MET during the 2007 full scope peer review were reassessed. Provide a self-assessment of the PRA model for the RG 1.200, Revision 2 clarifications and qualifications and indicate how any identified remaining gaps were dispositioned.

Probabilistic Risk Assessment RAI 66 Since the plant modification in Attachment S of the LAR have not been completed but have been credited directly or indirectly in the change-in-risk estimates provided in Attachment W, the models and values used in the PRA are necessarily estimates based on current plans. The as-built facility after the modification is completed may be different than the plans. Add an implementation item that, upon completion of all PRA credited modifications, verifies the validity of the reported change-in-risk. This item should include your plan of action should the as-built change-in-risk exceed the estimates reported in the LAR.

Probabilistic Risk Assessment RAI 67 Identify any changes made to the FPRA since the full-scope peer review that are consistent with the definition of a "PRA upgrade" in ASME/ANS-RA-Sa-2009, Addenda to ASME/ANS RA-S 2008, Standard for Level 1/Large Early Release Frequency Probabilistic Risk Assessment for Nuclear Power Plant Applications, American Society of Mechanical Engineers/American Nuclear Society, New York, NY, as endorsed by Regulatory Guide 1.200, An Approach for Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk-Informed Activities, Revision 2, March 2009. Specifically consider if the changes described in the LAR Table V-3 disposition of F&Os 4-22, 4-23, 4-25 and 4-32 are upgrades. If any such changes exist, describe what actions have been or will be implemented to address this review deficiency (i.e., lack of a focus-scope peer review when an upgrade occurs).

Probabilistic Risk Assessment RAI 68 The disposition of the peer review F&O 2-8, states that common cause failure (CCF) for fire induced failures do not impact the results. How is the CCF probability in the FPRA treated when redundancy is reduced as a result of a fire (e.g., redundancy is decreased from N to N-1)?

Probabilistic Risk Assessment RAI 69 Section 5.1.4 of the FSR, discusses treatment of sensitive electronics. From the discussion, it appears that it was assumed that since sensitive electronics are always in cabinets, protection provided by cabinets offsets the non-conservatism of using the cable damage criteria rather than a lower damage criteria for sensitive electronics based ZOI in the generic fire modeling report. Provide a revised analysis using the lower damage criteria for sensitive electronics or provide a demonstration that the use of the cable damage criteria is not non-conservative.

Probabilistic Risk Assessment RAI 70 FSR Section 5.1.5.1 states that for closed panels that are substantially sealed, damage is limited to the cabinet itself. Fire propagation from electrical cabinets is discussed extensively in FAQ 08-0042 (ADAMS Accession No. ML092110537). Summarize the guidelines for substantially sealed in FAQ 08-0042 are met.

Probabilistic Risk Assessment RAI 71 Appendix A of the FSR for the CR transient fire appears to include a 0.1 conditional probability that combustibles are stored near the specific location in the MCR. Provide justification for the use of this factor.

Probabilistic Risk Assessment RAI 72 In your letter dated April 23, 2012, (ADAMS Accession No. ML12117A052) you responded to PRA RAI 55 and stated:

"After 60 minutes, the fire is assumed to be "out".

There are non-suppression probabilities in NUREG/CR-6850 Table P-3 that extend beyond 60 minutes and the Table indicates that a minimum non-suppression probability of 1E-3 should be used. Clarify if non-suppression probabilities of less than 1E-3 were used (i.e., no fire damage possible after 60 minutes). If non-suppression probabilities of less than 1E-3 were used, provide justification for using a lower value and also perform a sensitivity study using the NRC accepted minimum value of 1E-3 and provide the resulting changes to the change in risk estimates.

Probabilistic Risk Assessment RAI 73 It is noted that the sensitivity studies described in response to PRA RAI 1 (ADAMS Accession No. ML12146A094) as well as the response to some of the other RAIs, for example: PRA RAIs:

8, 14, 20 and 35 (ADAMS Accession No. ML12146A094) may lead to the need for revised and updated FPRA and NFPA 805 LAR documentation. Describe the plans for developing the documentation for the updated models that meets the requirements of SR FQ-F1.

Probabilistic Risk Assessment RAI 74 Relative to F&O 5-26, concerning the assignment of transient influence factors, the guidance provided in Table 4.3-1 of the Plant Partitioning and Fire Ignition Frequency Development Report, 493080001.01 appears, in some instances to differ from the guidance in Table 6-3 of NUREG/CR-6850. For example, in Table 6-3 the maintenance influence factor should be based on the number of preventive maintenance/corrective maintenance (PM/CM) work orders compared to the average number of work orders for a typical compartment. The maintenance factors were assigned based on the frequency with which Mechanical/Electrical or Hotwork is performed such as occasionally (quarterly), or frequently (weekly). Further, Table 6-3 requires that to have a medium storage influence factor, all combustible/flammable material is stored in closed containers placed in dedicated fire safe cabinets and if not in a fire safe cabinet it is considered to have a high influence factor. Medium storage influence factor Table 4.3-1 appears to require storage normally in sealed drums/cabinets. Provide further discussion of the procedure for assigning influence factors and how it is consistent with NUREG/CR-6850.

Probabilistic Risk Assessment RAI 75 F&O 5-27 concerning the documentation of operator interviews was dispositioned by stating that the FPRA documentation was updated to include documentation of the operator interviews in Appendix E of the FSR. NRC staff review of Appendix E describing the operator interviews indicates that insufficient information is provided to conclude that the requirements of SR HRA-A4 are met. Specifically, the SR requires that interviews confirm the interpretation of procedures relevant to actions identified in SRs HRA-A1, HRA-A2 and HRA-A3 are consistent with plant operation and training. These SRs identify for each scenario: safe shutdown actions carried over from the FPIE PRA, new fire specific safe shutdown actions and new undesired actions that could result from failure of single instrument. Provide further discussion and documentation of meeting SR HRA-A4.

Probabilistic Risk Assessment RAI 76 It is noted that a fire in a PAU may result in an increase in the general environmental temperature in the area beyond that for which the equipment in the area is environmentally qualified. This may then lead to failures of this equipment, particularly for those requiring active room cooling, even if the item (and its normal room cooling) is outside of the ZOI for the fire.

Describe whether this has been considered in the NFPA 805 analysis and if not, why not.

Probabilistic Risk Assessment RAI 77 The disposition of F&O 5-34 indicated that the FPRA HEP consistency review was documented by Table E-4 added to the FSR. While this table appears to address the consistency of FPRA HEPs and FPIE HEP, the consistency of the FPRA HEP relative to each other as required by SR HR-G6 is not specifically included. Discuss how the consistency review requirement of SR HR-G6 is met for CCII.

Probabilistic Risk Assessment RAI 78 It is noted that the LAR identifies certain motor operated valves (MOVs) which are subject to spurious operation as described in Information Notice (IN) 92-18, Potential For Loss Of Remote Shutdown Capability During A Control Room Fire. These valves include steam supply valves in the HPCI and RCIC systems, Core Spray, residual heat removal (RHR) system, as well as others. Describe whether any CIVs are included as a IN 92-18 MOV in the FPRA, and, if so, describe the assumptions and methods that are applied in modeling them. Describe whether all the IN 92-18 MOVs are included in the FPRA and how they are addressed for this application (i.e., are they treated as VFDRs). Also, it is noted that Appendix S of the LAR does not show modifications for IN 92-18 valves. Confirm or clarify if there are no modifications associated with the IN 92-18 valves.

Probabilistic Risk Assessment RAI 79 Appendix G of the FSR provides fire damage probability for single cable bundles and multiple cable bundles. According to Section 5.1.5.1, single cable bundle results are used for ventilated/open high voltage switchgear, MCC, and AC/DC distribution panels, while multiple cable bundle results are used for ventilated/open load centers and other type of electric panels.

Provide the basis for the assignment of single or multiple cable bundle results to the specific cabinet types.

Probabilistic Risk Assessment RAI 80 Table V-1 of the LAR shows that the peer review had findings on SRs UNC-A1 and UNC-A2, which had been addressed. However, Table V-3 of the LAR does not include these. Clarify or provide the corresponding Table V-3 information for these SRs.

Probabilistic Risk Assessment RAI 81 Explain why SR FSS-C3 was determined to be not applicable in the LAR Table V-1.