ML14014A088

Browns Ferry, Units 1, 2 and 3, Response to NRC Request for Additional Information Regarding the License Amendment Request to Adopt NFPA 805 Performance-Based Standard for Fire Protection for Light Water Reactor Electric Generating Plant

ML14014A088
Person / Time
Site:	Browns Ferry
Issue date:	01/10/2014
From:	Shea J W Tennessee Valley Authority
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
CNL-14-001, TAC MF1185, TAC MF1186, TAC MF1187
Download: ML14014A088 (85)

Text

L44 140110 005 Tennessee Valley Authority, 1101 Market Street, Chattanooga, Tennessee 37 402 CNL-14-001 January 10, 2014 10 CFR 50.90 ATTN: Document Control Desk U.S. Nuclear Regulatory Commission Washington, D.C. 20555-0001 Browns Ferry Nuclear Plant, Units 1, 2, and 3 Renewed Facility Operating License Nos. DPR-33, DPR-52, and DPR-68 NRC Docket Nos. 50-259, 50-260, and 50-296

Subject:

Response to NRC Request for Additional Information Regarding the License Amendment Request to Adopt NFPA 806 Performance-Based Standard for Fire Protection for Light Water Reactor Electric Generating Plants for the Browns Ferry Nuclear Plant, Units 1, 2, and 3 (TAC Nos. MF1186, MF1186, and MF1187) -Set 2

References:

1. Letter from TVA to NRC, "License Amendment Request to Adopt NFPA 805 Performance-Based Standard for Fire Protection for Light Water Reactor Electric Generating Plants (2001 Edition)

(Technical Specification Change TS-480),"

dated March 27, 2013 (ADAMS Accession No. ML 13092A393)

Printed on recycled paper 2. Letter from TVA to NRC, "Response to NRC Request to Supplement License Amendment Request to Adopt NFPA 805 Performance-Based Standard for Fire Protection for Light Water Reactor Electric Generating Plants for the Browns Ferry Nuclear Plant, Units 1, 2, and 3 (TAC Nos. MF1185, MF1186, and MF1187),"

dated May 16, 2013 (ADAMS Accession No. ML 13141A291)

3. Letter from NRC to TVA, "Browns Ferry Nuclear Plant, Units 1, 2, and 3-Request for Additional Information Regarding License Amendment Request to Adopt National Fire Protection Association Standard 805 Based Standard for Fire Protection for Light Water Reactor Generating Plants (TAC Nos. MF1185, MF1186, and MF1187),"

dated November 19, 2013 (ADAMS Accession No. ML 13298A702)

U.S. Nuclear Regulatory Commission Page 2 January 10, 2014 By letter dated March 27, 2013 (Reference 1 ), Tennessee Valley Authority (TVA) submitted a license amendment request (LAR) for Browns Ferry Nuclear Plant (BFN), Units 1, 2, and 3, to transition to National Fire Protection Association Standard (NFPA) 805. In addition, by letter dated May 16, 2013 (Reference 2), TVA provided information to supplement the Reference 1 letter. By letter dated November 19, 2013 (Reference 3), the Nuclear Regulatory Commission (NRC) requested additional information to support the review of the LAR. The required dates for responding to the requests for additional information varied from a nominal 60 days to 120 days. Enclosure 1 provides the second set of TVA responses to some of the requests for additional information (RAis) identified in the Reference 3 letter. This enclosure provides some of the nominal 60 day responses that have not previously been submitted.

As stated in the Reference 3 letter, these nominal 60 day responses are due by January 14, 2014. Furthermore, this enclosure provides some of the nominal 90 day responses

, which are due by February 13, 2014. Enclosure 2 provides a listing of all RAis listed in the Reference 3 letter and the actual date of the TVA response to each of the RAis. Consistent with the standards set forth in Title 10 of the Code of Federal regulations (1 0 CFR), Part 50.92(c), TVA has determined that the additional information

, as provided in this letter, does not affect the no significant hazards consideration associated with the proposed application previously provided in Reference 1. There are no regulatory commitments contained in this submittal.

Please address any questions regarding this submittal to Mr. Edward D. Schrull at (423) 751-3850.

I declare under penalty of perjury that the foregoing is true and correct.

Executed on this 10th day of January 2014. Respectfully

,

Enclosures:

1. TVA Responses to NRC Request for Additional Information

Set 2 (nominal 60-day and 90-day) 2. Summary of BFN NFPA 805 RAI Response Dates cc (Enclosures)

NRC Regional Administra tor-Region II NRC Senior Resident Inspector-Browns Ferry Nuclear Plant State Health Officer, Alabama State Department of Health ENCLOSURE 1 Tennessee Valley Authority Browns Ferry Nuclear Plant, Units 1, 2, and 3 TVA Responses to NRC Request for Additional Information:

Set 2 (nominal 60

-day and 90-day) FPE RAI 03 LAR Attachment A, Table B

-1, Section 3.5.5, Water Supply Pump Separation, states that "Complies by Previous Approval" is the compliance strategy

. The documentation for such approval seems to be a general statement indicating that the safety evaluation approved the submitted program

. Provide more details, specifically regarding the description of fire pump separation from each other, as well as the rest of the plant, and the specific Nuclear Regulatory Commission (NRC) approval chain of correspondence and approvals that lead to this conclusion.

RESPONSE

Browns Ferry Nuclear Plant (B FN), Units 1, 2, and 3 have three electric motor

-driven fire pumps located in the Intake Pumping Station (i.e., Fire Area (FA) 25

-1) and one diesel

-driven fire pump located at the Number (No.)

2 Gate Structure (i.e., FA Yard). All four pumps provide high pressure fire water to the main fire header

. The three electric fire pumps are not spatially separated from each other but are separated from the diesel

-driven fire pump

. The Intake Pumping Station and the No

. 2 Gate Structure are separated from each other by over 400 feet (ft). The Intake Pumping Station and the No

. 2 Gate Structure are separated from the Reacto r Building by over 200 ft

. All fire scenarios other than the maximum switchyard fire (i.e., FA Switch) require only one operating fire pump to provide sufficient suppression capability

. For the maximum switchyard fire, two pumps are required

. All four fire pumps are available for any fire in FA Switch

. For FA Yard fires, all three electric fire pumps are available

. For a FA 25

-1 fire, the diesel

-driven fire pump is available

. At least one fire pump is available for all other fire areas. In a letter to the Nuclear Regulatory Commission (NRC) dated January 15, 1992, BFN submitted the Fire Protection Report (i.e., a Design Basis Document), that was supplemented with a comparison of BFN to Branch Technical Position (BTP) Chemical Engineering Branch (CMEB) 9.5-1. BTP CMEB 9.5

-1, Sections C.6.b(6)(a) and (b) state in part "Each pump and its driver and controls should be located in a room separated from the remaining fire pump by a fire wall with a minimum rating of three hours.

" This part of BTP CMEB 9.5-1, Sections C.6.b(6)(a) and (b) is equivalent to NFPA 805 Section 3.5.5

. In response to Sections C.6.b(6)(a) and (b), BFN discusses that there are three 100% capacity electric fire pumps that are physically separated from the one 100% diesel fire pump

. The NRC responded in a letter to BFN dated March 31, 1993, documenting the NRC Safety Evaluation (SE) of the Fire Protection Report (FPR). Section 2.14 of the NRC S E, titled "Fire Protection Water Supply Systems," addresses this fire pump configuration

. Within Section 2.14 of the NRC S E, numerous aspects of the system, including number of pumps, size and rating of pumps, and fire main design are listed

. In the NRC SE Section 3.0, "Conclusions," it is stated that the "licensee's Fire Protection Plan and Fire Hazards Analysis described in BFN

-FPR was reviewed and found acceptable.

" The E1-1 NRC SE Section 3.0 concludes by stating, "Consequently, based on this SE and the SEs mentioned above, the staff concludes that TVA's fire protection program described by the BFN-FPR submitted on January 15, 1992, conforms with BTP CMEB 9.5

-1 and is therefore acceptable.

" The current configuration of the fire pumps is identical to the configuration that existed at that time

.

Therefore, the "complies with previous approval" statement in LAR Attachment A, Section 3.5.5, Water Supply Pump Separation, relies on the existing four pump configuration, NRC approval of TVA's fire protection program's conformance with BTP CMEB 9.5

-1, and the equivalency of BTP CMEB 9.5-1, Sections C.6.b(6)(a) and (b) to NFPA 805 Section 3.5.5.

E1-2 FPE RAI 04 LAR Attachment S, Table S

-3, Implementation Item 44, states "Implement corrective actions as required to ensure that pressure is maintained in the fire protection system during normal operation without using a fire pump.

" Provide a more detailed description of this issue, and the intended/proposed resolution

. If necessary, explain the type of modification to the system this would entail

. If a modification is indicated, update LAR Attachment S, Table S-2 accordingly.

RESPONSE

High Pressure Fire Protection (HPFP) System pressure is maintained by an interconnection to the raw service water (RSW)

System. The National Fire Protection Association (NFPA) Standard 20 code compliance review identified that a fire pump is often utilized to supplement the RSW System when the RSW System cannot meet the total demand

. Challenges to RSW System demand have occurred when cooling towers are in service, which have resulted in inadequate flow and pressure to the bearing lubricating water supply for the cooling tower lift pumps. Design Change Notices (DCNs) 70332, 70337 and 70351 have been implemented to install duplex strainers upstream of the cooling tower lift pump stations to improve RSW System performance

. Subsequently, a test of the RSW System will be performed to confirm that these modifications were successful in precluding the need to run a fire pump to support system load.

License Amendment Request (LAR) Attachment S, Table S

-3, Implementation Item 44 is revised to state the following:

"Implement corrective actions as required to ensure that pressure is maintained in the fire protection system during normal operation without using a fire pump

. In addition, perform testing of the Raw Service Water System to confirm that pressure is maintained in the fire protection system during normal operation without using a fire pump." E1-3 FPE RAI 05 LAR Attachment V, Fire PRA [Probabilistic Risk Assessment] Quality,Section V.2.5, states that

"-in some cases, the 1

-hour rated ERFBS [electrical raceway fire barrier system] may not have automatic suppression

. In those instances, an Engineering Equivalency Evaluation will be performed to determine that the 1

-hour rated ERFBS is adequate for the hazard." Describe the method used from a regulatory standpoint for resolution of these identified ERFBS conditions including the use of variance from deterministic requirements (VFDRs) for identification, resolution, and evaluation of the safety margin and defense

-in-depth (DID)

. Describe how these ERFBS resolutions were identified (e.g., Table C

-2). If VFDRs were not identified, provide justification for not using the fire risk evaluation (FRE) process.

RESPONSE

There are six instances involving 1

-hour ERFBS in fire areas that do not have automatic suppression

. These are shown in the Table below

. In each of these instances, the ERFBS is not yet installed and a separation issue is tracked by one or more VFDRs identified in th e table below. The analysis results presented in the LAR were based on resolving the VFDRs by installing 1

-hour ERFBS and an Engineering Equivalency Evaluation (EEE) to demonstrate that the ERFBS is adequate for the hazard without credit for suppression

. This method resulted in a deterministic resolution of the VFDR, wherein the probability of ERFBS failure is not reflected in the delta Core Damage Frequency (

CDF) for the fire area.

Fire Area Cable Protected Associated Equipment Configuration VFDR No. LAR Table S

-2 Item No. 5 PP626-IB ES50-I 4kV Shutdown (Sd) Board (Bd) B AC load cables (two cables to prevent fault propagation) 1-hour rated ERFBS, Detection, Manual suppression VFDR-05-0006 87 12 3B180-B1 4kV Sd Bd 3EA DC control power cable 1-hour rated ERFBS, Detection, Manual suppression VFDR-12-0006 14 13 3PL6363-II 3PL451-II 480v Sd Bd AC power cables 1-hour rated ERFBS, Detection, Manual suppression VFDR-13-0005 VFDR-13-0008 17 21 3B188-B3 4kV Sd Bd 3EC control power cable 1-hour rated ERFBS, Detection, Manual suppression VFDR-21-0006 91 E1-4 Fire Area Cable Protected Associated Equipment Configuration VFDR No. LAR Table S

-2 Item No. 21 3B193-B2 4kV Sd Bd 3ED control power cable 1-hour rated ERFBS, Detection, Manual suppression VFDR-21-0007 92 23 3B180-B1 4kV Sd Bd 3EA DC control power cable 1-hour rated ERFBS, Detection, Manual suppression VFDR-23-0009 14 TVA has re

-considered the use of EEEs for this application and has decided not to disposition 1-hour ERFBS without automatic suppression as adequate for the hazard. The current plan for these applications is to install 1-hour ERFBS and to resolve the VFDRs using the fire risk evaluation process

. Documentation affected by this change will be revised

. In addition, LAR Attachment V,Section V.2.5 is revised by deleting the fourth sentence of the first paragraph and replacing it with the following:

"In those instances, the separation issues will be resolved using the fire risk evaluation process."

Thus, the first paragraph of LAR Attachment V,Section V.2.5 now reads: "BFN has Electrical Raceway Fire Barrier Systems (ERFBS) installed, and modifications planned to install ERFBS, in accordance with NFPA 805 Chapter 3 Section 3.11.5 to meet the separation requirements of NFPA 805 Chapter 4 Section 4.2.4

. In most cases the ERFBS are 1

-hour rated and are or will be installed in areas where automatic detection and suppression is available. In some cases, the 1

-hour rated ERFBS may not have automatic suppression

. In those instances, the separation issues will be resolved using the fire risk evaluation process

. The ERFBS are referenced in Tables C-1 and C-2 as a required Fire Protection System/Feature for the Fire Area

. In addition, the engineering equivalency evaluation is referenced in Tables C

-1 and C-2." Separation issues and resolutions which involve ERFBS are documented in the Fire Risk Evaluation calculation

. LAR Attachment C, Table C

-2 also identifies that credit will be taken for a modification to install ERFBS as a fire protection feature and identifies resolution by denoting the ERFBS modification as "Separation" or "Risk." After this change in method is implemented, the affected sections in the LAR will be revised to reflect the results of the updated fire risk evaluations

. The revision to the LAR will be provided to the NRC after the Fire PRA is updated and additional quantification is performed in response to remaining NRC Requests for Additional Information (RAIs). E1-5 SSA RAI 05 LAR Section 4.2.1.2 states that the at

-power analysis is based upon the ability to achieve and maintain hot shutdown conditions (Mode 3).

a. Describe the specific capabilities that will be required to meet the performance criteria beyond 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />

. One example is that "Liquid nitrogen tanks for safety relief valve pneumatics are sufficient for at least 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> after which sources may be required." b. Describe any system or component capacity limitations and time

-limited actions needed to replenish systems, make repairs, or otherwise maintain safe and stable conditions, (e.g. DC battery power, diesel generator fuel, water inventory, etc.).

c. Describe whether there are any actions to recover/repair nuclear safety capability assessment (NSCA) equipment to sustain safe and stable conditions

. Describe the resource (staffing) requirements and timing of these actions.

d. Describe how the feasibility of the actions in items b and c (above) was evaluated or addressed.

e. Provide a more detailed discussion (qualitative or quantitative) of the risk of failure of the actions necessary to sustain safe and stable conditions beyond 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.

f. LAR Section 4.2.1.2 states that off

-site power is available to provide long term power

. Explain the basis for this statement

. Describe what repairs are necessary to establish this availability

. RESPONSE:

Part a. As stated in the LAR, Attachment B, Item 3.1.1.9 - 72 Hour Coping, the Nuclear Safety Capability Assessment (NSCA) is not limited to a 72

-hour coping period

. As such, the credited success paths for achieving the Nuclear Safety Performance Criteria of NFPA 805, Section 1.5.1 remain the same throughout the event (i.e., the credited success paths do not change after 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />)

. The following is a description of the NSCA

-credited systems and capabilities that are required beyond 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.

1.5.1(a) - Reactivity Control After a reactor scram has occurred and control rods are inserted, subcritical conditions are achieved for all reactor conditions and reactivity control is a passive function

. No further capabilities are required to maintain the reactor subcritical. 1.5.1(b) - Inventory and Pressure Control Safety Relief Valves (SRVs) are credited for pressure control and low pressure systems using re-circulated water from the suppression pool are credited for inventory control.

1.5.1(c) - Decay Heat Removal Residual Heat Removal (

RHR) pumps/RHR heat exchangers and RHR Service Water (RHRSW) pumps are credited for decay heat removal

. RHR may be in Suppression Pool Cooling or Alternate Shutdown Cooling mode

. 1.5.1(d) - Vital Auxiliaries

Emergency Equipment Cooling Water (EECW) is credited for providing cooling to heating ventilation and air conditioning (HVAC) equipment and diesel generators (DGs). HVAC systems E1-6 are credited for cooling vital equipment spaces

. The liquid nitrogen tanks are credited to supply pneumatic pressure for SRVs.

Offsite alternating current (AC) power or D Gs are credited for supplying AC power and sustaining direct current (DC) power by supplying power to the battery chargers.

The Drywell Spray mode of RHR is also credited in some fire areas for Drywell Cooling.

1.5.1(e) - Process Monitoring

The process monitoring instruments credited in the NSCA remain the same throughout the event. Part b. DGs, drywell pneumatic supplies, batteries, and Reactor Pressure Vessel (RPV) makeup water are credited in the NSCA and have limited capacities. DGs require replenishment of fuel oil and lubricating oil after seven days

. Each DG has a seven day fuel oil tank which is routinely replenished from onsite storage tanks by Operations personnel

. In the event that these tanks and associated equipment are not useable, each engine seven day tank can be filled directly from a delivery truck

. This procedure is simple and is included in normal operating instructions as an infrequent operation

. Actions required to replenish lubricating oil are routine activities involving transfer of oil from containers to the engine by maintenance personnel

. Lubricating oil may be procured from offsite sources. Drywell pneumatic supply for the SRVs is maintained by liquid nitrogen storage tanks, which have a capacity of at least 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />

. The tanks are replenished by filling the tanks directly from a delivery truck

. This activity is performed by Operations personnel and is routine. Batteries credited in the NSCA beyond their calculated depletion time are placed on chargers prior to their calculated depletion time

. Restoration of chargers is required in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />) and is performed by Operations personnel as part of the recovery actions included in LAR Attachment G. Safe shutdown paths in the NSCA which utilize high pressure makeup from the High Pressure Coolant Injection (HPCI) System or the Reactor Core Isolation Cooling (RCIC) System are dependent on water from the condensate storage tanks (CSTs), which have limited capacity

. For those safe shutdown paths where high pressure inventory makeup is available, prior to the CSTs becoming depleted, the safe shutdown analysis assumes transition to low pressure injection using water from the suppression pool (i.e., transition is made from high pressure inventory makeup to low pressure inventory makeup)

. Water required for RPV makeup when using low pressure injection is then re

-circulated through the suppression pool

. Therefore replenishment of makeup water is not required.

Part c.

There are no repair actions credited in the NSCA to sustain safe and stable conditions

. Recovery actions are listed in LAR Attachment G

. Actions required to maintain systems or components with limited capacity are described in response to Part b above. E1-7 Part d. Feasibility of recovery actions in LAR Attachment G is addressed in accordance with Frequently Asked Question (FAQ)07-030. Actions to maintain safe and stable conditions as discussed in response to Part b above, were qualitatively determined to be feasible given the time available

. These are simple or routine activities that are covered by normal plant procedures, work processes and training

. Procurement and delivery of fuel oil, lubricating oil and nitrogen from offsite sources is routine and easily achievable given the time available. Part e.

The actions beyond 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> discussed in response to Part b above are qualitatively determined to be very low risk based upon the nature of the activities and the amount of time available

. In addition, assistance will be available long term in the form of trained emergency response personnel

. The availability of additional resources further ensure s that long-term actions will be reliably accomplished.

Part f.

The intent of the statement in LAR Section 4.2.1.2 is that Offsite AC Power would be available in the long term (i.e., >

24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />) if it is credited in the NSCA short term (i.e.,

24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />). No credit is taken for long term recovery of Offsite AC Power and no repairs are necessary or credited.

E1-8 SSA RAI 7 LAR Attachment S, Table S

-3, Items 27, 28, 29, 30, 31, 32, and 33, indicate the need to develop or revise post

-fire response procedures, identify required tools during procedure validation and verification, document new staffing requirements, conduct training, and update the human reliability analysis (HRA) and FPRA model

. Additionally, the methods of alternate and safe shutdown process will be transitioning from the "self-induced station blackout" (SISBO) method to the NFPA 805 post transitioned fire safe shutdown procedures

. There are also new success paths including an emergency high pressure makeup pump, and the condensate and booster pump feedwater injection success paths

. RAs have not been fully evaluated for feasibility because procedures have not been completed

. These outstanding work activities present a potential for significant impacts to the current NSCA presented in the LAR

. Provide justification regarding the confidence that the results presented in the LAR will not significantly change.

RESPONSE

Changes to the analysis presented in the LAR are expected when fire safe shutdown procedures and modifications are developed

. The analysis to support the LAR was performed assuming use of the new methods for fire safe shutdown, including equipment and procedures

. Recovery actions and attributes for new or modified equipment were developed and assumed in the modeling

. These assumptions were based on initial modification scoping or knowledge of how the equipment is currently operated in fire safe shutdown and other emergency procedures

. Subsequent development of modifications and procedures will focus on satisfying the analysis assumptions.

Although the details of procedures and equipment modifications are not finalized, TVA does not anticipate that they will differ significantly from the way systems and equipment are currently utilized and operated in fire safe shutdown and other emergency procedures

. For example, the condensate and condensate booster pumps are utilized as an injection source in the symptom-based Emergency Operating Instructions (EOIs) similar to the way they will be used in the new fire safe shutdown procedures

. Therefore, these changes are not expected to significantly affect risk, defense in depth or safety margin presented in the LAR.

LAR Attachment S, Table S

-3, Implementation Items 32 and 33 address the requirement to update both the Fire Probabilistic Risk Assessment (PRA) and Human Reliability Analysis (HRA) to reflect the effect of the finalized modifications and procedures

. In the TVA response to PRA RAI 14 contained in this enclosure, TVA is changing the wording of Implementation Item 33 to clearly state that the update to the HRA analyses will include a verification of the validity of the reported change in risk on as

-built conditions after the procedure updates, modifications, and training are completed.

E1-9 FM RAI 01.b.iii NFPA 805, Section 2.4.3.3, states:

"The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [authority having jurisdiction] ..." The NRC staff noted that fire modeling comprised the following:

The algebraic equations implemented in FDTs [Fire Dynamics Tools] and Fire Induced Vulnerability Evaluation, Revision 1 (FIVE) were used to characterize flame radiation (heat flux), flame height, plume temperature, ceiling jet temperature, and hot gas layer (HGL) temperature.

The Consolidated Model of Fire and Smoke Transport (CFAST) was used in the multi-compartment analysis (MCA), and for the temperature sensitive equipment hot gas layer study.

Fire Dynamics Simulator (FDS) was used to assess the MCR habitability, and in the plume/hot gas layer interaction and temperature sensitive equipment ZOI studies.

Section 4.5.1.2, "Fire PRA" of the LAR states that fire modeling was performed as part of the FPRA development (NFPA 805 Section 4.2.4.2)

. Reference is made to Attachment J, "Fire Modeling V&V, [Validation & Verification]" for a discussion of the acceptability of the fire models that were used.

Regarding the acceptability of the PRA approach, methods, and data:

b. Regarding fire propagation in cable trays:

iii. Explain if and how the presence of holes in cable tray covers was accounted for.

RESPONSE

Credited cable tray covers were reviewed for holes by performing plant walkdowns

. The plant walkdowns confirmed that all sections of cable tray covers, credited in the Fire PRA analysis to delay ignition or damage to cables, are robust and without holes

. Therefore, holes in the cable tray covers were not applicable to the fire modeling analyses.

E1-10 FM RAI 01.d.i NFPA 805, Section 2.4.3.3, states:

"The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [authority having jurisdiction] ..." The NRC staff noted that fire modeling comprised the following:

The algebraic equations implemented in FDTs [Fire Dynamics Tools] and Fire Induced Vulnerability Evaluation, Revision 1 (FIVE) were used to characterize flame radiation (heat flux), flame height, plume temperature, ceiling jet temperature, and hot gas layer (HGL) temperature.

The Consolidated Model of Fire and Smoke Transport (CFAST) was used in the multi-compartment analysis (MCA), and for the temperature sensitive equipment hot gas layer study.

Fire Dynamics Simulator (FDS) was used to assess the MCR habitability, and in the plume/hot gas layer interaction and temperature sensitive equipment ZOI studies.

Section 4.5.1.2, "Fire PRA" of the LAR states that fire modeling was performed as part of the FPRA development (NFPA 805 Section 4.2.4.2)

. Reference is made to Attachment J, "Fire Modeling V&V, [Validation & Verification]" for a discussion of the acceptability of the fire models that were used.

Regarding the acceptability of the PRA approach, methods, and data:

d. The HRR of electrical cabinets throughout the plant is based on the assumption that they are closed and contain multiple bundles of unqualified cable (Case 4 in Table E-1 of NUREG/CR-6850, Vol

. 2). i. During the audit the NRC staff noted that the door of an electric cabinet in the area behind the Unit 1 MCR horseshoe was partially open

. Provide justification for the assumption that there are no open cabinets in the plant.

RESPONSE

The assumption in the fire modeling analysis that there were no open cabinets was based on plant procedures and personnel expectations

. Plant procedures require that proper housekeeping be maintained related to fire prevention, fire protection, and protection of equipment

. Also, personnel are expected to report equipment problems, personnel hazards, and material condition deficiencies, following certain guidelines, when conditions cannot be corrected on the spot.

In addition, the fire modeling assumptions regarding the condition of cabinet doors will be included in the monitoring program

. LAR Attachment S, Table S

-3 describes the Implementation Items that will be completed prior to the implementation of the new NFPA 805 fire protection program

. A new implementation item, Implementation Item 46, is added to LAR Table S-3. The new Implementation Item reads: "Verification of the condition of electrical cabinet doors to meet Fire Modeling Assumptions will be included in the monitoring program." Based on current plant procedures and requirements, and future updates to the monitoring program, the observation of an open cabinet door during the audit walkdowns does not invalidate the assumption used in the fire modeling analysis.

E1-11 FM RAI 01.d.ii NFPA 805, Section 2.4.3.3, states:

"The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [authority having jurisdiction] ..." The NRC staff noted that fire modeling comprised the following:

The algebraic equations implemented in FDTs [Fire Dynamics Tools] and Fire Induced Vulnerability Evaluation, Revision 1 (FIVE) were used to characterize flame radiation (heat flux),

flame height, plume temperature, ceiling jet temperature, and hot gas layer (HGL) temperature.

The Consolidated Model of Fire and Smoke Transport (CFAST) was used in the multi-compartment analysis (MCA), and for the temperature sensitive equipment hot gas layer study.

Fire Dynamics Simulator (FDS) was used to assess the MCR habitability, and in the plume/hot gas layer interaction and temperature sensitive equipment ZOI studies.

Section 4.5.1.2, "Fire PRA" of the LAR states that fire modeling was performed as part of the FPRA development (NFPA 805 Section 4.2.4.2)

. Reference is made to Attachment J, "Fire Modeling V&V, [Validation & Verification]" for a discussion of the acceptability of the fire models that were used.

Regarding the acceptability of the PRA approach, methods, and data:

d. The HRR of electrical cabinets throughout the plant is based on the assumption that they are closed and contain multiple bundles of unqualified cable (Case 4 in Table E-1 of NUREG/CR-6850, Vol

. 2). ii. During the audit the NRC staff noted that several electrical cabinets in the auxiliary instrument rooms have Plexiglass doors

. Describe and technically justify the HRR that was used for these cabinets.

RESPONSE

Two electrical cabinets in each of the auxiliary instrument rooms have plexiglass doors

. These electrical cabinets were analyzed separately as part of the scoping fire modeling analysis and the multi

-compartment analysis

.

In the scoping fire modeling analysis, the fire scenarios for these electrical cabinets conservatively assume damage to all targets in the fire compartment, if the fire is able to spread outside the cabinet

. The fire is able to spread outside the cabinet only if it is not suppressed i n its incipient phase

. This methodology bounds the use of a heat release rate (

HRR) to determine target damage

.

For the multi

-compartment analysis (MCA), the electrical cabinets were modeled using the 98th percentile HRR for a vertical cabinet with unqualified cables, a fire in more than one cable bundle, and having closed doors

. The resulting HRR is 464 kilowatt (

kW), in accordance with NUREG/CR-6850, Table G-1. To technically justify the 464 kW HRR used in the MCA for these cabinets, the following calculation analyzes the combustibles within the cabinet (i.e., internal cables and the plexiglass door) to determine an appropriate HRR for these cabinets assuming the door is open. Each cabinet with a plexiglass door has a relatively small number of cables contained within the cabinet that were conservatively assumed to be thermoplastic

. Based on walkdowns of these cabinets, a full six-inch wide cable tray riser with a length equal to the height of the cabinet is a E1-12 bounding representation of the number of cables inside the cabinet

. Therefore, the HRR of the internal cable load of the cabinet can be bounded by calculating the complete ignition of a full six-inch wide cable tray riser with a length that is equal to the height of the cabinet (i.e., 100 inches). The calculation below determines the total HRR of a six inch wide cable tray riser based on the area of the tray and the heat release rate per unit area (HRRPUA) for thermoplastic cables (i.e., 23.2 kW/ft 2 as provided in NUREG/CR-7010, Section 9.2.2). ACABLES = 100 inches (in) x 6 in = 600 in2 = 4.17 ft2 QCABLES = ACABLES x HRRPUATP = 4.17 ft 2 x 23.2 kW/ft 2 = 96.7 kW

The HRR of the plexiglass door can be calculated by multiplying the area of the plexiglass by its HRRPUA. The Plexiglass door is 7 ft x 2 ft for a total area of 14 ft

2. The HRRPUA for plexiglass, or polymethyl methacrylate (PMMA), stored vertically is 250 kW/m2 (i.e., 23.2 kW/ft2) following the guidance in NUREG/CR

-6850, Table G-8. The following calculation determines the total HRR assuming the complete ignition of the plexiglass door

. QPLEXIGLASS

= APLEXIGLASS x HRRPUAPLEXIGLASS

= 14 ft2 x 23.2 kW/ft 2 = 324.8 kW Therefore, the combined HRR for the cables in the cabinet and the plexiglass is:

QCABLES + QPLEXIGLASS

= 96.7 kW + 324.8 kW = 421.5 kW Additional safety margin was included in the fire modeling analyses through the selection of the following multiple conservative inputs:

The fire elevation for these ignition sources was at the top of the cabinet

. This is conservative because the combustion process would occur where the fuel mixes with oxygen, which is not always at the top of the ignition source

. Because the fire elevation was located at the top of the cabinet, the above cables trays were ignited very early in the fire.

For hot gas layer calculations, no equipment or structural steel was credited as a heat sink. Fire spread to adjacent cabinets assumes the worst case cable configuration (i.e., cables are in contact with the panel sides) and used 10 minute spread time versus a 15 minute spread time in accordance with NUREG/CR

-6850, Appendix S.

Conservative screening criteria for damage temperatures and heat fluxes were used (i.e., 205°C and 6 kW/m 2 for thermoplastic cables).

Some cable trays in the area are coated with a fire resistant coating, however, this was not credited for any fire scenarios.

The HRR calculations for the six inch wide cable tray riser and full plexiglass door utilize conservative and bounding inputs to determine the total HRR of the cabinet

.

Based on the additional safety margin discussed above, and because the modeled 464 kW HRR exceeds the calculated HRR for the cabinet by 10%, the 464 kW HRR for the electrical cabinets with plexiglass doors is appropriate and bounding.

E1-13 FM RAI 01.f NFPA 805, Section 2.4.3.3, states:

"The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [authority having jurisdiction] ..." The NRC staff noted that fire modeling comprised the following:

The algebraic equations implemented in FDTs [Fire Dynamics Tools] and Fire Induced Vulnerability Evaluation, Revision 1 (FIVE) were used to characterize flame radiation (heat flux), flame height, plume temperature, ceiling jet temperature, and hot gas layer (HGL) temperature.

The Consolidated Model of Fire and Smoke Transport (CFAST) was used in the multi-compartment analysis (MCA), and for the temperature sensitive equipment hot gas layer study.

Fire Dynamics Simulator (FDS) was used to assess the MCR habitability, and in the plume/hot gas layer interaction and temperature sensitive equipment ZOI studies.

Section 4.5.1.2, "Fire PRA" of the LAR states that fire modeling was performed as part of the FPRA development (NFPA 805 Section 4.2.4.2)

. Reference is made to Attachment J, "Fire Modeling V&V, [Validation & Verification]" for a discussion of the acceptability of the fire models that were used.

Regarding the acceptability of the PRA approach, methods, and data:

f. During the audit the NRC staff noted transient combustibles in several areas that may not have been considered in the fire modeling analyses, e.g., upholstered chairs, cart s with combustibles and computer equipment in several areas; large quantities of resin near the Unit 1 oil tanks in the turbine building; self

-contained breathing apparatus cases, paper in cardboard boxes, a copier and printers in the control room areas, e tc. Describe whether the fire modeling analyses that were conducted are bounding for scenarios that involve these transient combustibles

. In addition, explain how it is ensured that the model assumptions in terms of transient combustibles in a fire area or zone will not be violated during and post

-transition.

RESPONSE

Transients were evaluated based on the 98th percentile HRR (i.e.,

317 kW) specified in NUREG/CR-6850, "EPRI/NRC-RES Fire PRA Methodology for Nuclear Power Facilities," Table G-1. The suggested 98 th percentile HRR for 98 th percentile transients listed in NUREG/CR-6850, Table G

-1 are based on tested fuel package configurations identified in NUREG/CR-6850, Table G-7. The configurations tested are various solid fuel packages comprised of materials commonly found in the plant (e.g., cardboard, paper, plastics, cotton rags, acetone)

. TVA utilizes combustible control procedures to identify areas that require enhanced combustible controls

. These procedures require that waste, debris, scraps, rags, or other combustibles resulting from work activities shall not be allowed to accumulate in any area except when placed in appropriate trash receptacles

. Therefore, the type and nature of transient combustible materials within the plant is expected to be a typical solid fuel package (i.e., a 317 kW fire as recommended by NUREG/CR

-6850). Walkdowns were performed and room usage was also considered when prescribing the transient HRR

. This provided assurance that the HRRs used for the transient fire scenarios, modeled in the Fire PRA, would be bounding.

The type and amount of transient combustibles expected to be found in a specific fire compartment at the plant, including those found during NRC walkdowns, with the exception of the resin, are bounded by the typical fuel package configurations identified in NUREG/CR

-6850, E1-14 Table G-7. Therefore, based on the guidance within NUREG/CR

-6850, the 317 kW 98 th percentile transient HRR identified in NUREG/CR

-6850, Table G

-1, is bounding for transient combustibles within the plant, with the exception of the Turbine Building

. For the Turbine Building, which is where the resin is located, the fire modeling analysis for transient combustibles conservatively assumed damage to all targets in the fire compartment

. This is a bounding analysis for the resin or any other transient combustibles located in the Turbine Building area.

Select fire compartments and transient zones have combustible control procedures, physical limitations, or a lack of maintenance activities that would preclude the presence of transient fuel packages of sufficient size necessary to create a fire that could generate a transient HRR in the 98th percentile transient HRR

. For these areas, a reduced transient HRR of 69 kW was applied to model the bounding transient fire effect. LAR Attachment S, Implementation Item 45, requires TVA to provide additional restrictions on the combustibles allowed in specific areas of the plant

. Additionally, the combustibles identified in the NRC RAI were located in either the Reactor Building or the Control Building, which are both safety

-related/critical areas that have strict combustible control requirements

. These strict combustible controls ensure that waste, debris, scraps, rags, and other combustibles are not allowed to accumulate in the area

. This HRR accounts for the placement of small amounts of transient combustibles in the area

. A review of the transient ignition source tests in NUREG/CR

-6850, Table G-7, indicates that of the type of transient fires that can be expected in these rooms (e.g., polyethylene trash can or bucket containing rags and paper) were measured at peak heat release rates of 50 kW or below. Therefore, based on the above discussion, the use of a 69 kW transient HRR bounds the expected transient fire size in these select fire compartments and transient zones

. E1-15 FM RAI 01.h.i NFPA 805, Section 2.4.3.3, states:

"The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [authority having jurisdiction] ..." The NRC staff noted that fire modeling comprised the following:

The algebraic equations implemented in FDTs [Fire Dynamics Tools] and Fire Induced Vulnerability Evaluation, Revision 1 (FIVE) were used to characterize flame radiation (heat flux), flame height, plume temperature, ceiling jet temperature, and hot gas layer (HGL) temperature.

The Consolidated Model of Fire and Smoke Transport (CFAST) was used in the multi-compartment analysis (MCA), and for the temperature sensitive equipment hot ga s layer study.

Fire Dynamics Simulator (FDS) was used to assess the MCR habitability, and in the plume/hot gas layer interaction and temperature sensitive equipment ZOI studies.

Section 4.5.1.2, "Fire PRA" of the LAR states that fire modeling was performed as part of the FPRA development (NFPA 805 Section 4.2.4.2)

. Reference is made to Attachment J, "Fire Modeling V&V, [Validation & Verification]" for a discussion of the acceptability of the fire models that were used.

Regarding the acceptability of the PRA approach, methods, and data:

h. Specifically regarding the use of the algebraic models:

i. Explain how the elevation of ignition source fires was determined, and describe whether the assumptions are consistent with plant conditions or if they lead to conservative ZOI and HGL temperature estimates.

RESPONSE

The height of each fixed ignition source was determined through plant walkdowns for all fire compartments

. The height was selected as the highest location on the source where fire or smoke could propagate

. This was typically an opening, vent, or the top of the door of a cabinet where warping could occur

. If the combustible material inside the cabinet was visible through vents and the elevation of the material could be observed, then the fire elevation was placed at the location of the vent openings

. The approach of using a fire elevation at the location of the vent openings was consistent with plant conditions

. However, when the inside of the cabinet was not visible and the approach of using the location of the vent opening could not be used, the top of the electrical cabinet was conservatively selected as the elevation of the fire

. As walked down, this approach is consistent with plant conditions.

When the height of the fire was assumed to be the top of the electrical panel or pump body, this resulted in a conservative zone of influence (ZOI) for the following reasons:

This results in a conservative plume ZOI because the damaging plume height was added to the fire elevation

. Therefore, damaging plume temperatures would affect targets located at a greater distance above the ignition source than if the fire elevation was assumed to be lower (e.g., at the level of the ignition source inside the cabinet).

This results in a conservative ceiling jet ZOI because the fire elevation was factored into the ceiling jet ZOI

. Therefore, a larger radial ceiling jet ZOI is calculated based on the higher fire elevation than if the fire elevation was assumed to be lower (e.g., at the level of the ignition source inside the cabinet).

E1-16 This results in a conservative radiant heat ZOI because the radial ZOI was applied from the floor to the top of the flame to determine target effects. Therefore, the flame height was increased when the base of the fire was located at a greater distance above the floor than if the fire elevation was assumed to be lower (e.g., at the level of the ignition source inside the cabinet)

. The fire elevation is not applicable to models that utilized the McCaffrey, Quintiere, Harkleroad (MQH) or Beyler methods to determine Hot Gas Layer (HGL) temperatures

. These methods consider the fire elevation to be zero when calculating HGL temperatures.

The height of transient fire sources composed of ordinary combustibles (i.e., paper, wood, anti

-contamination clothing, rags, and plastic) was selected as two feet for most fire compartments

. The fire origin was chosen to be two feet above the floor at the center of the postulated location

. In some fire compartments, the height of transient fire sources was selected as 3.3 ft (i.e., 1 meter). Although transient fire elevations are consistent with most plant conditions, these fire elevations were selected for conservatism

. A transient fire elevation higher than floor elevation is considered conservative because many transient fires occur at floor level

. Assuming the fire elevation at a height above the floor (e.g., 2 ft) leads to conservatism in the plume ZOI and HGL temperatures as outlined above for fixed ignition sources. E1-17 FM RAI 01.h.ii NFPA 805, Section 2.4.3.3, states:

"The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [authority having jurisdiction] ..." The NRC staff noted that fire modeling comprised the following:

The algebraic equations implemented in FDTs [Fire Dynamics Tools] and Fire Induced Vulnerability Evaluation, Revision 1 (FIVE) were used to characterize flame radiation (heat flux), flame height, plume temperature, ceiling jet temperature, and hot gas layer (HGL) temperature.

The Consolidated Model of Fire and Smoke Transport (CFAST) was used in the multi-compartment analysis (MCA), and for the temperature sensitive equipment hot gas layer study.

Fire Dynamics Simulator (FDS) was used to assess the MCR habitability, and in the plume/hot gas layer interaction and temperature sensitive equipment ZOI studies.

Section 4.5.1.2, "Fire PRA" of the LAR states that fire modeling was performed as part of the FPRA development (NFPA 805 Section 4.2.4.2)

. Reference is made to Attachment J, "Fire Modeling V&V [Validation & Verification]," for a discussion of the acceptability of the fire models that were used.

Regarding the acceptability of the PRA approach, methods, and data:

h. Specifically regarding the use of the algebraic models:

ii. Provide technical justification for using a fixed Froude number value of one to determine the diameter of the fire as opposed to the actual diameter of the fire

. Determine the range of fire diameters that correspond to a Froude number of one for the HRRs of the fires that were considered in the fire modeling analyses and show that this range is reasonably consistent with the dimensions of transient combustibles in the plant.

RESPONSE

In accordance with NUREG

-1934, Table 2-5, the typical accidental fire has a Froude number of 1.0. The fire scenarios modeled at BFN were not momentum driven (e.g., jet flares), and therefore, this value is valid

. In addition, buoyancy

-driven fire plumes, such as those involving cable insulation, lube oil, transient materials, etc., generally have fire Froude numbers toward the lower end of the validation range (i.e., 0.4)

. A Froude number equal to 1.0 is within the validation range of NUREG

-1824. Therefore, selection of a Froude number of 1.0 was conservative because a higher fire Froude number leads to more severe estimates of centerline plume temperature and flame height.

In the BFN analyses, the fire diameters that correspond to a Froude number of 1.0 range from less than 0.5 ft for a small cabinet fire in the early t² growth stage to over 10 ft for a large 100% oil spill fire

. Fire diameters determined by using a fixed Froude number of 1.0 for typical fixed sources burning at peak HRR were reasonably consistent with the dimensions of the fixed sources at BFN

. For example, 1.69 ft was used as the fire diameter for a typical 211 kW electrical cabinet fire and 2.32 ft was used for a 464 kW cabinet fire

. This equates to an area of 2.2 ft² and 4.2 ft², respectively

. These areas are reasonably consistent with the dimensions of the fixed sources because the 211 kW fire is limited to one cable bundle and the 464 kW fire involves multiple cable bundles.

For both thermoset and thermoplastic cable fires, TVA utilized the HRRPUAs prescribed in NUREG/CR-6850 and NUREG/CR

-7010. The HRRPUAs prescribed in these documents yield E1-18 a Froude number below the validation range (i.e., below 0.4)

. Therefore, using a Froude number of 1.0 in the fire modeling at BFN is conservative.

The maximum spill area as calculated by the methods found in NUREG/CR

-6850, Appendix G was used to determine the HRR for oil fires and the adjusted fire diameter was calculated based on the HRR and a Froude number of 1.0

. The fire area calculated when using the adjusted fire diameter is often below that of the original spill area used

. Although the fire area analyzed (i.e., calculated using a Froude number of 1.0) may be reduced, the approach was conservative because the reduced fire diameter coupled with the HRR lead to more severe zone of influence estimates.

For 317 kW transient fires, the fire diameter that equates to a Froude number of 1.0 is 2.0 ft

. This fire diameter was consistent with the fuel package expected for a 317 kW fire

. BFN plant procedures require that waste, debris, scraps, rags, and other combustibles resulting from work activities shall be removed from all plant operating areas, or placed in an appropriate trash receptacle right after the work is completed

. The most likely transient fire scenario would be a trash can fire.

For 69 kW transient fires, the fire diameter which equates to a Froude number of 1.0 is 1.1 ft. This fire diameter was consistent with the relatively small fuel package expected for a 69 kW fire, which was expected to be something similar to a small polyethylene trash can or bucket containing rags and paper.

E1-19 FM RAI 01.i.iii NFPA 805, Section 2.4.3.3, states:

"The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [authority having jurisdiction] ..." The NRC staff noted that fire modeling comprised the following:

The algebraic equations implemented in FDTs [Fire Dynamics Tools] and Fire Induced Vulnerability Evaluation, Revision 1 (FIVE) were used to characterize flame radiation (heat flux), flame height, plume temperature, ceiling jet temperature, and hot gas layer (HGL) temperature.

The Consolidated Model of Fire and Smoke Transport (CFAST) was used in the multi-compartment analysis (MCA), and for the temperature sensitive equipment hot gas layer study.

Fire Dynamics Simulator (FDS) was used to assess the MCR habitability, and in the plume/hot gas layer interaction and temperature sensitive equipment ZOI studies.

Section 4.5.1.2, "Fire PRA" of the LAR states that fire modeling was performed as part of the FPRA development (NFPA 805 Section 4.2.4.2)

. Reference is made to Attachment J, "Fire Modeling V&V, [Validation & Verification]" for a discussion of the acceptability of the fire models that were us ed.

Regarding the acceptability of the PRA approach, methods, and data:

i. Specifically regarding the use of FDS in the MCR abandonment calculations:

iii. Confirm that the 10 in

. by 6 in. holes that were created in the polycarbonate ceiling are consistent with the actual openings to the interstitial space, or explain why the assumed hole dimensions are conservative in terms of CR abandonment.

RESPONSE

The Main Control Room (MCR) analysis modeled ventilation openings in the drop ceiling to represent the natural ventilation openings in the ceiling to the interstitial space

. The FDS models assumed ventilation openings at the locations where fully enclosed cable tray risers penetrate the drop ceiling for simplified modeling of the vent distribution

. The ventilation openings were conservatively modeled as 10 inch by 6 inch holes, which is representative of the openings in the ceiling to the interstitial space

. The modeled ventilation holes are located throughout the MCRs and provide consistent venting for all fires.

The openings in the drop ceiling were not explicitly modeled in the MCR abandonment analysis, but they are bounded by conservatively modeling smaller ventilation openings

. For example, between each ceiling tile there are small openings, approximately 1 inch wide, which run the entire length and width of the MCRs

. The ventilation area from these openings result in approximately 16 ft 2 of flow area to the interstitial space in the Units 1 and 2 MCR and approximately 9 ft 2 in the Unit 3 MCR.

E1-20 The table below provides the total area of the modeled openings in the FDS scenarios versus the actual area of the openings around the ceiling tiles in the drop ceilings.

Room Total Area of Assumed Openings From FDS Models (ft²)

Total Area of Actual Openings Around Ceiling Tiles (ft²)

Units 1 and 2 MCR 7.7 16 Unit 3 MCR 2.0 9 In addition to the actual ventilation gaps around the ceiling tiles, there are numerous additional openings in the ceiling of both the Units 1 and 2 MCR and the Unit 3 MCR (e.g., the open area around the perimeter of the large cable tray risers and large conduit penetrations that are mostly open to the interstitial space).

As shown above, the actual openings in the ceiling of both MCRs are significantly greater than the area of the openings modeled in the MCR analysis

. Additionally, the modeled ventilation holes are located throughout the MCRs, providing consistent ventilation for any fire

. Therefore, the assumed 10 inch by 6 inch holes conservatively bound the actual hole dimensions, in terms of MCR abandonment. E1-21 FM RAI 01.i.vii NFPA 805, Section 2.4.3.3, states:

"The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [authority having jurisdiction] ..." The NRC staff noted that fire modeling comprised the following:

The algebraic equations implemented in FDTs [Fire Dynamics Tools] and Fire Induced Vulnerability Evaluation, Revision 1 (FIVE) were used to characterize flame radiation (heat flux), flame height, plume temperature, ceiling jet temperature, and hot gas layer (HGL) temperature.

The Consolidated Model of Fire and Smoke Transport (CFAST) was used in the multi-compartment analysis (MCA), and for the temperature sensitive equipment hot gas layer study.

Fire Dynamics Simulator (FDS) was used to assess the MCR habitability, and in the plume/hot gas layer interaction and temperature sensitive equipment ZOI studies.

Section 4.5.1.2, "Fire PRA" of the LAR states that fire modeling was performed as part of the FPRA development (NFPA 805 Section 4.2.4.2)

. Reference is made to Attachment J, "Fire Modeling V&V, [Validation & Verification]" for a discussion of the acceptability of the fire models that were used.

Regarding the acceptability of the PRA approach, methods, and data:

i. Specifically regarding the use of FDS in the MCR abandonment calculations: vii. Provide technical justification for assuming that transient fires in the MCR reach peak HRR in 8 minutes. RESPONSE:

NUREG/CR-6850, Supplement 1, Chapter 17 provides guidance related to transient fire growth rate. Chapter 17 apportions transient fires and their growth rates into three categories: commo n trash cans, common types of plant trash, and transients associated with flammable liquid fuels

. Common trash cans (i.e., plastic or metal receptacles up to 33 gallons in size intended for temporary trash collection) that contain routine types of refuse (e.g., paper, plastics, and other solid materials) may be assumed to grow from zero to peak heat release rate in eight minutes. Common types of plant trash (e.g., paper, plastics, and other solid materials) that are contained in plastic trash bags but that are not contained within a plastic or metal receptacle may be assumed to grow from zero to peak heat release rate i n two minutes. Transients associated with spilled solvents or other combustible or flammable liquid fuels should be assumed to reach peak intensity immediately upon ignition.

As required by the TVA Transient Combustible Program, the MCRs do not contain a significant amount of solvents or other combustible liquids

. The MCRs do contain trash cans with transient combustibles; however, trash bags left outside of trash cans or other receptacles are prohibited and must be removed and placed in a designated storage area

. Due to these combustible controls and walkdowns of both M CRs, the HRR growth rate for transients was determined to be that of a common trash can fire scenario, that is, eight minutes to peak HRR.

E1-22 FM RAI 01.j.ii NFPA 805, Section 2.4.3.3, states:

"The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [authority having jurisdiction] ..." The NRC staff noted that fire modeling comprised the following:

The algebraic equations implemented in FDTs [Fire Dynamics Tools] and Fire Induced Vulnerability Evaluation, Revision 1 (FIVE) were used to characterize flame radiation (heat flux), flame height, plume temperature, ceiling jet temperature, and hot gas layer (HGL) temperature.

The Consolidated Model of Fire and Smoke Transport (CFAST) was used in the multi-compartment analysis (MCA), and for the temperature sensitive equipment hot gas layer study. Fire Dynamics Simulator (FDS) was used to assess the MCR habitability, and in the plume/hot gas layer interaction and temperature sensitive equipment ZOI studies.

Section 4.5.1.2, "Fire PRA" of the LAR states that fire modeling was performed as part of the FPRA development (NFPA 805 Section 4.2.4.2)

. Reference is made to Attachment J, "Fire Modeling V&V, [Validation & Verification]" for a discussion of the acceptability of the fire models that were used.

Regarding the acceptability of the PRA approach, methods, and data:

j. Specifically regarding the MCA:

ii. In the MCA it is assumed that there is a natural ventilation opening of 10 x 10 ft or 3 x 7 ft in the exposing compartment, and that there are no natural ventilation openings in the exposed compartment

. During the audit the NRC staff observed that these assumptions may not be consistent with plant conditions

. Provide technical justification for the vent dimensions in the exposing and exposed compartments assumed in the CFAST multi

-compartment calculations.

RESPONSE

The ventilation openings from the exposing compartment to other areas of the plant were conservatively characterized by a single, standard

-sized (i.e., 3 ft by 7 ft) open door

. Once a fire is detected, fire brigade personnel will be dispatched to the room and are expected to open a door to perform suppression activities, which would provide the 3 ft by 7 ft opening assumed in the fire modeling analysis

. Prior to this action, the single door is a representation of the various natural ventilation openings within the room (e.g., door gaps, vents, and openings) because the fire compartments are connected to other areas of the plant to facilitate ventilation

. The 10 ft by 10 ft ventilation opening was not modeled in any of the CFAST calculations.

The exposed compartments were modeled in CFAST as closed compartments

. A sensitivity analysis was completed to justify this approach

. Results of the sensitivity analysis are documented below and indicate that adding a ventilation opening in the exposed compartment allowing flows in and out of the computational domain does not significantly increase the HGL temperature in the exposed compartment

. The NUREG

-1934, "Nuclear Power Plant Fire Modeling Analysis Guidelines (NPP FIRE MAG)," normalized parameters for the sensitivity models remain unchanged from the original analysis provided in Attachment O of the MCA

. All CFAST inputs remain identical to the original model with the exception of additional ventilation openings in the exposed compartment as described below.

E1-23 Two worst case scenarios were considered for the sensitivity analysis as these scenarios resulted in the highest HGL temperatures without reaching the target damage threshold temperature of 205 degrees Celsius (ºC). Therefore, the results bound the other MCA scenarios with lower peak HGL temperatures. The first scenario analyzed was Fire Compartment 16

-K exposed to a fire in Fire Compartment 16-L. A ventilation opening of 3 ft by 7 ft was selected for Fire Compartment 16-K due to the presence of a door in the north wall of Fire Compartment16

-K leading to the hallway (i.e., Fire Compartment 16

-E). The maximum temperature calculated in Fire Compartment 16-K, when exposed to a fire in Fire Compartment 16

-L, decreased from 182ºC t o 176ºC when a 3 ft by 7 ft ventilation opening was added that allowed flows in and out of the computational domain

. The results are provided in the following figure. The second scenario analyzed was Fire Compartment 7 exposed to a fire in Fire Compartment 5. A door between Fire Compartment 6 and Fire Compartment 7 exists and was analyzed in both an open and closed position to further evaluate the effects of ventilation on the exposed compartment

. The door between Fire Compartment 6 and Fire Compartment 7 was modeled as a 3 ft by 7 ft opening

. Due to the small size of Fire Compartment 7 (i.e., 210 ft2), a smaller ventilation opening (i.e., 4 ft by 2 ft) was selected to account for compartment construction leakage and miscellaneous natural ventilation openings, allowing flows in and out of the computational domain

. The sensitivity analysis modeled with the door between Fire Compartment 6 and Fire Compartment 7 closed, resulted in an increased peak HGL temperature in the exposed compartment from 186ºC to 196ºC

. The sensitivity analysis that assumed the door was open resulted in peak HGL temperature of 177ºC. The results ar e provided in the following figure. E1-24 The sensitivity analyses above resulted in an increased peak HGL temperature within 10ºC of the results reported in the MCA

. In all cases, the peak HGL temperature was below the target damage threshold of 205ºC

. Therefore, the results and conclusions of the MCA assuming no ventilation openings in the exposed compartment remain valid.

Assuming that all penetrations fail between the exposing and exposed compartment is conservative as this allows for the maximum volume of hot gases to flow from the exposing compartment into the exposed compartment

. Additional conservatism is built into the analysis by assuming the fire cannot become ventilation limited and the oxygen lower limit is set to 1% in CFAST. The sensitivity analysis justifies that the MCA CFAST models assuming a closed exposed compartment provide acceptable results and the conclusions of the MCA are valid.

E1-25 FM RAI 03.a NFPA 805, Section 2.7.3.2, "Verification and Validation," states: "Each calculational model or numerical method used shall be verified and validated through comparison to test results or comparison to other acceptable models."

LAR Section 4.5.1.2, "Fire PRA" states that fire modeling was performed as part of the FPRA development (NFPA 805, Section 4.2.4.2)

. Reference is made to LAR Attachment J, "Fire Modeling V&V," for a discussion of the verification and validation (V&V) of the fire models that were used

. Furthermore, LAR Section 4.7.3 "Compliance with Quality Requirements in Section 2.7.3 of NFPA 805" states that "Calculational models and numerical methods used i n support of compliance with 10 CFR 50.48(c) were verified and validated as required by Section 2.7.3.2 of NFPA 805."

Regarding the V& V of fire models:

a. Explain how the workbooks that were developed to calculate the ZOI and HGL temperatures were verified.

RESPONSE

The Fire Modeling Workbook was developed to calculate the ZOI and HGL temperatures

. The Fire Modeling Workbook was verified by "black box testing to ensure that the results were identical to the verified and validated models

. "Black box testing, also referred to as functional testing, is testing that ignores the internal mechanism of a system or component and focuses solely on the outputs generated in response to selected inputs and execution conditions.

The process consisted of entering identical inputs into the Fire Modeling Workbook and the NUREG-1805, "Fire Dynamic Tools and Fire

-Induced Vulnerability Evaluation," Revision 1, and comparing the results produced by each document

. Because the correlations from NUREG-1805 were verified and validated in NUREG

-1824, "Verification & Validation of Selected Fire Models for Nuclear Power Plant Applications," and the results from NUREG

-1805 match the results produced by the Fire Modeling Workbook, the Fire Modeling Workbook is verified and validated with respect to NUREG

-1824. E1-26 FM RAI 03.b.i NFPA 805, Section 2.7.3.2, "Verification and Validation," states: "Each calculational model or numerical method used shall be verified and validated through comparison to test results or comparison to other acceptable models."

Section 4.5.1.2, "Fire PRA" states that fire modeling was performed as part of the FPRA development (NFPA 805, Section 4.2.4.2)

. Reference is made to LAR Attachment J, "Fire Modeling V&V," for a discussion of the verification and validation (V&V) of the fire models that were used

. Furthermore LAR Section 4.7.3 "Compliance with Quality Requirements in Section 2.7.3 of NFPA 805" states that "Calculational models and numerical methods used in support of compliance with10 CFR 50.48(c) were verified and validated as required by Section 2.7.3.2 of NFPA 805."

Regarding the V& V of fire models:

b. Beyler's method to estimate the HGL temperature in closed compartments was used in the MCA, but there is no discussion of this model in Attachment J of the LAR and the V&V report (reference 16 in the LAR).

i. Describe how Beyler's method (as implemented) was verified.

RESPONSE

Beyler's method to estimate the HGL temperature in closed compartments was used in the MCA in accordance with NUREG

-1805 Fire Dynamics Tool (FDT) 2.3, "Predicting Hot Gas Layer Temperature in a Room Fire with Door Closed.

" This FDT is discussed in NUREG

-1805, Section 2.6 and verified and validated in NUREG

-1824, Volume 3, Section 3.1.2

. The correlation was applied using verified methods within the validation range and limitations, or the use was justified as acceptable, as further discussed in RAI responses for FM RAI 03.b.ii and FM RAI 03.b.iii. In addition, LAR Attachment J is revised to include the following line item:

E1-27 Calculation Application V&V Basis Discussion Hot Gas Layer (Method of Beyler) Calculates the hot gas layer temperature for a closed compartment with no ventilation.

NUREG-1805, Chapter 2, 2004 NUREG-1824, Volume 3, 2007 SFPE Handbook, 4 th Edition, Chapter 3

-6, Walton W. and Thomas, P., 2008 MDQ0009992012000099,

"Verification and Validation of Fire Modeling Tools and Approaches for use in NFPA 805 and Fire PRA Applications," Appendix A The correlation is used in the NUREG-1805 fire model, for which V&V was documented in NUREG-1824. The correlation is documented in an authoritative publication of the SFPE Handbook of Fire Protection Engineering.

The correlation has been applied within the validated range reported in NUREG-1824 or, if applied outside the validated range, the model has been justified as acceptable, either by qualitative analysis, or by quantitative sensitivity analysis.

E1-28 FM RAI 03.b.ii NFPA 805, Section 2.7.3.2, "Verification and Validation," states: "Each calculational model or numerical method used shall be verified and validated through comparison to test results or comparison to other acceptable models."

Section 4.5.1.2, "Fire PRA" states that fire modeling was performed as part of the FPRA development (NFPA 805, Section 4.2.4.2)

. Reference is made to LAR Attachment J, "Fire Modeling V&V," for a discussion of the verification and validation (V&V) of the fire models that were used

. Furthermore LAR Section 4.7.3 "Compliance with Quality Requirements in Section 2.7.3 of NFPA 805" states that "Calculational models and numerical methods used in support of compliance with10 CFR 50.48(c) were verified and validated as required by Section 2.7.3.2 of NFPA 805."

Regarding the V& V of fire models:

b. Beyler's method to estimate the HGL temperature in closed compartments was used in the MCA, but there is no discussion of this model in Attachment J of the LAR and the V&V report (reference 16 in the LAR).

ii. Provide the V&V basis for Beyler's method.

RESPONSE

Verification and Validation (

V&V) Basis - Hot Gas Layer - Closed Compartment (Method of Beyler)

HGL temperatures

, for those cases where a closed compartment with no ventilation is assumed, were calculated using the method of Beyler found in NUREG-1824, Section 3.1.2. This method is also found on page 3

-213 (Chapter 3

-6) of the 4 th edition of the Society of Fire Protection Engineering (SFPE) Handbook. For a constant HRR, the compartment HGL temperature increase above ambient was determined using the following equation:

)1(211212tKgetKKKT where: pTmcAckK)4.0(21 pmcQK2 Input Parameters:

gT= upper-layer gas temperature rise above ambient (T g - Ta) [K] t= time after ignition [s]

k=t =density of the compartment surface (kg/m

3) c=s TA= total area of the compartment enclosing surfaces, excluding area of vent opening(s) [m 2] m = mass of the gas in the compartment (kg)

E1-29 pc= specific heat of air [kJ/kg-K] Q= heat release rate [kW]

The method of Beyler for HGL temperature was designated as YELLOW+ in NUREG

-1824. A YELLOW+ designation is assigned "If the first criterion is satisfied and the calculated relative differences are outside the experimental uncertainty but indicate a consistent pattern of model over-prediction or under

-prediction, then the model predictive capability is characterized as YELLOW+ for over

-prediction, and YELLOW- for under

-prediction

. The model prediction for the specific attribute may be useful within the ranges of experiments in this study, and as described in Tables 2

-4 and 2-5, but the users should use caution when interpreting the results of the model

. A complete understanding of model assumptions and scenario applicability to these V&V results is necessary

. The model may be used if the grade is YELLOW+ when the user ensures that model over

-prediction reflects conservatism

. The user must exercise caution when using models with capabilities described as YELLOW+/-."

NUREG-1824, Volume 3, Section 6.1 further states that "The FDT models for HGL temperature capture the appropriate physics and are based on appropriate empirical data

. FDTs generally over-predicts HGL temperature, outside of uncertainty.

" The over-prediction is expected to lead to conservative results and increased safety margin

.

The following assumptions and limitations were applied to the Beyler HGL temperature calculations, similar to all other HGL temperature calculations:

The method was applied to both transient and steady

-state fire growth with a known HRR.

Compartment geometry assumed that a given space could be analyzed as a rectangular space with no beam pockets

. This assumption affected the smoke filling rate within a space if the space had beam pockets

. For irregularly shaped compartments, equivalent compartment dimensions (i.e., length, width, and height) were calculated and yielded slightly higher HGL temperatures than would actually be expected from a fire in the given compartment

. The method predicted average temperatures and did not apply to cases in which prediction of local temperature was desired

. For example, the method was not used to predict detector or sprinkler actuation or the material temperatures resulting from direct flame impingement

. Caution was exercised when the compartment overhead was highly congested wit h obstructions such as cable trays, conduits, and ducts, to ensure appropriate use of the correlation

. A single heat transfer coefficient was used for the entire inner surface of the compartment.

The heat flow to and through the compartment boundaries was unidimensional (i.e., corners and edges are ignored, and thermally thick boundaries are assumed to be semi

-infinite slabs).

E1-30 FM RAI 03

.b.iii NFPA 805, Section 2.7.3.2, "Verification and Validation," states: "Each calculational model or numerical method used shall be verified and validated through comparison to test results or comparison to other acceptable models."

LAR Section 4.5.1.2, "Fire PRA" states that fire modeling was performed as part of the FPRA development (NFPA 805, Section 4.2.4.2)

. Reference is made to LAR Attachment J, "Fire Modeling V&V," for a discussion of the verification and validation (V&V) of the fire models that were used

. Furthermore, LAR Section 4.7.3 "Compliance with Quality Requirements in Section 2.7.3 of NFPA 805" states that "Calculational models and numerical methods used in support of compliance with 10 CFR 50.48(c) were verified and validated as required by Section 2.7.3.2 of NFPA 805."

Regarding the V& V of fire models:

b. Beyler's method to estimate the HGL temperature in closed compartments was used in the MCA, but there is no discussion of this model in Attachment J of the LAR and the V&V report (reference 16 in the LAR).

iii. Provide technical details to demonstrate whether the method has been applied within the validation range of input parameters, or to justify the application of the equations outside the validation range reported in the V&V basis documents.

RESPONSE

Beyler's Method to calculate the HGL temperature in closed compartments was applied, in most cases, within the validated ranges reported in NUREG

-1934, "Nuclear Power Plant Fir e Modeling Analysis Guidelines," Final Report, November 2012, Table 2-5. When Beyler's Method was applied outside the validated ranges reported in NUREG

-1934, Table 2-5, a qualitative analysis was performed to justify the acceptability of using this method.

Technical details demonstrating that the models are within range, as well as any justification of models outside the range, are documented in the table below for the HGL (i.e., Beyler's Method) calculations in the MCA.

E1-31 Normalized Parameters

- Multi-Compartment Analysis NUREG 1805 FDT 02.3 Beyler's Method Quantity Normalized Parameter Validation Range Validity Statement Fire Froude Number N/A 0.4 - 2.4 The Froude Number is predominantly used to validate the plume temperatures and flame heights. Because Beyler's Method was used exclusively to calculate HGL temperatures, the item of importance is the amount of energy (HRR) being released into the fire zone, and a Froude Number outside of the validated range would not invalidate the results. Flame Length relative to Ceiling Height N/A 0.2 - 1.0 The primary application of this parameter is to determine whether the flame length exceeds the ceiling height

. The concern is that for this type of configuration when the normalized parameter would be calculated as greater than one, aside from being outside of the validated range, the models for predicting this phenomenon have not been verified or validated

. The Beyler Method analyzes HGL development exclusively and does not calculate damage to targets within the flame height or targets that may be subjected to flame radiation. Therefore, this parameter is not applicable to the MCA.

Ceiling Jet Radial Distance relative to Ceiling height N/A 1.2 - 1.7 The primary application of this parameter is to determine the temperature of targets at the ceiling. This correlation is primarily used, for example, in conjunction with the smoke detection activation correlation to determine the time to detection and suppression activation

. NUREG-1805 FDT 02.3 is not used to determine the time to detection and sprinkler activation

. Further, the evaluation of other ceiling jet targets is not included in the MCA. Equivalency Ratio N/A 0.04 - 0.6 The underlying consideration for this parameter is that conditions in the enclosure are not expected to be worse in a fire where the combustion process is affected by lack of oxygen than they would be under fire conditions where the combustion process is assumed unaffected

. This parameter is not applicable to MCA fire modeling calculations using Beyler's Method because oxygen levels are not taken into account with the equations employed by this model.

E1-32 Therefore, based on the above justifications, the use of Beyler's Method to estimate the HGL in the MCA is justified.

Compartment Aspect Ratio - Length See Discussion in Validity Statement 0.6 - 5.7 Most of the room geometries for the fire compartments modeled in the MCA using the Beyler Method are within the validated range of NUREG-1934 with respect to Compartment Aspect Ratio

. For those exceptions which were outside of the validation range for compartment aspect ratio, the use of this model is justified because:

- The primary purpose of the aspect ratio is to determine whether there could be localized HGL effects surrounding the fire that could not be evenly distributed over the volume of the compartment

. In an MCA, localized effects are not a concern because a compartment screens from damage if an HGL causing whole room damage is not obtained. Further, the localized HGL concerns which affect fire compartments outside the validation range mostly apply to conditions early in the fire scenario development

. Because fire models using the Beyler Method in the screening process for the MCA were analyzed for at least 20 minutes, the effects of a potential localized HGL on the model are minimized

. - The volume of space available between the top of the door soffits and the ceiling provides a substantial volume for any potentially localized HGL gases to collect and mix before affecting an adjacent fire compartment.

- Safety margin exists between the highest HRR for an initiator present in the fire compartments when compared to the limiting HRRs needed to create an HGL

. This safety margin is increased when considering the fact that the 98 th percentile HRRs were used as the initiating fire sizes

. Compartment Aspect Ratio - Width See Discussion in Validity Statement 0.6 - 5.7 Refer to the validity statement for Compartment Aspect Ratio - Length. Radial Distance relative to Fire Diameter N/A 2.2 - 5.7 This parameter is only applicable when the heat flux parameter is computed

. Beyler's Method does not analyze radiant targets

. Hot gas layer development is the only fire effect analyzed.

E1-33 FM RAI 03.c NFPA 805, Section 2.7.3.2, "Verification and Validation," states: "Each calculational model or numerical method used shall be verified and validated through comparison to test results or comparison to other acceptable models."

Section 4.5.1.2, "Fire PRA" of the Transition Report states that fire modeling was performed as part of the Fire PRA development (NFPA 805, Section 4.2.4.2). Reference is made to Attachment J, "Fire Modeling V&V," for a discussion of the verification and validation (V&V) of the fire models that were used

. Furthermore Section 4.7.3 "Compliance with Quality Requirements in Section 2.7.3 of NFPA 805" of the Transition Report states that "Calculational models and numerical methods used in support of compliance with10 CFR 50.48(c) were verified and validated as required by Section 2.7.3.2 of NF PA 805." Regarding the V& V of fire models:

c. Attachment J in the LAR states that the Smoke Detection Actuation Correlation (Method of Heskestad and Delichatsios) has been applied within the validated range reported in NUREG-1824. However the latter does not report a validation range for this correlation

. Provide technical details to demonstrate that the correlation has been applied within the validation range of input parameters, or to justify the application of the equations outside the validation range reported in the V&V basis documents.

RESPONSE

The Heskestad and Delichatsios Smoke Detection Actuation Correlation is based upon the ceiling jet temperature predicted by Alpert's Ceiling Jet Correlation

. Therefore, the normalized parameters for the ceiling jet correlation are applicable

. The normalized parameter that applies to the Alpert's Ceiling Jet Correlation is the ceiling jet radial distance relative to the ceiling height and the validation range is 1.2 to 1.7. The Alpert's Ceiling Jet Correlation, used for the Heskestad and Delichatsios Smoke Detection Actuation Correlation, was performed within the validated range reported in NUREG

-1824; therefore, the Heskestad and Delichatsios correlation was appropriately used

.

In addition to being applied within the validation range for Alpert's Ceiling Jet, the smoke detection correlation was applied to fuels, configurations, and environmental conditions consistent with those described in the SFPE Handbook and NUREG

-1805. It was also applied within the limitations described in these publications.

Heskestad and Delichatsios correlated a smoke temperature change of 10°C (i.e., 18 degrees Fahrenheit (

°F)) based upon typical fire fuels

. The materials tested to create the Heskestad and Delichatsios smoke detector correlation are representative of the fuels modeled for smoke detector activation

. These tested materials include various plastics, foams, and paper that have smoke properties similar to the fires modeled at BFN. E1-34 FM RAI 05.a LAR Section 4.5.1.2, "Fire PRA" states that fire modeling was performed as part of the FPRA development (NFPA 805, Section 4.2.4.2)

. This requires that qualified fire modeling and PRA personnel work together

. Furthermore, LAR Section 4.7.3, "Compliance with Quality Requirements in Section 2.7.3 of NFPA 805," states that "For personnel performing fire modeling or FPRA development and evaluation, TVA will develop and maintain qualification requirements for individuals assigned various tasks

. Position Specific Guides will be developed to identify and document required training and mentoring to ensure individuals are appropriately qualified per the requirements of NFPA 805 Section 2.7.3.4 to perform assigned work."

Regarding qualifications of users of engineering analyses and numerical models (i.e., fire modeling techniques):

a. Describe the requirements to qualify personnel for performing fire modeling calculations in the NFPA 805 transition.

RESPONSE

Fire modeling calculations have been, and will be, performed by engineers who meet the qualification requirements of Section 2.7.3.4 of NFPA 805 (2001)

. Fire modeling to support the LAR and Fire PRA development was performed by contract personnel using their compan ies' procedures and their Quality Assurance programs

. These procedures require that project personnel assigned to each task have the proper experience and training to perform the work

. This is verified by contractor company management.

These contractor personnel were chosen based on their experience and expertise in fire modeling. The qualifications needed to perform fire modeling related tasks depends in part on the specific role of the personnel

. Appropriate qualifications for contractor personnel using, applying, and approving fire modeling tools include required reading on fire modeling Project Instructions, relevant industry methodology and/or guidance documents such as NUREG/CR-6850, NUREG

-1934, NUREG

-1805, and applicable fire modeling software user's guide documents

. Other requirements include training and/or mentoring in Fire Growth Analysis, ZOI calculations, and Fire Modeling Tools

. Qualification requirements also involve a demonstration of comprehension and proficiency in fire modeling.

The qualification requirements to perform other fire modeling related tasks depend in part on the personnel's specific assigned role

. Some sub-tasks of fire modeling, may be assigned to other staff with experience and skill set commensurate with the task

. For example walk down data collection, raceway drawing reviews, and System Assurance and Fire Protection Engineering (SAFE) data entry.

During the NFPA 805 transition phase (i.e., LAR submittal to receipt of the NRC Safety Evaluation), TVA will continue to utilize qualified personnel

. In addition, as stated in LAR Section 4.7.3, TVA will develop qualification requirements for TVA personnel to perform fire modeling during and after the transition

. Position Specific Qualification Guides and Task Qualifications will be developed by first performing a job analysis using TVA's Systematic Approach to Training (SAT) process via training program procedures

. These Task Qualifications will be similar to those described above and will be accomplished by means of the SAT mentoring process.

E1-35 FM RAI 05.b LAR Section 4.5.1.2, "Fire PRA" states that fire modeling was performed as part of the FPRA development (NFPA 805, Section 4.2.4.2)

. This requires that qualified fire modeling and PRA personnel work together

. Furthermore, LAR Section 4.7.3, "Compliance with Quality Requirements in Section 2.7.3 of NFPA 805," states that "For personnel performing fire modeling or FPRA development and evaluation, TVA will develop and maintain qualification requirements for individuals assigned various tasks

. Position Specific Guides will be developed to identify and document required training and mentoring to ensure individuals are appropriately qualified per the requirements of NFPA 805 Section 2.7.3.4 to perform assigned work."

Regarding qualifications of users of engineering analyses and numerical models (i.e., fire modeling techniques):

b. Describe the process for ensuring that the fire modeling personnel meet the qualifications, not only before the transition but also during and following the transition.

RESPONSE

Fire modeling to support the LAR and Fire PRA development (i.e., before the transition) was performed by contractor personnel using their compan ies' procedures and their Quality Assurance programs

. Their procedures require that their management be responsible for the overall project performance of fire modeling tasks and for ensuring that project personnel assigned to each task have the proper experience and training to perform the work

. Engineers performing fire modeling are required to perform their duties in accordance with the fire modeling procedure and Quality Assurance program

. This process was followed to ensure the personnel performing fire modeling were qualified.

During the transition phase (i.e., LAR submittal to receipt of the NRC Safety Evaluation), TVA will continue to utilize qualified personnel to perform fire modeling and will continue to use the process described above

. Additionally, TVA personnel will develop qualification requirements for TVA engineers to perform fire modeling during and after the transition

. This requirement is also included as Implementation Item 23 in LAR Attachment S, Table S-3.

Position Specific Qualification Guides and Task Qualifications will be developed and implemented via site training program procedures

.

When TVA personnel perform fire modeling post transition, current TVA processes will be followed to ensure that assigned personnel are qualified

. TVA procedure NPG

-SPP-17.1, "Systematic Approach to Training (SAT) Overview," requires that personnel be qualified to perform assigned tasks and managers and supervisors are responsible to ensure that personnel are qualified

. Once the fire modeling Qualification Guides and Task Qualifications are developed, the process described in NPG

-SPP-17.1 will ensure personnel performing fire modeling are qualified. E1-36 FM RAI 05.c LAR Section 4.5.1.2, "Fire PRA" states that fire modeling was performed as part of the FPRA development (NFPA 805, Section 4.2.4.2)

. This requires that qualified fire modeling and PRA personnel work together

. Furthermore, LAR Section 4.7.3, "Compliance with Quality Requirements in Section 2.7.3 of NFPA 805," states that "For personnel performing fire modeling or FPRA development and evaluation, TVA will develop and maintain qualification requirements for individuals assigned various tasks

. Position Specific Guides will be developed to identify and document required training and mentoring to ensure individuals are appropriately qualified per the requirements of NFPA 805 Section 2.7.3.4 to perform assigned work

.'

Regarding qualifications of users of engineering analyses and numerical models (i.e., fire modeling techniques):

c. When fire modeling is performed in support of Fire PRA, describe how proper communication between the fire modeling and FPRA personnel is ensured. RESPONSE:

When fire modeling was performed to support the Fire PRA, the fire modeling analyses were developed into approved calculations which were then used as input to the Fire PRA

. These calculations are controlled under the design and configuration management processes and procedures within TVA

. These procedures ensure proper communication between fire modeling and PRA engineers via the identification of affected documents.

During the development of the Fire PRA, the personnel performing fire modeling and the PRA engineers maintained frequent communications

. PRA and fire modeling personnel were active participants on the project team

. The team participated in various re

-iterative analyses to support final issuance of the approved fire modeling and Fire PRA documentation

. For example, engineers conducting fire modeling and PRA engineers participated in periodic cutset review meetings during the Fire PRA development

. Participation in cutset reviews facilitated the identification of fire modeling refinements necessary for supporting risk quantification activities.

E1-37 FM RAI 06.a LAR Section 4.7.3, "Compliance with Quality Requirements in Section 2.7.3 of NFPA 805" states that "Uncertainty analyses were performed as required by 2.7.3.5 of NFPA 805 and the results were considered in the context of the application

. This is of particular interest in fire modeling and Fire PRA development used to support performance

-based approach." Regarding the uncertainty analysis for fire modeling:

a. Describe how the uncertainty associated with the fire model input parameters (compartment geometry, radiative fraction of heat transfer methods, thermophysical properties, etc.) was addressed for this application and accounted for in the analyses.

RESPONSE

Fire modeling was performed within the Fire PRA utilizing codes and standards developed by industry and NRC staff and that were verified and validated in authoritative publications such as NUREG-1824, "Verification and Validation of Selected Fire Models for Nuclear Power Plant Applications.

" In general, the fire modeling was performed using conservative methods and input parameters that were based upon NUREG/CR

-6850, "Fire PRA Methodology for Nuclear Power Facilities.

" This approach was used based upon the current state of knowledge regarding the uncertainties related to the application of the fire modeling tools and associated input parameters for specific plant configurations

. A discussion of uncertainties associated with detailed fire modeling is summarized below.

The detailed fire modeling task developed a probabilistic output in the form of target failure probabilities that were subject to both aleatory (i.e., statistical) and epistemic (i.e., systematic) uncertainties.

NUREG/CR-6850, Appendix V, recommends that to the extent possible, modeling parameters should be expressed as probability distributions and propagated through the analysis to arrive at target failure probability distributions

. These distributions should be based on the variation o f experimental results as well as the analyst's judgment

. To the extent possible, more than one fire model can be applied and probabilities assigned to the outcome which describe the degree of belief that each model is the correct one.

The propagation of fire for each non

-screened fire source has been described by a fire model (i.e., represented by a fire growth event tree) that addresses the specific characteristics of the source and the configuration of secondary combustibles.

Aleatory uncertainties identified within the fire modeling parameters include:

Detector response reliability and availability Automatic suppression system reliability and availability Manual suppression reliability with respect to time available

Epistemic uncertainties which affect the ZOI and time to damage range include:

Heat release rates (peak HRR, time to reach peak, steady burning time, decay time)

Number of cabinet cable bundles Ignition source fire diameter Room ventilation conditions Fire growth assumptions (cable tray empirical rule set, barrier delay)

Cable fire spread characteristics for horizontal and vertical trays E1-38 Transient fires (peak HRR, time to reach peak, location factor, detection time)

Oil fires (spill assumptions)

Assumed target location Target damage threshold criteria Manual detection time Mean prompt suppression rate Manual suppression rate Welding and cutting target damage set Transient target effects Due to the uncertainty with each of these parameters, conservative values were selected during the fire modeling task for each to provide safety margin

. In accordance with NEI 04-02, there is no clear definition of an adequate safety margin

. However, the safety margin should be sufficient to bound the uncertainty within a particular calculation or application. The Detailed Fire Modeling Report provides a list of items that were modeled conservatively and that provide safety margin

. Some examples include the following items:

The majority of fire scenarios involving electrical equipment, including the electrical split fraction of pump fires, utilize the 98th percentile HRR for the severity factor calculated out to the nearest Fire PRA target

. This was considered conservative.

The fire elevation in most cases was at the top of the cabinet or pump body, which is conservative because the combustion process would occur where the fuel mixes with oxygen, which is not always at the top of the ignition source.

The radiative fraction utilized was 0.4, which represents a 33% increase over the normally recommended value of 0.3.

The convective heat release rate fraction utilized was 0.7

. The normally recommended value is between 0.6 and 0.65, and thus the use of 0

.7 is conservative.

For transient fire effects, a large bounding transient zone assumes all targets within its ZOI are affected by a fire

. Time to damage is calculated based on the most severe (i.e., closest) target, which is conservative because a transient fire would actually have a much smaller zone of influence and varying damage times

. This approach was implemented to minimize the multitude of transient scenarios to be analyzed.

For HGL calculations, no equipment or structural steel was credited as a heat sink, because the closed

-form correlations used do not account for heat loss to these items.

Not every cable tray is filled to capacity

. The fire modeling assumed all cable trays were filled to capacity, which provided a conservative estimate of the contribution of cable insulation to the fire and the corresponding time to damage.

As the fire propagated to secondary combustibles, the fire was conservatively modeled as one single fire using the fire modeling closed

-form correlations

. The resulting plume temperature estimates used in this analysis were therefore conservative, because in actuality, the fire would be distributed over a large surface area, and would be less severe at the target location.

For most scenarios, target damage was assumed to occur when the exposure environment met or exceeded the damage threshold

. No additional time delay due to thermal response was allowed.

The fire elevation for transient fires was assumed to be two feet. This was conservative because most transient fires are expected to be below this height (e.g., at floor level).

Oil fires were analyzed as both unconfined and confined spills with 20

-minute durations

. While unconfined spills resulted in large heat release rates, they usually burn for seconds, not minutes

. However, the oil fires were conservatively analyzed for a 20-minute burn time to account for the uncertainty in the oil spill size.

E1-39 High energy arcing fault scenarios were conservatively assumed to be at peak fire intensity for 20

-minutes from time zero (ignition), even though the initial arcing fault is expected to consume the contents of the cabinet and burn for only a few minutes.

For many fire scenarios, fire brigade intervention was not credited prior to 85 minutes

. A review of the Fire Brigade drills indicated that typical manual suppression times are about 20 minutes). E1-40 FM RAI 06.b LAR Section 4.7.3, "Compliance with Quality Requirements in Section 2.7.3 of NFPA 805" states that "Uncertainty analyses were performed as required by 2.7.3.5 of NFPA 805 and the results were considered in the context of the application

. This is of particular interest in fire modeling and Fire PRA development used to support performance

-based approach."

Regarding the uncertainty analysis for fire modeling:

b. Describe how the "model" and "completeness" uncertainties were addressed for this application and accounted for in the analyses

. NUREG-1934, "Nuclear Power Plant Fire Modeling Analysis Guidelines," provides guidance on quantifying model/completeness uncertainty.

An example of these uncertainties is the effect of ignoring the contents of a compartment on the HGL calculations

. Cabinets and cable trays reduce the effective volume of a compartment, but also act as a heat sink

. Hence, how ignoring these contents affects the HGL temperature is not clear

. Explain how the corresponding uncertainties were accounted for, or show that ignoring compartment contents leads to conservative HGL temperatures.

RESPONSE

NUREG-1934 states that "model" uncertainties can be estimated using the processes of verification and validation

. Model uncertainty is based primarily on comparisons of model predictions with experimental measurements as documented in NUREG

-1824 and other model validation studies.

The fire models used a s listed in LAR Attachment J were within or very near the experimental uncertainty, as determined by NUREG

-1824, "Verification and Validation of Selected Fire Models for Nuclear Power Plant Applications," Final Report, dated May 2007. Where applicable, the fire models listed below were applied within the validation ranges or the use was justified as acceptable with a subsequent analysis

. Each model is discussed below. Hot Gas Layer Temperature using FDTs

The predictive capability of this parameter using FDTs is characterized as YELLOW+ according to NUREG-1824, Table 3-1.

As stated in NUREG

-1824, Volume 3, Section 6.1, a YELLOW+/- characterization is assigned "If the first criterion is satisfied and the calculated relative differences are outside the experimental uncertainty but indicate a consistent pattern of model over

-prediction or under

-prediction, then the model predictive capability is characterized as YELLOW+ for over

-prediction, and YELLOW- for under

-prediction

. The model prediction for the specific attribute may be useful within the ranges of experiments in this study, and as described in Tables 2

-4 and 2-5, but the users should use caution when interpreting the results of the model

. A complete understanding of model assumptions and scenario applicability to these V&V results is necessary

. The model may be used if the grade is YELLOW+ when the user ensures that model over

-prediction reflects conservatism

. The user must exercise caution when using models with capabilities described as YELLOW+/-."

NUREG-1824, Volume 3, Section 6.1 states that:

"The FDTs models for HGL temperature capture the appropriate physics and are based on appropriate empirical data

. FDTs generally over-predict HGL temperature, outside of uncertainty." The over-prediction is expected to lead to conservative results and increased safety margin.

E1-41 Hot Gas Layer Height and Temperature using Fire Dynamics Simulator (FDS)

The predictive capability of these parameters in FDS is characterized as GREEN according to NUREG-1824, Table 3-1.

As stated in NUREG

-1824, Volume 7, Section 6.1, a GREEN characterization is assigned "If both criteria are satisfied (i.e., the model physics are appropriate for the calculation being made and the calculated relative differences are within or very near experimental uncertainty), then the V&V team concluded that the fire model prediction is accurate for the ranges of experiments in this study, and as described in Tables 2

-4 and 2-5. A grade of GREEN indicates the model can be used with confidence to calculate the specific attribute

. The user should recognize, however, that the accuracy of the model prediction is still somewhat uncertain and for some attributes, such as smoke concentration and room pressure, these uncertainties may be rather large. It is important to note that a grade of GREEN indicates validation only in the parameter space defined by the test series used in this study; that is, when the model is used within the ranges of the parameters defined by the experiments, it is validated." NUREG-1824, Volume 7, Section 6.1 further states: "FDS is suitable for predicting HGL temperature and height, with no specific caveats, in both the room of origin and adjacent rooms

. In terms of the ranking system adopted in this report, FDS merits a Green for this category, based on FDS predictions of the HGL temperature and height are, with a few exceptions, within experimental uncertainty." Hot Gas Layer Temperature and Height using the Consolidated Model of Fire and Smoke Transport (CFAST)

The predictive capability of these parameters in CFAST was characterized as GREEN according to NUREG

-1824, Table 3-1. The GREEN designation is discussed above under the "Hot Gas Layer Height and Temperature using FDS" heading. Specifically, the GREEN designation was assigned to the CFAST HGL temperature parameter calculated in the fire compartment of origin

. Compartments remote from the fire were assigned a YELLOW designation

. HGL temperatures were calculated in CFAST for the compartment of fire origin only.

NUREG-1824, Volume 5, Section 6.1 summary states:

"The CFAST predictions of the HGL temperature and height are, with a few exceptions, within or close to experimental uncertainty

. The CFAST predictions are typical of those found in other studies where the HGL temperature is typically somewhat over

-predicted and HGL height somewhat lower than experimental measurements

. These differences are likely attributable to simplifications in the model dealing with mixing between the layers, entrainment in the fire plume, and flow through vents

. Still, predictions are mostly within 10% to 20% of experimental measurements." Ceiling Jet Temperature using Alpert Correlation The predictive capability of this parameter using the Alpert correlation in the Fire Induced Vulnerability Evaluation (FIVE) fire model is characterized as YELLOW+ according to NUREG-1824, Table 3-1. The YELLOW+ designation is discussed above under the "Hot Gas Layer Temperature using FDTs" heading. Specifically, NUREG-1824, Volume 4, Section 6.2 summary states:

"The Alpert correlation under

-predicts ceiling jet temperatures in compartment fires with an established hot gas layer

. This result is expected because the correlation was developed without considering HGL effects

. The original version of FIVE accounted for HGL effects by adding the ceiling jet and HGL temperature

. This practice results in consistent over

-predictions of the ceiling jet temperature

. The approach of adding ceiling jet temperatures to the calculated E1-42 hot gas layer continues to be the recommended method for FIVE-Rev 1 users

. Based on the above discussion, a classification of Yellow+ is recommended if HGL effects on the ceiling jet temperature are considered using the approach described in the above bullet

. The Alpert correlation by itself is not intended to be used in rooms with an established hot gas layer." The approach of adding the HGL temperature to the ceiling jet temperature was not used in the BFN fire modeling analysis. The primary application of the ceiling jet correlation at BFN was the determination of detection and suppression timing, in which the ceiling jet velocity is a sub

-model in the analysis

. Including the effects of a HGL would have resulted in shorter detection and suppression times, and therefore the BFN approach is conservative

. The use of the ceiling jet correlation for target damage was bounded by the use of the point source radiation model and is justified and discussed in the BFN fire modeling V&V documentation. Plume Temperature using FDTs The predictive capability of this parameter using FDTs is characterized as YELLOW- according to NUREG-1824, Table 3-1. The YELLOW- designation is discussed above under the "Hot Gas Layer Temperature using FDTs" heading. NUREG-1824, Volume 3, Section 6.2 summary states "The FDTs model for plume temperature is based on appropriate empirical data and is physically appropriate

. FDTs generally under

-predicts plume temperature, outside of uncertainty, because of the effects of the hot gas layer on test measurements of plume temperature

. The FDTs model is not appropriate for predicting the plume temperatures at elevations within a hot gas layer."

The FDTs plume correlation for fire modeling applications was used within the limitations provided in NUREG-1824. The effects of the plume and hot gas layer interaction were analyzed and documented in the BFN fire modeling V&V documentation

. The use of the FDTs plume correlation was used in accordance with the results of this analysis.

Plume Temperature using FDS The predictive capability of this parameter using FDS is characterized as YELLOW according to NUREG-1824, Table 3

-1. As stated in NUREG

-1824, Volume 7, Section 6.3, a YELLOW characterization is assigned "If the first criterion is satisfied and the calculated relative differences are outside experimental uncertainty with no consistent pattern of over- or under-prediction, then the model predictive capability is characterized as YELLOW

. A YELLOW classification is also used despite a consistent pattern of under or over

-prediction if the experimental data set is limited

. Caution should be exercised when using a fire model for predicting these attributes

. In this case, the user is referred to the details related to the experimental conditions and validation results documented in Volumes 2 through 6

. The user is advised to review and understand the model assumptions and inputs, as well as the conditions and results to determine and justify the appropriateness of the model prediction to the fire scenario for which it is being used."

NUREG-1824, Volume 7, Section 6.3 summary states:

"The FDS hydrodynamic solver is well-suited for this application

. FDS over-predicts the lower plume temperature in BE #2 because it over

-predicts the flame height

. FDS predicts the FM

/SNL plume temperature to within experimental uncertainty

. The simulations of BE #2 and the FM

/SNL series are the most time-consuming of all six test series, mainly because of the need for a fairly fine numerical grid near the plume

. It is important that a user understand that considerable computation time may be necessary to well

-resolve temperatures within the fire plume

. Even with a relatively fine grid, it is still challenging to accurately predict plume temperatures, especially in the fire itself or just above the flame tip

. There are only nine plume temperature measurements in the data set

. A E1-43 more definitive conclusion about the accuracy of FDS in predicting plume temperature would require more experimental data." In accordance with the guidance provided in NUREG-1934, a D*/ dx ratio of 5 to 10 produces favorable FDS results at moderate computational cost

. This guidance was used in the BFN FDS applications that analyzed plume temperatures

. The meshes used in BFN fire modeling were considered to be sufficiently fine to analyze plume temperatures in each case

. In addition, the plume temperatures within the flaming region are not the focal point of either study.

Flame Height using FDTs The predictive capability of this parameter using FDTs is characterized as GREEN according to NUREG-1824, Table 3-1. The GREEN designation is discussed above under the "Hot Gas Layer Height and Temperature using FDS" heading. NUREG-1824, Volume 3, Section 6.3 summary states "The FDTs model predicted flame heights consistent with visual test observations." Smoke Concentration using FDS

The predictive capability of this parameter in FDS is characterized as YELLOW according to Table 3-1 of NUREG

-1824. The YELLOW designation is discussed above under the "Plume Temperature using FDS" heading.

NUREG-1824, Volume 7, Section 6.6 summary states:

"FDS is capable of transporting smoke throughout a compartment, assuming that the production rate is known and that its transport properties are comparable to gaseous exhaust products

. This assumption may break down in closed-door fires, or if an appreciable part of the flame extends into the upper layer

. FDS over-predicts the smoke concentration in all of the BE #3 tests

. For the open

-door tests, it is possible to explain the discrepancy in terms of the uncertainty of both the specified smoke yield and the optical measurement of the smoke concentration

. There is no clear explanation for the discrepancy in the closed

-door tests

. FDS does not over

-predict the CO concentration, another fixed yield product of incomplete combustion, in either the open- or closed

-door tests

. No firm conclusions can be drawn from this one data set

. The measurements in the closed

-door experiments are inconsistent with basic conservation of mass arguments, or there is a fundamental change in the combustion process as the fire becomes oxygen

-starved. FDS does not have a submodel to adjust the production rate or the optical properties of smoke, regardless of whether or not this would explain the discrepancy between the measurements and the model predictions." The smoke concentration was used to determine the probability of M CR abandonment at BFN following a fire scenario in the M CR. Because the smoke concentration was over

-predicted for both the open

-door and closed

-door test configurations as indicated in NUREG

-1824, the BFN FDS results were conservative.

The smoke production rates used in the FDS model were conservatively selected from Table 3-4.16 of the SFPE Handbook of Fire Protection Engineering, 4th Edition. Because transport properties of the smoke are expected to be comparable to gaseous exhaust products

, the use of the model is within the limitations and the experimental uncertainty.

E1-44 Radiant Heat using FDTs The predictive capability of this parameter in FDTs is characterized as YELLOW according to NUREG-1824, Table 3-1. The YELLOW designation is discussed above under the "Plume Temperature using FDS" heading. NUREG-1824, Volume 3, Section 6.4 summary states:

"The FDTs point source radiation and solid flame radiation model in general are based on appropriate empirical data and is physically appropriate with consideration of the simplifying assumptions

. The FDTs point source radiation and solid flame radiation model are not valid for elevations within a hot gas lay er. FDTs predictions had no clear trend

. The model under- and over-predicted, outside uncertainty

. The point source radiation model is intended for predicting radiation from flames in an unobstructed and smoke

-clear path between flames and targets."

Only the FDTs point source radiation model was used in the BFN fire modeling

. NUREG-1824 states that there is no clear trend in under- or over-prediction for the point source model

. The model over

-predicted heat flux for locations immersed in a HGL, which is likely due to smoke and the HGL preventing radiation from reaching the gauges

. This over

-prediction lead s to conservative results and increased safety margin

. In a smaller number of cases, the model under-predicted heat flux due to contribution of radiation from the HGL

. In order to account for this potential under

-prediction, conservatism was built into the use of the radiation model at BFN, including the use of a radiant heat release rate fraction of 0.4, versus the normally recommended value of 0.3.

In addition, NUREG

-1824 states that the point source model is not intended to be used for locations relatively close to the fire

. In the BFN fire modeling analysis, targets located close to the fire were conservatively failed within the early stages of fire growth.

Radiant Heat using FDS The predictive capability of this parameter in FDS is characterized as YELLOW according to NUREG-1824, Table 3-1. The YELLOW designation is discussed above under the "Plume Temperature using FDS" heading.

Even though the FDS Radiant Heat Model was designated as YELLOW, NUREG 1824, Volume 7, Section 6.8 states that:

"FDS has the appropriate radiation and solid phase models for predicting the radiative and convective heat flux to targets, assuming the targets are relatively simple in shape

. FDS is capable of predicting the surface temperature of a target, assuming that its shape is relatively simple and its composition fairly uniform

. FDS predictions of heat flux and surface temperature are generally within experimental uncertainty, but there are numerous exceptions attributable to a variety of reasons

. The accuracy of the predictions generally decreases as the targets move closer to, or go inside of the fire

. There is not enough near-field data to challenge the model in this regard." FDS was used to calculate radiant heat exposure to determine the radiant heat exposure to an electrical cabinet from a transient fire

. The limitations outlined in NUREG

-1824 were not of concern based upon the following:

Heat flux was not calculated for any targets inside of the fire

. For the FDS analyses performed, all potential radiant heat targets were located a minimum of 3 feet horizontally away from the fire.

All targets were simple in shape and not complex in nature

. The targets analyzed were a flat sheet metal panel and heat flux monitoring devices located independent of obstructions

. In both instances, the targets were of simple geometry and uniform composition.

E1-45 Because the model was not used outside of the limitations identified, the FDS predictions of heat flux were within the experimental uncertainty.

Regarding "completeness" uncertainties, these refer to the fact that a model may not be a complete description of the phenomena it is designed to predict

. Completeness uncertainty was addressed by the same process used to address the model uncertainty

. Model and completeness uncertainty are closely related, and it would be impractical to evaluate them separately

. Therefore, the discussion above for "model" uncertainties, as well as the conservative approaches discussed in the response to Fire Modeling (FM) RAI 06.a address "completeness" uncertainty.

For uncertainties specifically involved with ignoring the contents of a compartment, there were several areas of conservatism that mitigate the reduction in volume in HGL calculations

. The following assumptions were utilized within the fire modeling which lead to conservative results or reduced the effect of ignoring the contents of a compartment in the fire modeling analysis: If equipment was included in HGL calculations, a large heat sink was provided in the fire compartment, which would have generated lower HGL temperatures

. No heat passage through fire doors or dampers was considered in the HGL temperature calculations

. The material properties of concrete were applied to all exterior boundaries of the fire compartment

. Realistically, the heat from the HGL would be transferred to adjacent spaces more easily by a door or fire damper, which have a higher thermal conductivity than concrete

. Including these passages to adjacent compartments would have generated lower HGL temperatures

. Although obstructions within the room could reduce the effective volume when analyzing HGL temperatures, many of these obstructions (e.g., electrical cabinets, transformers) are not totally solid obstructions

. Electrical cabinets are generally not full of electrical components on the inside (i.e., they have large empty spaces within the cabinets)

. These empty spaces within the electrical cabinets reduce the effect of including obstructions for HGL temperature calculations.

The volume of some fire compartments was reduced in the fire modeling analysis to meet the validation range for compartment aspect ratio

. For fire compartments having an aspect ratio outside the validated range where detailed fire modeling was performed and whole room damage was not postulated, the height, length, or width of the fire compartment was "shortened" to values that fall within the validation range

. Shortening the dimensions of the fire compartment decrease s the overall volume of the compartment and create s a more severe condition

. E1-46 PRA RAI 01.a Section 2.4.3.3 of NFPA 805 states that the probabilistic safety assessment (PSA is also referred to as PRA) approach, methods, and data shall be acceptable to the AHJ, which is the

NRC. Regulatory Guide (RG) 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a FPRA and endorses, with exceptions and clarifications,

NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a FPP consistent with NFPA 805

. RG 1.200, "An Approach For Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk

-Informed Activities,

" describes a peer review process utilizing an associated ASME/ANS standard (currently ASME/ANS

-RA-Sa-2009) as one acceptable approach for determining the technical adequacy of the PRA once acceptable consensus approaches or models have been established for evaluations that could influence the regulatory decision

. The primary results of a peer review include the F&Os identified by the peer review and the ir subsequent resolution. Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessment s identified in LAR Attachment V that have the potential to impact the FPRA results and do not appear to be fully resolved:

a. F&O 2-47 against CF

-A1: This F&O 2

-47 disposition explains that additional components were added to what was originally a small population of components analyzed for their circuit failure mode likelihood

. A larger scope (more than 200 components) was analyzed to address this F&O. SR CF-A1 requires that conditional failure probabilities be assigned to risk significant contributors

. The resolution to this F&O stated that the top 50 scenarios were reviewed to identify potentially important components

. State the percentage of risk that these scenarios represent, and summarize how this review provides confidence that components associated with risk

-significant spurious operations have been identified. RESPONSE:

The top 50 scenarios for Core Damage Frequency (

CDF) and Large Early Release Frequency (LERF) for each unit contain between 83

-95% of the total model Fire CDF and LERF and included all scenarios that contributed greater than 1%, as well as many that contributed greater than 0.5% to the total Fire CDF and LERF

. The definition of significant for individual scenarios cited in the Fire Risk Quantification report is the same as the American Society of Mechanical Engineers (ASME) PRA Standard, which is scenarios that contribute greater than 1% of the applicable hazard group

. Thus the review identified components associated with risk significant spurious operations

. This review provides confidence that components associated with risk significant spurious operations have been identified in the Fire PRA model.

E1-47 PRA RAI 01.c Section 2.4.3.3 of NFPA 805 states that the probabilistic safety assessment (PSA is also referred to as PRA) approach, methods, and data shall be acceptable to the AHJ, which is the NRC. Regulatory Guide (RG) 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a FPRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a FPP consistent with NFPA 805

. RG 1.200, "An Approach For Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk

-Informed Activities,

" describes a peer review process utilizing an associated ASME/ANS standard (currently ASME/ANS

-RA-Sa-2009) as one acceptable approach for determining the technical adequacy of the PRA once acceptable consensus approaches or models have been established for evaluations that could influence the regulatory decision

. The primary results of a peer review include the F&Os identified by the peer review and their subsequent resolution.

Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessments identified in LAR Attachment V that have the potential to impact the FPRA results and do not appear to be fully resolved:

c. F&O 1-7 against FSS

-A4, FSS-A5 and FSS

-A6: Appendix D of the MCR categorizes a number of detached panels, which appear to be associated with offsite power (i.e., 9 1 through 9 8), as part of the main control board rather than as electrical panels

. In addition, the disposition to the F&O notes that electrical panels 3

-PNLA-009-0023CD and 3PNLA

-009-0023BA were reclassified as part of the main control board following the peer review

. Provide the basis for including these electrical panels as part of the main control board using guidance provided in Supplement 1 to NUREG/CR

-6850. RESPONSE:

According to NUREG/CR

-6850, Appendix L: The main control board (MCB) is defined as the collection of control panels inside the MCR of a nuclear power plant from which operators control the plant on a day

-to-day basis. The MCB would typically include the front face of the "horse shoe," other control or instrumentation display panels that are typically placed in full view of the areas where control room operators are expected to spend most of their time, and other panels in the control room proper that contain control switches or instrumentation displays that are used for plant control or emergency shutdown.

The MCB would typically not include the back panel of the main board, if such exists. The MCB would also not include those electrical panels devoted primarily to housing control relays, printed cards (e.g., signal conditioning cards), or all other devices that the operators do not directly use to maintain plant control or safe shutdown.

NUREG/CR-6850, Supplement 1, Section 5 explains that the above guidance from NUREG/CR-6850, Appendix L was not intended to allow inclusion of many various types of electrical cabinets and panels into the MCB count and analysis, but that it was intended to allow for some flexibility given the wide variability among control rooms around the country

. The supplement details a "bench-board" panel, with the following characteristics, that is routinely included as a n MCB: E1-48 These panels were serving as an integral part of the main plant monitoring and control functions.

These panels were located in the center of the operators' main work area.

These panels were manned on a nearly continuous basis.

Panels 9-23-1 through 9 8 in the Units 1 and 2 MCR and panels 3

-PNLA-009-0023CD and 3-PNLA-009-0023BA in the Unit 3 MCR meet the above criteria

. However, they would not be considered "bench-board" panels, for the following reasons:

Panels 9-23-1 through 9 8 make up the Unit 0 MCB, which contains several controls, including those for the diesel generators and off

-site power, for all three units

. These panels are of a similar design to those panels making up the MCB horseshoes in each of the control rooms and are located in front of the operators' desks

. The Unit 0 MCB serves an integral part to the main plant monitoring and control functions and is manned on a nearly continuous basis

.

Panels 3-PNLA-009-0023CD and 3

-PNLA-009-0023BA are connected to the Unit 3 MCB horseshoe in front of the operators' desks

. These panels were reclassified as MCBs following the peer review because controls for the emergency diesels are located within these panels

. This was also done for consistency with the Units 1 and 2 MCB panels 9-23-1 through 9 8 that contain similar controls

. Panels 3-PNLA-009-0023CD and 3-PNLA-009-0023BA were not analyzed using NUREG/CR-6850, Appendix L methodology, but instead failure of the entire control board was conservatively assumed.

E1-49 PRA RAI 01.g Section 2.4.3.3 of NFPA 805 states that the probabilistic safety assessment (PSA is also referred to as PRA) approach, methods, and data shall be acceptable to the AHJ, which is the NRC. Regulatory Guide (RG) 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a FPRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a FPP consistent with NFPA 805

. RG 1.200, "An Approach For Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk

-Informed Activities,

" describes a peer review process utilizing an associated ASME/ANS standard (currently ASME/ANS

-RA-Sa-2009) as one acceptable approach for determining the technical adequacy of the PRA once acceptable consensus approaches or models have been established for evaluations that could influence the regulatory decision

. The primary results of a peer review include the F&Os identified by the peer review and their subsequent resolution.

Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessments identified in LAR Attachment V that have the potential to impact the FPRA results and do not appear to be fully resolved:

g. F&O 2-55 against FSS

-E3: According to the disposition provided for F&O 2

-55, a quantitative characterization of the parameters used in the fire modeling of significant fire scenarios has not been completed

. Identify and characterize sources of uncertainty and related assumptions associated with parameters used for the modeling significant fire scenarios (e.g., a mean value and statistical representation uncertainty intervals)

. Provide a discussion of the criteria utilized to assess their importance to fire risk, and for those determined to be key sources of uncertainty for the FPRA, provide a sensitivity study of their impact on fire risk . If no more than a qualitative characterization is performed, provide detailed justification describing why meetin g FSS-E3 below CC

-II is acceptable for this application.

RESPONSE

Quantitative sensitivity analyses were performed after the peer review for some of the fire modeling methods used in the Fire PRA

. These are discussed in LAR Attachment V,Section V.2, with additional clarification provided in the LAR Supplement dated May 16, 2013. The subject of these sensitivity analyses are: The use of generic ignition frequencies based upon NUREG/CR

-6850, Supplement 1, Chapter 10 The reduction of the HRR to 69 kW for selected transient fires The crediting of very early warning fire detection systems (VEWFDS) and automatic suppression for fire scenarios in the Cable Spreading Room and Unit 1 Auxiliary Instrument Rooms For each of the above quantitative sensitivity analyses, BFN met the guidance specified in Regulatory Guide 1.174 for a Region II plant, with total CDF and LERF below 1E

-04/reactor (rx)-year (yr) and 1E-05/rx-yr for overall plant risk

. BFN also met the CDF/LERF criteria for a Region II plant, which allows a positive CDF of 1E

-05/rx-yr and positive -06/rx-yr for acceptable risk increases with the above sensitivities.

For the remaining sources of uncertainty, additional quantitative analysis of uncertainty intervals does not provide sufficient benefit over a qualitative characterization

. Because of the wide E1-50 range in modeling choices in NUREG/CR

-6850, the fire modeling guidance was applied in a conservative manner

. Several sources of uncertainty and related assumptions in the fire modeling analysis are discussed in the TVA response to FM RAI 06.a. Additional examples of the conservatisms present in the Fire PRA are: The 98th percentile HRR was generally used as the mean value, which would result in an overestimation of fire severity

. The ASME/ANS Fire PRA standard requires use of two point fire modeling, and for most risk

-significant fires, two points were modeled, one being the severity factor for damage to the ignition source only, and the other for effects beyond the source

. For the latter, the 98th percentile was predominately used which conservatively bounds a large range of HRRs

. This method results in the use of the 98 th percentile HRR for fire damage beyond the source, resulting in conservative target set damage assumptions.

NUREG/CR-6850 fire modeling assumptions that involved the growth and propagation of a fire include conservative peak heat release rates, conservative cable flame spread rates, and conservative cable tray propagation rule sets

. This directly leads to a reduction in effectiveness of detection/suppression and the time available for manual suppression, and tend to produce conservative estimates of the damage. These conservatisms were implicitly included in the fire damage states. The quantitative sensitivity analyses that were performed, combined with a qualitative assessment of fire modeling uncertainty, provides sufficient insight into conservatisms of the fire modeling analyses, as well as the effect of input parameter uncertainty on the results of the Fire PRA. The conservatisms in the selection of methods and data associated with NUREG/CR-6850 are expected to outweigh and influence the variability in results more than the parametric data uncertainty

. Therefore, meeting Fire Selection Scenario (FSS)-E3 Category I through the use of the current qualitative assessment of fire modeling input parameter uncertainty is sufficient and acceptable for the NFPA 805 application. E1-51 PRA RAI 01.k Section 2.4.3.3 of NFPA 805 states that the probabilistic safety assessment (PSA is also referred to as PRA) approach, methods, and data shall be acceptable to the AHJ, which is the NRC. Regulatory Guide (RG) 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a FPRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a FPP consistent with NFPA 805

. RG 1.200, "An Approach For Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk

-Informed Activities,

" describes a peer review process utilizing an associated ASME/ANS standard (currently ASME/ANS

-RA-Sa-2009) as one acceptable approach for determining the technical adequacy of the PRA once acceptable consensus approaches or models have been established for evaluations that could influence the regulatory decision

. The primary results of a peer review include the F&Os identified by the peer review and the ir subsequent resolution. Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessment s identified in LAR Attachment V that have the potential to impact the FPRA results and do appear to be fully resolved:

k. F&O 4-3 against SC

-B1: The disposition to this F&O states that "MSO scenario 1a is closed as the decision has been made on how it will be addressed.

" Multiple spurious operation (MSO) 1a could prevent a full scram

. The NFPA 805 Multiple Spurious Operation Review report indicates that emergency operating instructions will be modified to instruct operators to take actions, both internal and external to the MCR, in case of the inability to scram; however, as stated in Appendix 3, "this action was not modeled in the FPRA since ATWS [anticipated transients without scram]

is not modeled." Since fire induced failure could prevent a full scram it is not clear why this failure is not modeled in the FPRA. Provide further justification for not modeling this MSO scenario and related recovery actions in the FPRA. RESPONSE:

Multiple Spurious Operation (MSO) scenario 1a is based on the hot short conditions that could affect scram capabilities, as postulated in NRC Information Notice (IN) 2007

-07, "Potential Failure of All Control Rod Groups to Insert in a Boiling Water Reactor Due to a Fire.

" The three cases identified in IN 2007

-07 are assessed by Boiling Water Reactor Owners Group (BWROG)-Technical Paper (

TP)-11-011 (NEDO-33638, Rev. 0), "BWROG Assessments of Generic Multiple Spurious Operations (MSOs) in Post

-Fire Safe Shutdown Circuit Analysis for Operating BWR Plants.

" The BWROG assessment of the scram circuitry is consistent with and applicable to the BFN design.

The overall assessment of the condition described in NRC IN 2007-07 by the BWROG is that it represents a condition with a low likelihood of occurrence, with low safety significance and with multiple layers of defense

-in-depth currently in place each with the capability to either prevent the condition from occurring or to effectively mitigate the effects of the occurrence without consequence.

There were several reasons for this low probability

. First, because scram occurs early in a sequence following the early stages of fire growth, there is a high probability that the reactor would be manually scrammed before the postulated fire damage developed sufficiently to cause the hot short

. Second, it is unlikely that multiple hot shorts would exist simultaneously prior to initiating the automatic or manual scram signal

. Cable fire testing has shown that grounded AC hot shorts last for only a few minutes prior to shorting to ground

. Electric Power Research E1-52 Institute, Inc. (EPRI) testing determined that each hot short is more likely to ground before multiple shorts can occur that would fail the automatic scram function. Furthermore, the alternate rod insertion function (a DC powered system) is unaffected by the cases described in NRC IN 2007-07, and manual operator actions to de-energize the Reactor Protection Systems (RPS) reactor trip bus, which would result in a full scram, are available

. Therefore, there are no unique insights or conclusions from modeling the fire

-induced ATWS related to MSO scenario 1a

. This scenario is best addressed with a qualitative risk assessment, as is discussed above.

E1-53 PRA RAI 01.m Section 2.4.3.3 of NFPA 805 states that the probabilistic safety assessment (PSA is also referred to as PRA) approach, methods, and data shall be acceptable to the AHJ, which is the NRC. Regulatory Guide (RG) 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a FPRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a FPP consistent with NFPA 805

. RG 1.200, "An Approach For Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk

-Informed Activities,

" describes a peer review process utilizing an associated ASME/ANS standard (currently ASME/ANS

-RA-Sa-2009) as one acceptable approach for determining the technical adequacy of the PRA once acceptable consensus approaches or models have been established for evaluations that could influence the regulatory decision

. The primary results of a peer review include the F&Os identified by the peer review and the ir subsequent resolution. Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessment s identified in LAR Attachment V that have the potential to impact the FPRA results and do not appear to be fully resolved:

m. F&O 4-18 against AS

-B3 The disposition to this F&O only addresses flooding resulting from fire

-induced interfacing system loss

-of-coolant accidents (ISLOCAs). Summarize the consideration given to other potential fire

-induced flooding scenarios (e.g., location, extent, and consequences) and describe how such scenarios were treated in the FPRA. RESPONSE:

In addition to the ISLOCA scenarios addressed in the F&O response, several plant-specific scenarios that could result in fire

-induced flooding were identified by the MSO expert panel

. These potential fire

-induced flooding scenarios were the result of the draining of normally full discharge piping prior to spurious start of Emergency Core Cooling System (ECCS) pumps, which then would have caused a water hammer to the pump's discharge piping, resulting in potential flooding due to piping failure

. The scenarios were evaluated by the Fire PRA, and addressed both qualitatively and quantitatively.

E1-54 PRA RAI 01.n Section 2.4.3.3 of NFPA 805 states that the probabilistic safety assessment (PSA is also referred to as PRA) approach, methods, and data shall be acceptable to the AHJ, which is the NRC. Regulatory Guide (RG) 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a FPRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a FPP consistent with NFPA 805

. RG 1.200, "An Approach For Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk

-Informed Activities,

" describes a peer review process utilizing an associated ASME/ANS standard (currently ASME/ANS

-RA-Sa-2009) as one acceptable approach for determining the technical adequacy of the PRA once acceptable consensus approaches or models have been established for evaluations that could influence the regulatory decision

. The primary results of a peer review include the F&Os identified by the peer review and the ir subsequent resolution. Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessment s identified in LAR Attachment V that have the potential to impact the FPRA results and do not appear to be fully resolved: n. F&O 4-28 against FQ

-D1: Describe the FPRA model LERF

-reducing refinements discussed in the disposition to this F&O. RESPONSE:

The LERF-reducing refinements discussed in the disposition to the F&O are as follows:

The first LERF

-reducing refinement was modeling a modification that allows the control air headers in the individual reactor buildings to be depressurized

. Depressurization of these control air headers results in closure of the containment isolation valves

. This was modeled in the alternate shutdown analysis and the modification is described in LAR Attachment S, Table S-2, Modification 93

.

The second LERF

-reducing refinement was modeling the availability of the Electro

-Hydraulic Control (EHC) system which can be used to mitigate a stuck open Main Steam Isolation Valve (MSIV) event

. This is described in the disposition to F&O 4

-9 in LAR Attachment V, Table V-7. The stuck open MSIV event was modeled as a containment bypass (break outside containment) and was significant to the LERF for both fire

-induced and random failures

. The third LERF

-reducing refinement was applying circuit failure probabilities to basic events that model fire

-induced spurious opening of containment isolation valves

. These probabilities are more realistic than the probabilities that were initially applied to these basic events in the initial Fire PRA model.

E1-55 PRA RAI 01.t Section 2.4.3.3 of NFPA 805 states that the probabilistic safety assessment (PSA is also referred to as PRA) approach, methods, and data shall be acceptable to the AHJ, which is the NRC. Regulatory Guide (RG) 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a FPRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a FPP consistent with NFPA 805

. RG 1.200, "An Approach For Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk

-Informed Activities,

" describes a peer review process utilizing an associated ASME/ANS standard (currently ASME/ANS

-RA-Sa-2009) as one acceptable approach for determining the technical adequacy of the PRA once acceptable consensus approaches or models have been established for evaluations that could influence the regulatory decision. The primary results of a peer review include the F&Os identified by the peer review and their subsequent resolution.

Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessments identified in LAR Attachment V that have the potential to impact the FPRA results and do not appear to be fully resolved:

t. F&O 5-24 against FSS

-D10: The disposition to this F&O does not address the deficiencies identified by the peer review team and appears to be an editorial error

. Provide a disposition to this F&O.

RESPONSE

This is an editorial error within LAR Attachment V, Table V-7. The disposition to F&O 5-24 against FSS

-D10 is revised to read:

"As stated by the F&O, walkdowns were performed at Browns Ferry in accordance with applicable procedure

s. The analysis documentation has been updated to reflect this.

During target data entry into the electronic database and automatic validation of raceway targets, it was found that many raceways had formatting differences between the field ID tags and the electronic data

. In these cases, the targets were manually validated using drawings, the electronic database, and other resources

. The validation was used as a means to ensure the walkdown data was converted to the same format as each raceway appears in the electronic database

. The analysis documentation was also updated to discuss with more detail the raceway identification issues encountered and their resolution."

E1-56 PRA RAI 02 Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the NRC

. RG 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a fire PRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a fire protection program consistent with NFPA

-805. In letter dated July 12, 2006, to NEI (ADAMS Accession No

. ML061660105

), the NRC established the ongoing FAQ process where official agency positions regarding acceptable methods can be documented until they can be included in revisions to RG 1.205 or NEI 04

-02. Methods that have not been determined to be acceptable by the NRC Staff require additional justification to allow the NRC Staff to complete its review of the proposed method. Per NUREG/CR

-6850 Section 11.5.1.6, transient fires should at a minimum be placed in locations within the plant physical analysis units (PAUs) where CCDPs are highest for that PAU

(i.e., at "pinch points"). Pinch points include locations of redundant trains or the vicinity of other potentially risk

-relevant equipment

. Transient fires should be placed at all appropriate locations in a PAU where they can threaten pinch points

. Hot work should be assumed to occur in locations where hot work is a possibility, even if improbable, keeping in mind the same philosophy

. Describe how transient and hot work fires are distributed within the PAUs at your plant. In particular, identify the criteria for your plant used to determine where an ignition source is placed within the PAUs.

RESPONSE

The fire modeling analysis, which is an input to the Fire PRA, considered transient fires and hot work fires in each fire compartment

. For all fire compartments, with the exception of the cable spreading room (CSR) portion of Fire Compartment 16

-A, all accessible floor areas were postulated as possible transient and hot work ignition source locations

. By analyzing transient fires at all accessible floor areas within these fire compartments, all potential pinch point locations were considered for damage.

Transient and hot work fires were not postulated in locations within PAUs that were considered inaccessible (i.e., where precluded by design)

. Inaccessible areas are defined as those occupied by permanent fixtures such as plant equipment, structural features, piping, and cable trays. These permanent fixtures must either occupy the floor space entirely or be sufficiently low to the floor (i.e., 2 ft or less), so as to obstruct the placement of transient material.

As discussed in the fire modeling documentation, each transient and hot work fire scenario, analyzed in the CSR portion of Fire Compartment 16

-A, assumed damage to one of the cable trays with the highest Conditional Core Damage Probability (

CCDP) in the CSR

. If damage was considered to spread beyond the tray, the fire scenario conservatively assumed all targets in the CSR were damaged

. Using this methodology, the entire transient frequency of the CSR was apportioned to risk significant cable trays, which ensured that transient and hot work scenarios were analyzed at risk significant locations.

E1-57 PRA RAI 06 Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the NRC

. RG 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a fire PRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a fire protection program consistent with NFPA

-805. In letter dated July 12, 2006, to NEI (ADAMS Accession No

. ML061660105), the NRC established the ongoing FAQ process where official agency positions regarding acceptable methods can be documented until they can be included in revisions to RG 1.205 or NEI 04

-02. Methods that have not been determined to be acceptable by the NRC Staff require additional justification to allow the NRC Staff to complete its review of the proposed method. Fire-induced instrument failure should be addressed in the HRA per NUREG/CR

-6850 and NUREG-1921, "EPRI/NRC-RES Fire Human Reliability Analysis Guidelines." Describe how fire-induced instrument failure (including no readings, off

-scale readings, and incorrect/misleading readings) is addressed in the fire HRA.

RESPONSE

The HRA was addressed in accordance with NUREG/CR

-6850 and NUREG

-1921 guidance

. For each Human Failure Event (HFE), the EOI s, Alarm Response Procedures (ARPs), Abnormal Operating Instructions (AOIs), and applicable normal operating procedures were reviewed to determine the instruments needed for successful implementation of a required action. While the ARPs identify some instruments important to the required action, if the operator is only directed to the instrument by the ARP then the instruments and associated alarms were not credited in the Fire PRA model

. It should be noted that the instruments are cable routed and directly modeled in the Fire PRA model to logically fail if they are affected by the fire. Because they are treated this way in the model, they are not considered in the probability of failure of the individual HFEs.

When the credited instruments are placed in the logical structure of the fault tree, they are grouped according to the parameter they are measuring

. If any one of these parameters (i.e., groups) fails, then the fault tree logic fails the associated human action (i.e., logically OR'd). Each of the above groups (i.e., parameters) consists of events representing every instrument credited for measuring that parameter

. These instrument events are combined in a manner that requires all of them to fail in order to fail the parameter (i.e., logically AND'ed)

. For example, if reactor vessel level, suppression pool level, and containment pressure are needed for an action, the logic will have three groups of sub

-logic, one for each parameter

. These groups are logically OR'd so if any one parameter fails, the action cannot be successful

. Under each parameter group, the instrument trains events are logically AND'ed. The NSCA credited instrumentation trains that are known to be free of fire effects have been assumed to be identified in the fire safe shutdown procedures so the operator would know which instrument trains are reliable and which are not.

The analysis of undesired operator actions due to false or misleading instrument readings was performed as part of the undesired operator response evaluation

. This involved review of the EOIs, AOIs, and ARPs

. This review is documented in Section 5.6 and Attachment E of the Fire HRA calculation.

E1-58 PRA RAI 08 Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the NRC

. RG 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a fire PRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a fire protection program consistent with NFPA

-805. In letter dated July 12, 2006, to NEI (ADAMS Accession No

. ML061660105), the NRC established the ongoing FAQ process where official agency positions regarding acceptable methods can be documented until they can be included in revisions to RG 1.205 or NEI 04

-02. Methods that have not been determined to be acceptable by the NRC Staff or acceptable methods that appear to have been applied differently than described require additional justification to allow the NRC Staff to complete its review of the proposed method.

Explain how the effect of smoke on equipment was evaluated, e.g., by using the guidance provided in Appendix T of NUREG/CR- 6850. RESPONSE:

Smoke damage to equipment was evaluated using the guidance provided in NUREG/CR

-6850, Appendix T

. Consistent with Section T.3.1, short-term smoke damage was only assumed to result from a severe smoke exposure condition

. Therefore, components housed in the same electrical panel as the fire source, or in an electrical panel directly connected via an open bus duct, were assumed damaged by smoke, unless a specific installation feature, such as the features identified in NUREG/CR

-6850, Section T.3.1, precludes such damage

. E1-59 PRA RAI 09 Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the NRC. RG 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a fire PRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a fire protection program consistent with NFPA

-805. In letter dated July 12, 2006, to NEI (ADAMS Accession No

. ML061660105), the NRC established the ongoing FAQ process where official agency positions regarding acceptable methods can be documented until they can be included in revisions to RG 1.205 or NEI 04

-02. Methods that have not been determined to be acceptable by the NRC Staff or acceptable methods that appear to have been applied differently than described require additional justification to allow the NRC Staff to complete its review of the proposed method.

Describe how sensitive electronics components are identified and treated

. Additionally, indicate whether this treatment is consistent with guidance in NUREG/CR-6850 and FAQ 13-0004. RESPONSE:

TVA considers temperature sensitive electronic equipment to be equipment that is susceptible to lower thermal damage thresholds (i.e., solid

-state control components)

. Due to the limited guidance at the time of the fire modeling analysis, all electronic cabinets were conservatively identified as sensitive electronics

. This treatment is consistent with the guidance in NUREG/CR-6850, "EPRI/NRC-RES Fire PRA Methodology for Nuclear Power Facilities," and Fire PRA FAQ 13

-0004. All fire modeled compartments have been examined to determine whether sensitive electronic equipment may be exposed to conditions exceeding the damage criteria recommended by NUREG/CR-6850, Section H.2 (i.e., 3 kW/m2 and 65°C)

. Damage was considered for the electronic cabinets based on the following criteria.

The HGL heights and temperatures were evaluated for varying room sizes and configurations using CFAST; the results are included in the fire modeling documentation

. The analyses identified generic scenarios that may generate an HGL that was capable of damaging sensitive equipment and categorized the results based on the fire compartment sizes and configurations

. All fire modeled compartments were compared to these generic categories

. If the applicable category and fire conditions indicated damage to sensitive electronics for a specific fire compartment, all electronic cabinets within the compartment were assumed damaged.

The heat flux for cabinet internals was evaluated using FDS; the results are included in the fire modeling documentation

. The analysis determined that the metal housing on temperature sensitive equipment is capable of significantly reducing damaging heat fluxes to internal components such that damage to temperature sensitive equipment, inside of cabinets, via radiant heat is bounded by the use of a ZOI for thermoset cables. Temperature sensitive equipment, located inside of cabinets, was assumed damaged when the critical damage criteria for thermoset cable, as recommended by NUREG/CR

-6850, Section H.2 (i.e., 11 kW/m2), was exceeded on the outer surface of the cabinet

. This treatment is consistent with the guidance in Fire PRA FAQ 13-0004. E1-60 PRA RAI 14 Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the NRC

. Section 2.4.4.1 of NFPA 805 further states that the change in public health risk arising from transition from the current fire protection program to an NFPA 805 based program, and all future plant changes to the program, shall be acceptable to the NRC

. RG 1.174 provides quantitative guidelines on CDF, LERF, and identifies acceptable changes to these frequencies that result from proposed changes to the plant's licensing basis and describes a general framework to determine the acceptability of risk

-informed changes

. The NRC staff review of the information in the LAR has identified the following information that is required to fully characterize the risk estimates.

Implementation Items 32 and 33 do not commit to verifying the validity of the reported risk results. When the effect of a plant modification or procedure change has been included in the PRA before the modification or procedural change has been completed, the models and values used in the PRA are necessarily estimates based on current plans

. The as-built, as

-operated facility after the modification or procedural change is completed may be different than the current estimates

. Add an implementation item that verifies the validity of the reported change in risk subsequent to completion of all PRA

-credited modifications, procedures updates, and implementation items

. This item should include a plan of action should the as

-built, as

-operated change-in-risk exceed the acceptance guidelines.

RESPONSE

The Fire PRA sources of uncertainty, including those associated with planned modifications and recovery actions, were assumed in the base case Fire PRA model

. TVA recognized that this approach introduced uncertainty in the results because the actual modification may vary from the modification assumed or the modification may not function as modeled

. These assumed modifications were documented in the Fire PRA Notebooks

. Plant and model configuration and control mechanisms are in place to ensure that the fire model will be updated to reflect the

as-installed modifications

. Sensitivity analyses were performed on key modifications; therefore, key attributes of the modification are understood and allow for PRA input (insights) in the design modification process.

LAR Attachment S, Table S-3, Implementation Items 32 and 33 are the TVA commitments to update the Fire PRA model to reflect design modifications and fire response human actions

. These two implementation items require that the Fire PRA and Fire HRA are reviewed after the modifications are complete and to make any necessary changes to the Fire PRA model and Fire

HRA. However, these implementation items are revised as follows to ensure clarity of the commitment.

Implementation Item 32 is modified to read:

"Update the Fire PRA model after all modifications are complete (returned to operation) and in their as

-built configuration

. The update will include a verification of the validity of the reported change in risk on as

-built conditions after the modifications are completed." Implementation Item 33 is modified to read:

"Update the Fire HRA upon completion of all procedure updates, all modifications and all training

. The update will include a verification of the validity of the reported change in risk on as

-built conditions after the procedures updates, modifications, and training are completed."

The modified implementation items will ensure that the Fire PRA model and Fire HRA are reviewed when a modification or procedure change is finalized to ascertain deltas from the

as-modeled configurations to the as

-built, as

-operated configuration and make appropriate E1-61 updates. Verification of the validity of the reported risk results will be performed as the Fire PRA model and Fire HRA are updated.

TVA is committed to meeting the risk acceptance criteria in RG 1.174, to ensure the as

-built, as-operated change

-in-risk does not exceed the acceptance guidelines

. As specified in LAR Section 4.8.2, the change

-in-risk is not expected to be significant for any refinements made to the as-built, as

-operated model after modifications are completed

. However, if the change-in-risk exceeds the acceptance guidelines, the model will be reassessed and new modifications/refinements will be implemented, as necessary, to meet the acceptance guidelines.

E1-62 PRA RAI 18 Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the NRC

. Section 2.4.4.1 of NFPA

-805 further states that the change in public health risk arising from transition from the current fire protection program to an NFPA

-805 based program, and all future plant changes to the program, shall be acceptable to the NRC

. RG 1.174 provides quantitative guidelines on CDF, LERF, and identifies acceptable changes to these frequencies that result from proposed changes to the plant's licensing basis and describes a general framework to determine the acceptability of risk

-informed changes

. The NRC staff review of the information in the LAR has identified the following information that is required to fully characterize the risk estimates.

VFDRs and how the additional risk of RAs for each of the fire areas were determined but does not include any discussion about the use of acceptable methods and no discussion of how the special case of MCR was assessed

. The described approach is based on "removing" VFDRs by eliminating fire induced failure and giving perfect credit to RAs associated with VFDRs

. It is not clear whether there were any exceptions to this approach, such as in the case of the MCR

.

Please clarify the following:

a. ing room abandonment scenarios.

b. Explain whether there were any exceptions to the approach described in Section W.2.1 (e.g., when there was not enough resolution to the PRA), and describe the approach used for those exceptions.

RESPONSE

Part a.

Delta () scenarios are calculated in a similar method as any other scenarios. That is, the CDF and LERF for the applicable compl iant abandonment scenario was subtracted from the CDF and LERF for the applicable post

-transition (PT) abandonment scenario.

The abandonment scenarios in the PT case were modeled with two parallel shutdown strategies

. The first shutdown strategy utilized the installed backup control panel (i.e., panel 25-32 for each unit), and utilized the RCIC System or the Low Pressure Coolant Injection (LPCI) System for injection and suppression pool cooling (SPC) for decay heat removal

. The risk related to the failure of required operator actions, including those that occur outside the primary control station to ensure success of this strategy, was modeled for this shutdown strategy. The second, parallel shutdown strategy utilized the Emergency High Pressure Makeup (EHPM) pump (i.e., the pump being added as described in LAR Attachment S, Table S-2 Modification 35) as a source of injection and hardened wet

-well vent as a means of decay heat removal

. The risk related to the failure of required operator actions is credited along with the first shutdown strategy, providing two independent redundant, safe shutdown strategies.

The second shutdown strategy is based on the implementation of two modifications: 1) the installation of an EHPM pump that provides an additional source of makeup, and 2) the hardened wet

-well vent modification described in LAR Attachment S, Table S

-2, Modification 51, that allows increased usage of the wet

-well vent

. Additionally, there is a LERF-specific modification (i.e., LAR Attachment S, Table S

-2, Modification 93) that allows for isolating and E1-63 venting control air to the containment isolation, such that the containment isolation valves will fail closed.

The abandonment scenarios in the compliant plant were modeled with operator actions to utilize the installed backup control panel, similar to the PT case

. However, in the compliant case, the actions that need to occur outside of the primary control station were considered as occurring at the primary control station or to be completely successful (i.e., are not required) for safe shutdown, and do not contribute to the risk of the compliant plant.

Part b. The approach described in LAR Attachment W, Section W.2.1 was used for all Variance from Deterministic Requirements (VFDRs) where the Fire PRA modeled the NSCA equipment associated with the VFDR

. An extension of the process was used for cases where the Fire PRA did not model the exact equipment associated in the VFDR as identified in the NSCA

. For example, the PRA did not model the logic associated with the Alternate Shutdown Cooling mode of RHR

. This mode was credited in the NSCA as a means of Decay Heat Removal (DHR). To model the associated with Alternate Shutdown Cooling VFDRs

, the compliant plant model was reviewed and basic events associated with an equivalent higher level function (i.e., suppression pool cooling as a means of DHR) were "toggled off." The generic approach was to ensure the compliant model restored the higher level function associated with failure of the VFDR

. Another example of this approach included process monitoring VFDRs, in which the Fire PRA credited different instrumentation than the specific VFDR. The basic events associated with failure of the function supported by the instrumentation were reviewed and the basic events "toggled off" for the equivalent function, as required. For systems not required to be modeled in the Fire PRA such as electric board room HVAC, there was no or associated with the VFDR because the equipment did not contribute to the risk of compliant plant

. Thus there was no toggling of basic events or recovery actions for these cases.

The specific case of the modeling of MCR abandonment scenarios and calculation of in LAR Attachment W, Section W.2.1; by modeling success of recovery actions or by modeling the actions being performed at the primary control station

. A description of the modeling and treatment of abandonment scenarios is provided in the response to PRA RAI 18, Part a. E1-64 PRA RAI 19 Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the N RC. Section 2.4.4.1 of NFPA 805 further states that the change in public health risk arising from transition from the current fire protection program to an NFPA 805 based program, and all future plant changes to the program, shall be acceptable to the NRC

. RG 1.174 provides quantitative guidelines on CDF, LERF, and identifies acceptable changes to these frequencies that result from proposed changes to the plant's licensing basis and describes a general framework to determine the acceptability of risk

-informed changes

. The NRC staff review of the information in the LAR has identified the following information that is required to fully characterize the risk estimates.

LAR Attachment W, Table W-2 presents CDF for each fire area including values that represent exceptionally large risk reductions (e.g., Fire Areas 03

-02, 08, 16, and 26)

. The total CDF for Units 1, 2, and 3 respectively is

-5.37E-4/yr, -4.67E-4/yr, and

-5.45E-4/yr. Section 3.2.5 of RG 1.205 states that risk decreases may be combined with risk increases for the purposes of evaluating combined changes in accordance with Regulatory Positions 2.1.1 and 2.1.2 of RG 1.174. Accordingly, both individual and cumulative risk effects should be evaluated in detail.

a. Given that the submitted application represents a change that combines risk increases with risk decreases, please provide the total increase and total decrease in the CDF and LERF. b. LAR Attachment W summarizes the risk significant scenarios in the variant case

. Summarize the top few risk significant scenarios for risk significant fire areas in the compliant case.

c. Tables W.8, W.9, and W.10 of the LAR show that LERF is generally two to three orders of magnitude less than the CDF for fire areas

. Discuss why the fire LERFs are significantly lower percentage

-wise than those for internal events for most fire areas. RESPONSE:

Part a.

sum of the risk increases and decreases

. The total risk increase and decrease at the Fire Area level for BFN Units 1, 2, and 3 are provided below:

Unit 1 -1 1 -1 Total 1 Total 1 Unit 1 1.89E-06 -5.39E-04 2.70E-07 -7.74E-08 -5.37E-04 1.93E-07 Unit 2 9.92E-07 -4.67E-04 1.46E-07 -1.14E-07 -4.67E-04 3.19E-08 Unit 3 0.00E+00 -5.45E-04 1.58E-07 -3.92E-08 -5.45E-04 1.18E-07

. The differences are minor and are the result of rounding

. columns in this table.

Part b. Summarized below are the top few risk significant scenarios for risk significant fire areas in the compliant case. In each scenario in the compliant case, basic events associated with VFDRs E1-65 on the credited train in NSCA have been removed from being failed by the fire

. The core damage and large early release scenarios described here are the results of:

Fire-induced failures (with the exception of those associated with the credited train). Random failures of the credited train.

The following discussion of the risk significant scenarios in FA 16 applies to BFN Units 1, 2, and 3: Unit Scenario IGF (Ignition Frequency)

CCDP (Conditional Core Damage Probability)

CDF (Core Damage Frequency)

Unit 1 16-K.023-CAB-SUP 8.31E-05 9.64E-02 8.00E-06 Unit 1 16-K.024-CAB-SUP 8.31E-05 9.64E-02 8.00E-06 Unit 2 16-M.022-CAB-SUP 8.31E-05 9.64E-02 8.00E-06 Unit 2 16-M.023-CAB-SUP 8.31E-05 9.64E-02 8.00E-06 Unit 3 16-O.025-CAB-SUP 8.31E-05 9.64E-02 8.00E-06 Unit 3 16-O.024-CAB-SUP 8.31E-05 9.64E-02 8.00E-06 The risk in FA 16 is dominated by fire scenarios that are modeled as control room abandonment scenarios

. The core damage risk for control room abandonment is modeled as the set of operator actions required in the main control room prior to transferring command and control to the backup control panel, and the set of actions that need to occur at the backup control panel (i.e., panel 25-32) and locally to utilize the RCIC and LPCI Systems with control of SRVs

. In the compliant case, the local actions that need to occur outside of the primary control station are considered as occurring at the primary control station or to be completely successful (i.e., are not required) for safe shutdown, and do not contribute to the risk of the compliant plant.

The following discussion applies to the risk significant fire scenarios in FA 03-03 for the Unit 1, 2 and 3 compliant Fire PRA models: Unit Scenario IGF CCDP CDF Unit 1 03-03.4001-C 1.31E-04 1.13E-01 1.48E-05 Unit 1 03-03.3000-T-3-G-2 3.62E-05 1.13E-01 4.10E-06 Unit 2 03-03.4001-C 1.31E-04 1.13E-01 1.48E-05 Unit 2 03-03.3000-T-3-G-2 3.62E-05 1.13E-01 4.10E-06 Unit 3 03-03.4001-C 1.31E-04 1.25E-01 1.64E-05 The fire scenarios result in accident sequence GTRAN

-5A, which is a general transient sequence. Accident sequence GTRAN

-5A is an isolation accident (Power Conversion System not credited in the Fire PRA) with successful early high pressure injection from either the HPCI or RCIC Systems. Long term high pressure injection from the HPCI or RCIC System, or early suppression pool cooling fails, and Control Rod Drive (

CRD) System fails (not credited in the Fire PRA). Manual depressurization with two SRVs at Heat Capacity Temperature Limit (HCTL) about four hours after the scram is successful and subsequent low pressure injection with either the RHR or Core Spray (

CS) System fails. In these fire scenarios, the SPC, LPCI, and CS Systems are failed due to loss of the Emergency Equipment Cooling Water (EECW) System to both RHR and CS Room Coolers

. Both loops of E1-66 the RHR System are failed due to a combination of fire induced and random failure of loss of power supplies to EECW pumps and failure of the pumps themselves

. Off-site power is failed due to the fire; EECW Pumps A3 and C3 are failed due to the fire, and random failure to run of DG C which supplies power to EECW Pump B3 results in the loss of EECW

. The loss of the EECW System results in the loss of the NSCA credited path for Units 1 (LPCI Pump 1A), 2 (LPCI Pump 2D), and 3 (LPCI Pump 3B). Core damage is caused by loss of injection and the RPV is at low pressure.

The following discussion applies to the risk significant fire scenarios in Fire Area 03-03 for the Unit 3 compliant Fire PRA model:

Unit Scenario IGF CCDP CDF Unit 3 03-03.022-BCHG-2 7.46E-05 1.05E-01 7.82E-06 The fire scenario results in accident sequence GTRAN

-7, which is a general transient sequence. Accident sequence GTRAN-7 is an isolation accident (PCS not credited in the Fire PRA) with all high pressure injection failed immediately after scram

. Depressurization with two SRVs is successful

. Low pressure injection from the RHR or CS System is successful but must be initiated within 30 minutes

. Suppression pool coo ling fails but decay heat removal wit h drywell sprays or the primary containment vent is successful

. Without SPC, the Primary Containment Pressure Limit is reached in approximately 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />. The successful vent fails the RHR and CS Systems due to inadequate net positive suction head (NPSH). Drywell (

DW) spray is not successful and late (post vent) low pressure injection from the CRD System (not credited in the Fire PRA), condensate, RHRSW, SDC, CS from CST, and RHR from CST fails. Core damage is caused by loss of injection and the RPV is at low pressure.

In this fire scenario, all high pressure injection is failed due to the fire affecting the HPCI steam supply valves, HPCI instrumentation, and HPCI suction and discharge valves

. The RCIC System is failed due to the fire affecting the RCIC steam supply valves and instrumentation.

SPC Loop II is failed due to the fire affecting the RHR Pump D and the fire affecting the RHRSW Pumps B1 and off

-site power failure in addition to random failure of DG C power supply for RHRSW Pump B2 for RHR Heat Exchanger B cooling

. The loss of RHR Heat Exchanger B cooling results in the loss of the NSCA credited LPCI path using RHR Pump 3B. SPC Loop I is failed due to the fire affecting the beakers that supply Reactor Motor Operated Valve (RMOV) Board 3D. This also results in the loss of Loop I SDC and RHRSW injection, and LPCI from CST.

The following discussion applies to the risk significant fire scenarios in FA 08 for the Unit 1, 2 and 3 compliant Fire PRA models Unit Scenario IGF CCDP CDF Unit 1 08.001-CAB 6.07E-06 2.90E-02 1.76E-05 Unit 2 08.001-CAB 6.07E-06 3.59E-02 2.18E-05 Unit 3 08.001-CAB 6.07E-06 2.77E-02 1.68E-05 The fire scenarios result in accident sequence GTRAN

-5A, which is a general transient sequence. Accident sequence GTRAN

-5A is an isolation accident (i.e., PCS not credited in the Fire PRA) with successful early high pressure injection from either the HPCI or RCIC System. Long term high pressure injection from the HPCI or RCIC System, or early suppression pool cooling fails, and the CRD System fails (not credited in the Fire PRA). Manual depressurization E1-67 with two SRVs at HCTL about four hours after the scram is successful and subsequent low pressure injection with either the RHR or CS System fails. In this fire scenario, Loop II of SPC, LPCI, and CS for Units 1 and 2 are failed due to the fire affecting the RHR Pumps B, D, and associated valves

. Loop I for Unit 3 is lost due to the fire affecting the power from 4 kV Shutdown Board C and D to EECW Pumps B3 and D3, which supply RHR and CS Room Coolers

. The credited NSCA path for Units 1 and 2 (RHR Pumps 1A and 2C) and the path for Unit 3 (RHR Pump 3B) is lost due to the loss RHR Room cooling of the credited Loop

. Room cooling is failed due to a combination of fire induced and random failure

. Specifically room cooling to Loop I for Units 1 (LPCI Pump 1A) and 2 (LPCI Pump 2C) and Loop II for Unit 3 (LPCI Pump 3B) is lost due to test and maintenance unavailability of EECW Pump A3 or C3 (i.e., one of the two NSCA credited and required EECW pumps). Core damage is caused by loss of injection and the RPV is at low pressure.

Part c. The ratio of CDF to LERF in the Fire PRA is lower than the ratio of CDF to LERF in the internal events model for one primary reason

. Fire PRA CDF results contain a higher proportion of late sequence failures compared to early sequence failures than in the internal events model

.

Additionally, sequences that result directly in a Large Early Release (LER) in both the internal events and the Fire PRAs can be influenced by the fundamental modeling treatments in each analysis, which means the ratios become dependent on unique aspects of the particular PRA modeling used in that analysis.

There are a number of core damage sequences that are modeled to result directly in a LER in both the internal events PRA and the Fire PRA

. An example of this type of sequence is an ISLOCA on the RHR Shutdown Cooling suction lines

. In the internal events model, the likelihood of the ISLOCA scenario is modeled using generic spurious opening probabilities based on internal events failure data from industry references

. In the Fire PRA, the likelihood of the ISLOCA scenario is modeled using generic fire ignition frequencies from industry data and also depends on fire modeling, which accounts for hazard intensity

. The likelihood of the ISLOCA scenario depends on the ignition source, as well as distance to the targets and other spatial considerations

. In addition, the Fire PRA model s circuit failure likelihood, which is dependent on the circuit configuration

. These differences between the internal events modeling and the Fire PRA modeling demonstrate how a similar LER sequence likelihood can be affected by the fundamental modeling treatments used in the internal events model and the Fire PRA model. Secondly, the addition of the EHPM pump and the hardened wet

-well valve modifications reduced the likelihood of all core damage sequences that include loss of makeup and/or loss of decay heat removal

. Most of the core damage sequences that are considered early for the LERF analysis include credit for the EHPM pump providing an additional source of makeup water. The EHPM pump is not modeled in the Internal Events Model

. In particular, the additional source of makeup water has greater importance for early core damage sequences than for those that are late core damage sequences

. Therefore, one effect of adding the EHPM pump and hardened wet

-well vent modifications is that the CDF becomes more heavily weighted by late core damage sequences in the Fire PRA, and late sequences have a significant lower contribution to LERF

. This results in a relatively higher CDF as compared to LERF for internal events.

E1-68 PRA RAI 21 Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the NRC

. Section 2.4.4.1 of NFPA

-805 further states that the change in public health risk arising from transition from the current fire protection program to an NFPA

-805 based program, and all future plant changes to the program, shall be acceptable to the NRC

. RG 1.174 provides quantitative guidelines on CDF, LERF, and identifies acceptable changes to these frequencies that result from proposed changes to the plant's licensing basis and describes a general framework to determine the acceptability of risk

-informed changes

. The NRC staff review of the information in the LAR has identified the following information that is required to fully characterize the risk estimates.

LAR Attachment S identifies numerous proposed plant modifications

. Although many modifications are extensive most are only briefly described

. Please provide the following:

a. Summarize how the design of the new features has been provided to the PRA analysts for use in modeling the risk impact (e.g., as brief descriptions or completed design package).

b. Summarize the new models that have been developed (e.g., what basic events, fault trees, event trees, and failure data). c. Describe how the effect of all new cables has been evaluated (e.g., have areas that credit for the new equipment is being taken been identified as areas where required cables may not be routed).

d. In regard to the disposition to F&

O 9-2, which indicates that the single event (i.e., 0.1) representation of the new emergency high

-pressure injection pump has been replaced with more detailed modeling logic, describe the scope and content of any interviews, thermal-hydraulic analyses, and procedure reviews performed to support the PRA modeling of this future modification

. Additionally, provide a review of modified accident sequences (including timing considerations), system dependencies, flow path considerations, required operator actions, and any new parameter estimates (including unavailability due to testing and maintenance

). RESPONSE: Part a.

Development of the modifications and reflecting the effect of the modifications in the Fire PRA is an iterative process

. As the need for modifications was identified during the FRE process, modifications to resolve separation issues were scoped

. The design scoping involved the review of plant documentation (i.e., reports, drawing, calculations, specifications)

. These were described and translated to the PRA analysts in terms of Basic Events that would be eliminated for each fire area or fire scenario

. This information was then used to reflect the proposed modifications in the Fire PRA and became criteria for further development of the design

. Adjustments to the Fire PRA may be needed as the modifications progress

. This requirement is covered by LAR Attachment S, Table S

-3, Implementation Item 32.

For modifications that provide a new capability, such as the EHPM pump, the feature was modeled from a design concept

. The design concept was developed through review of plant documentation (i.e., reports, drawing, calculations, specifications) in order to provide attributes and capabilities that could be modeled

. PRA modeling assumptions then became part of the design requirements for the modification.

E1-69 Part b. Plant modifications have been implemented into the Fire PRA model using the following approaches.

1) Basic Events (BEs) were altered to not fail due to fire to reflect added separation.

The majority of plant modifications were implemented into the Fire PRA model using this approach. These were primarily plant modifications addressing separation issues

. For these plant modifications, the Fire PRA model BEs were altered in the "FireImpact" table of the FRANX software

. These BEs were altered to not fail due to fire for the specific fire scenarios where the plant modification would prevent the fire damage.

2) Modeling assumptions were developed based on plant modification scoping.

A smaller set of plant modifications were developed to ensure the applicability of certain modeling assumptions and the availability of systems or components

. These plant modifications did not directly relate into the failure of BEs

. Examples of these types of plant modifications are the removal of abandoned equipment to remove ignition sources, installation of incipient detection, and installation or changes to fire suppression systems.

3) Modeling new BEs, HRAs, and fault tree logic to represent new equipment or controls. Examples of modifications for which new fault tree logic was developed are the EHPM pump and the hardened wet

-well vent modifications.

The EHPM pump is a new system to provide reactor inventory

. The response to PRA RAI 21, Part d describes the modeling of the EHPM pump.

For the hardened wet-well vent modifications, a new backup pneumatic actuation capability was added

. Basic events with failure rates were added to the fault tree logic for all three units. Failure rates were based on existing type codes and data used in the internal events PRA model.

Part c.

New cable routes are developed as part of the design process for each modification

. Therefore, the PRA model will not reflect future cable locations until after the design for each modification is completed; the PRA would be updated after the modification is implemented

. A qualitative assessment will be done during modification scoping to ensure that new cable locations will not have a negative effect on risk results

. Modifications involving new cables or cable routes will be developed such that the proposed new route avoids where practical, fire areas where the dependent equipment is otherwise available

. This is usually done by not introducing the cable route to any new fire areas or by placing the new cable route in fire areas where the dependent equipment already has other fire vulnerabilities

. LAR Attachment S, Table S

-3, Implementation Item 32 as supplemented in the TVA response to PRA RAI 14, requires a check following modification implementation to verify the validity of the reported change

-in-risk on as

-built conditions after the modifications are completed. TVA will take action as appropriate if change in risk is not within acceptance guidelines.

E1-70 Part d. On the event tree level, the Fire PRA model credits the EHPM system as a final injection system for all sequences that involve a loss of all other injection sources

. To achieve an acceptable end state, decay heat removal must also be successful

. This final EHPM system is also credited in sequences where decay heat removal has failed, leading to primary containment failure, but only for the purpose of assessing a different level 1 plant damage state (i.e., containment fails before core uncovery). Similar but distinct fault tree models were developed to accommodate timing constraints for the initiation of the system

. This is the only sequence dependent variable because the system does not depend on vessel depressurization, reactor building environment, or the suppression pool as a suction source

. These timing criteria were based on the structure of the general transient event tree that incorporated a split fraction for various success criteria timings.

The timing criteria were not based on any specific thermal-hydraulic analysis during the modeling of this initial design concept

. System successes in these cases were judgments made in order to estimate the potential contribution of this system for these sequences

. The four hour timing criterion was based on the timing value typically used in the PRA to separate initial and long term injection

. The PRA definition of initial injection success was continued injection for at least four hours. The next major timing criterion for the EHPM system occurs more than eight hours after the scram, and would be after normal injection was lost due to a successful primary containment vent

. In this case, the EHPM pump's suction source would not be dependent on suppression pool temperature

. Separate timing was not considered for this initiation action because it would not provide enough human error probability (HEP) improvement over the four hour case to warrant the additional model complexity.

Each EHPM system fault tree is connected to the PRA fault tree via the top logic. Each case includes the normal power supply dependency (i.e., 4kV Unit Board 1C for Unit 1, 4kV Unit Board 2C for Unit 2, and 4kV Unit Board 3C for Unit 3), DC control power, the suction supply dependency (i.e., the CST), an estimated test and maintenance unavailability, random check valve failure events, and failure to start and failure to run events

. This suction supply dependency model also includes an operator action to refill the CST because it was assumed the CST would not have the necessary inventory to provide 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of makeup

. For the four hour initiation case an additional alternate power source from the proposed supplemental diesel was built into the fault tree logic and includes an operator action to transfer to that power supply

. Because final design information is not available, all of these reliability numbers with the exception of the normal power supply were best estimates based on existing similar equipment

. The normal power supply models were already in the internal events fault tree

.

The development of the system design information to support the PRA modeling was a joint effort between the BFN NFPA 805 site team and the PRA team

. E1-71 PRA RAI 23.c Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the NRC

. RG 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a fire PRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a fire protection program consistent with NFPA

-805. RG 1.200 describes a peer review process utilizing ASME/ANS-RA-Sa-2009 as one acceptable approach for determining the technical adequacy of the PRA once acceptable consensus approaches or models have been established

. The primary results of a peer review are the F&Os recorded by the peer review and the subsequent resolution of these F&Os.

Clarify the following dispositions to fire F&Os and SR assessment identified in LAR Attachment U that have the potential to impact the fire PRA results and do not appear to be fully resolved:

c. 2-35 (Finding against LE

-D1 Met at CC

-I/II) Section 3.1.4 of the internal events LERF report (i.e., LE.01 - BFN Probabilistic Risk Assessment - LERF Analysis) presents unit differences including differences in the ability to cross

-tie Residual Heat Removal (RHR) and differences in RHR emergency power alignment

. These differences are related to ultimate containment structural capability, but appear to impact LERF values, given that RHR is a contributor to In

-Vessel Recovery

. Clarify, whether unit differences related to RHR were explicitly modeled in the internal events PRA to determine LERF

. If these differences were not incorporated, justify this modeling exclusion or provide an estimate of the impact of this exclusion on internal events and fire PRA CDF and LERF.

RESPONSE

The unit differences related to the RHR System and other systems were explicitly modeled in the internal events PRA to determine LERF

. These differences were carried over into the Fire PRA model

. The system failure logic for the LERF model is fed from the level 1 CDF model

. Therefore the RHR System model differences that are present in the level 1 CDF model are also accounted for in the LERF model.

The following is a summary of the RHR unit differences explicitly accounted for in the PRA model: Unit 2 has a unit

-to-unit cross

-tie from both Unit 1 and Unit 3

. One loop on Unit 2 can be cross-tied from Unit 1; the other loop on Unit 2 can be cross

-tied from Unit 3

. Units 1 and 3 only have the cross

-tie from Unit 2 and only one loop of those units can be backed up. Unit 2 has a 24

-inch Loop I to Loop II cross

-tie line including an isolation valve on the discharge side of the RHR heat exchangers

. The Loop I to Loop II cross

-tie is not modeled by the PRA because power has been removed from the isolation valve and there is no procedural application for using this alignment

. A similar cross

-tie was removed from Unit 1 (i.e., the line was cut and capped)

. The Unit 3 cross

-tie has a manual isolation valve

. This loop

-to-loop cross

-tie capability is not modeled in the PRA. The breakers associated with the unit

-to-unit cross

-tie isolation valves are maintained in the open configuration (i.e., OFF) during normal power operations

. An exception is the Unit 3 discharge cross

-tie valve breaker which is maintained closed (i.e., ON) during power operations.

E1-72 PRA RAI 23.f Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the NRC

. RG 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a fire PRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a fire protection program consistent with NFPA

-805. RG 1.200 describes a peer review process utilizing ASME/ANS-RA-Sa-2009 as one acceptable approach for determining the technical adequacy of the PRA once acceptable consensus approaches or models have been established

. The primary result s of a peer review are the F&Os recorded by the peer review and the subsequent resolution of these F&Os.

Clarify the following dispositions to fire F&Os and SR assessment identified in LAR Attachment U that have the potential to impact the FPRA results and do not appear to be fully resolved:

f. 3-31 (Finding against QU

-D6 Met at CC

-I/II/III, QU

-F3 Met at CC

-I/II/III, QU

-F6 Not Met at CC-I/II/III, and LE

-G6 Not Met at CC

-I/II/III)

Section 6.3.2.3 of the internal events Quantification report (QU - Probabilistic Risk Assessment - Quantification) states that the top 100 CDF and LERF cutsets were considered to define "significance

". Explain what contribution the top 100 CDF and LERF cutsets make to the total values, and justify the definition of significance used to define and review significant basic events, cutsets, and accident sequences.

RESPONSE

The top 100 CDF and LERF cutsets represent 52% of the Unit 1 CDF, 77% of the Unit 1 LERF, 53% of the Unit 2 CDF, 81% of the Unit 2 LERF, 42% of the Unit 3 CDF and 74% of the Unit 3 LERF. 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," defines significant accident sequence s as one of the set of accident sequences resulting from the analysis of a specific hazard group

. These groups are defined at the functional or systematic level that, when rank

-ordered by decreasing frequency, sum to a specified percentage of the CDF for that hazard group, OR that individually contribute more than a specified percentage of CDF. ASME/ANS RA

-Sa-2009 states that the summed percentage is 95% of the CDF for the applicable hazard group and the individual percentage is 1% of the CDF for the applicable hazard group.

ASME/ANS RA

-Sa-2009 defines significant accident progression sequence as one of the set of accident sequences contributing to LERF resulting from the analysis of a specific hazard group that, when rank

-ordered by decreasing frequency, sum to a specified percentage of the LERF, or that individually contribute more than a specified percentage of LERF for that hazard group

. These groups are defined at the functional or systematic level, that, when rank

-ordered by decreasing frequency, sum to a specified percentage of the CDF for that hazard group, OR that individually contribute more than a specified percentage of LERF. ASME/ANS RA

-Sa-2009 states that the summed percentage is 95% of the LERF for the applicable hazard group and the individual percentage is 1% of the LERF for the applicable hazard group.

TVA uses the individual contribution to the hazard group as the measure of a significant accident sequence and significant accident progression sequence

. Therefore, the cutsets that contribute at least one percent of the internal events CDF and LERF are characterized as significant accident sequences and significant accident progression sequences, respectively

. Each of the top 100 CDF cutsets individually contribute more than 0.1 7% to CDF for Unit 1, E1-73 0.16% for Unit 2, and 0.14% for Unit 3 and therefore meet the criterion for significant accident sequences

. Each of the top 100 LERF cutsets individually contribute more than 0.09% to LERF for Unit 1, 0.

08% for Unit 2, and 0.1% for Unit 3 and therefore meet the criterion for significant accident progression sequences

. E1-74 PRA RAI 23.i Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the NRC

. RG 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a fire PRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a fire protection program consistent with NFPA

-805. RG 1.200 describes a peer review process utilizing ASME/ANS-RA-Sa-2009 as one acceptable approach for determining the technical adequacy of the PRA once acceptable consensus approaches or models have been established

. The primary result s of a peer review are the F&Os recorded by the peer review and the subsequent resolution of these F&Os.

Clarify the following dispositions to fire F&Os and SR assessment identified in LAR Attachment U that have the potential to impact the FPRA results and do not appear to be fully resolved:

i. 4-45 (Finding against LE

-C3 Met at CC

-I/II/III)

The disposition to this F&O refers to repair credited in the LERF model

. Explain whether repair is credited in the FPRA and whether fire impacts on repair actions are considered.

RESPONSE

Repair is not credited in the Fire PRA

. The response to F&O 4

-45 inappropriately states that "repair" is credited in the LERF model in the resolution statement

. No repair actions were credited in the internal events model or the Fire PRA model

. The BFN Large Early Release Frequency Analysis notebook will be updated to reflect the fact that repair is not credited in the LERF analysis

. The disposition for F&O 4

-45 (i.e., the Resolution Statement) in LAR Attachment U, Table U

-1, is revised to state the following:

"A review of sequences was performed

. Repair actions were not credited in the internal events model. Recovery of offsite power is modeled in the Level 1 PRA and is credited in the LERF model under in-vessel recovery (UxIVR2)." E1-75 PRA RAI 23.l Section 2.4.3.3 of NFPA 805 states that the PRA approach, methods, and data shall be acceptable to the NRC

. RG 1.205 identifies NUREG/CR

-6850 as documenting a methodology for conducting a fire PRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting a fire protection program consistent with NFPA

-805. RG 1.200 describes a peer review process utilizing ASME/ANS-RA-Sa-2009 as one acceptable approach for determining the technical adequacy of the PRA once acceptable consensus approaches or models have been established

. The primary result s of a peer review are the F&Os recorded by the peer review and the subsequent resolution of these F&Os.

Clarify the following dispositions to fire F&Os and SR assessment identified in LAR Attachment U that have the potential to impact the fire PRA results and do not appear to be fully resolved:

l. 6-41 (Finding against SY-B11 Met at CC

-I/II/III)

The emergency diesel generators (EDGs) boundary definition from NUREG/CR

-6928 cited in the F&O disposition does not provide explicit guidance on whether the diesel fuel oil transfer pump and line from the 7

-day diesel storage tank to each diesel generator should be defined to be within the EDG boundary, but it is noted that the fuel oil transfer pumps and lines from the 7

-day diesel storage tanks are not physically collocated with the EDGs. This non- proximity could produce fire impacts that must be considered in the fire PRA. Describe how these components are modeled in the FPRA. RESPONSE:

The fuel oil transfer pumps and lines are collocated within the room of the associated DG

. In the Fire PRA, fuel oil transfer pumps have been included within the DG boundary and are tied to their DG by equipment

-to-equipment logic ties

. The fire effects on the cables associated with these DG components are included in the Fire PRA model.

E1-76 ENCLOSURE 2 Tennessee Valley Authority Browns Ferry Nuclear Plant, Units 1, 2, and 3 Summary of BFN NFPA 805 RAI Response Dates RAI Question Number Type of Response (days)

Actual Date of Response Fire Protection Engineering (FPE)

FPE 01 60 CNL-13-141 December 20, 2013 FPE 02 60 CNL-13-141 December 20, 2013 FPE 03 60 CNL-14-001 January 10, 2014 FPE 04 60 CNL-14-001 January 10, 2014 FPE 05 60 CNL-14-001 January 10, 2014 FPE 06 60 CNL-13-141 December 20, 2013 FPE 07 60 CNL-13-141 December 20, 2013 FPE 08 60 Future letter FPE 09 60 Future letter FPE 10 90 Future letter FPE 11 90 Future letter FPE 12 90 Future letter FPE 13 120 Future letter E2-1 RAI Question Number Type of Response (days)

Actual Date of Response Safe Shutdown Analysis (SSA)

SSA 01 60 CNL-13-141 December 20, 2013 SSA 02 60 Future letter SSA 03 60 Future letter SSA 04 60 CNL-13-141 December 20, 2013 SSA 05 60 CNL-14-001 January 10, 2014 SSA 06 60 Future letter SSA 07 60 CNL-14-001 January 10, 2014 SSA 08 60 CNL-13-141 December 20, 2013 SSA 09 60 Future letter SSA 10 60 Future letter SSA 11 60 CNL-13-141 December 20, 2013 SSA 12 60 CNL-13-141 December 20, 2013 SSA 13 60 Future letter SSA 14 60 CNL-13-141 December 20, 2013 SSA 15 60 Future letter Programmatic (PROG)

PROG 01 60 CNL-13-141 December 20, 2013 E2-2 RAI Question Number Type of Response (days)

Actual Date of Response PROG 02 60 CNL-13-141 December 20, 2013 Fire Modeling (FM)

FM 01, part a 90 Future letter FM 01, part b.i 60 CNL-13-141 December 20, 2013 FM-01, part b.ii 60 CNL-13-141 December 20, 2013 FM-01, part b.iii 90 CNL-14-001 January 10, 2014 FM 01, part c 60 Future letter FM 01, part d.i 60 CNL-14-001 January 10, 2014 FM-01, part d.ii 90 CNL-14-001 January 10, 2014 FM 01, part e 60 CNL-13-141 December 20, 2013 FM 01, part f 60 CNL-14-001 January 10, 2014 FM 01, part g 90 Future letter FM 01, part h.i 60 CNL-14-001 January 10, 2014 FM 01, part h.ii 60 CNL-14-001 January 10, 2014 FM 01, part h.iii 90 Future letter FM 01, part i.i 120 Future letter FM 01, part i.ii 60 CNL-13-141 December 20, 2013 E2-3 RAI Question Number Type of Response (days)

Actual Date of Response FM 01, part i.iii 90 CNL-14-001 January 10, 2014 FM 01, part i.iv 120 Future letter FM 01, part i.v 60 CNL-13-141 December 20, 2013 FM 01, part i.vi 60 CNL-13-141 December 20, 2013 FM 01, part i.vii 90 CNL-14-001 January 10, 2014 FM 01, part i.viii 60 CNL-13-141 December 20, 2013 FM 01, part j.i 60 CNL-13-141 December 20, 2013 FM 01, part j.ii 90 CNL-14-001 January 10, 2014 FM 02, part a 120 Future letter FM 02, part b 120 Future letter FM 02, part c 60 CNL-13-141 December 20, 2013 FM 02, part d 60 CNL-13-141 December 20, 2013 FM 02, part e 90 Future letter FM 03 60 CNL-14-001 January 10, 2014 FM 04 90 Future letter FM 05 60 CNL-14-001 January 10, 2014 FM 06 60 CNL-14-001 January 10, 2014 E2-4 RAI Question Number Type of Response (days)

Actual Date of Response Probabilistic Risk Assessment (PRA)

PRA 01, part a 60 CNL-14-001 January 10, 2014 PRA 01, part b 60 Future letter PRA 01, part c 90 CNL-14-001 January 10, 2014 PRA 01, part d 90 Future letter PRA 01, part e 120 Future letter PRA 01, part f 120 Future letter PRA 01, part g 60 CNL-14-001 January 10, 2014 PRA 01, part h 90 Future letter PRA 01, part i 60 CNL-13-141 December 20, 2013 PRA 01, part j 60 Future letter PRA 01, part k 60 CNL-14-001 January 10, 2014 PRA 01, part l 60 Future letter PRA 01, part m 60 CNL-14-001 January 10, 2014 PRA 01, part n 60 CNL-14-001 January 10, 2014 PRA 01, part o 120 Future letter PRA 01, part p 90 Future letter PRA 01, part q 60 CNL-13-141 December 20, 2013 PRA 01, part r 60 CNL-13-141 December 20, 2013 E2-5 RAI Question Number Type of Response (days)

Actual Date of Response PRA 01, part s 90 Future letter PRA 01, part t 60 CNL-14-001 January 10, 2014 PRA 01, part u 60 CNL-13-141 December 20, 2013 PRA 01, part v 120 Future letter PRA 02 60 CNL-14-001 January 10, 2014 PRA 03 60 CNL-13-141 December 20, 2013 PRA 04 90 Future letter PRA 05 60 Future letter PRA 06 60 CNL-14-001 January 10, 2014 PRA 07 60 Future letter PRA 08 60 CNL-14-001 January 10, 2014 PRA 09 60 CNL-14-001 January 10, 2014 PRA 10 90 Future letter PRA 11 60 Future letter PRA 12 120 Future letter PRA 13 60 Future letter PRA 14 60 CNL-14-001 January 10, 2014 PRA 15 90 Future letter PRA 16 90 Future letter E2-6 RAI Question Number Type of Response (days)

Actual Date of Response PRA 17 90 Future letter PRA 18 60 CNL-14-001 January 10, 2014 PRA 19 60 CNL-14-001 January 10, 2014 PRA 20 120 Future letter PRA 21 60 CNL-14-001 January 10, 2014 PRA 22 60 CNL-13-141 December 20, 2013 PRA 23 60 Parts c, f, i, and l: CNL-14-001 January 10, 2014 Remaining Parts: Future letter Radioactive Release (RR)

RR 01 60 Future letter RR 02 60 Future letter E2-7